From danmcd at omniti.com Wed Oct 1 14:18:53 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 1 Oct 2014 10:18:53 -0400 Subject: [OmniOS-discuss] Coming soon: r151013 version of "bloody" In-Reply-To: <385B7FCC-249D-4A65-B625-6F3118E5B2BD@omniti.com> References: <385B7FCC-249D-4A65-B625-6F3118E5B2BD@omniti.com> Message-ID: <6BF9972B-A64E-4303-B0ED-E72D0441ED4F@omniti.com> On Sep 30, 2014, at 2:12 PM, Dan McDonald wrote: > > So heads up, r151013 will be coming very soon. I will have release media for it at the same time. If anyone out there uses Kayak, I'd be especially interested in bloody testing for Kayak. If I hear no objections, I plan on cutting all of this over before COB today, US/Eastern (which is between 2100-2200 GMT if I did my timezone calculations correctly). Dan From simon at nostalgeek.com Wed Oct 1 15:45:43 2014 From: simon at nostalgeek.com (Simon Boulet) Date: Wed, 1 Oct 2014 11:45:43 -0400 Subject: [OmniOS-discuss] Time zone for Quebec, Canada Message-ID: Hi, I just recently tried OmniOS r151012. In the installer, when prompted for the time zone, under America, the "Eastern Time - Ontario & Quebec - places that did not observe DST.... " time zone is a bit misleading. I believe most part of Quebec (and Ontario) do observe DST. I think the "Eastern Time - Ontario - most locations", should be renamed to "Eastern Time - Ontario & Quebec - most locations". Thanks Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Screenshot - 011014 - 11:39:52 AM.png Type: image/png Size: 10095 bytes Desc: not available URL: From danmcd at omniti.com Wed Oct 1 17:09:58 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 1 Oct 2014 13:09:58 -0400 Subject: [OmniOS-discuss] Time zone for Quebec, Canada In-Reply-To: References: Message-ID: On Oct 1, 2014, at 11:45 AM, Simon Boulet wrote: > Hi, > > I just recently tried OmniOS r151012. > > In the installer, when prompted for the time zone, under America, the "Eastern Time - Ontario & Quebec - places that did not observe DST.... " time zone is a bit misleading. I believe most part of Quebec (and Ontario) do observe DST. I think the "Eastern Time - Ontario - most locations", should be renamed to "Eastern Time - Ontario & Quebec - most locations". I believe those are generated from IANA's timezone database, which we updated for r151012. It looks like there's been a new update post-freeze, but generally, we just compile those and run. Here's an up to date list for Canada: http://home.kpn.nl/vanadovv/time/Multizones.html#ca I notice Quebec is split in three here, including one zone that does not honor DST. Dan From simon at nostalgeek.com Wed Oct 1 17:28:25 2014 From: simon at nostalgeek.com (Simon Boulet) Date: Wed, 1 Oct 2014 13:28:25 -0400 Subject: [OmniOS-discuss] Time zone for Quebec, Canada In-Reply-To: References: Message-ID: Hi Dan, On Wed, Oct 1, 2014 at 1:09 PM, Dan McDonald wrote: > > I believe those are generated from IANA's timezone database, which we > updated for r151012. It looks like there's been a new update post-freeze, > but generally, we just compile those and run. > > Here's an up to date list for Canada: > http://home.kpn.nl/vanadovv/time/Multizones.html#ca > > I notice Quebec is split in three here, including one zone that does not > honor DST. > Most locations in Quebec do honor DST though, thus my recommendation for renaming "Eastern Time - Ontario - most locations" to "Eastern Time - Ontario & Quebec - most locations". Hopefully this can be fixed in the next release. Not a big issue though. Thanks Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: From johan.kragsterman at capvert.se Wed Oct 1 17:29:15 2014 From: johan.kragsterman at capvert.se (Johan Kragsterman) Date: Wed, 1 Oct 2014 19:29:15 +0200 Subject: [OmniOS-discuss] add second storage device in KVM? Message-ID: Hi! Funny, I never thought about this until now: How do I add a second storage device in a KVM VM? It came up now, when I installed FreeNAS 9.3 beta as a VM, and wanted to add storage pools. I looked at my start script, and of coarse I could do some experimenting, but perhaps there are also other people on this list that want to know this too. My script looks like this: root at omni:~# cat /usr/bin/freenas.sh #!/usr/bin/bash # on omnios command is /usr/bin/qemu-system-x86_64 # configuration NAME="freenas" NUM=5 VNIC0=freenas0 HDD=/dev/zvol/rdsk/mainpool/vm/freenas/os CD=/mainpool/nfs/iso/freenas.iso MEM=8192 # don't change below here! TLN=`expr 7000 + $NUM` mac0=`dladm show-vnic -po macaddress $VNIC0` /usr/bin/qemu-system-x86_64 \ -name $NAME \ -boot cd \ -enable-kvm \ -vnc 0.0.0.0:$NUM \ -smp cores=2,threads=2,sockets=1 \ -m $MEM \ -no-hpet \ -localtime \ -drive file=$HDD,if=virtio,index=0 \ -drive file=$CD,media=cdrom,if=ide,index=2 \ -net nic,vlan=0,name=net0,model=virtio,macaddr=$mac0 \ -net vnic,vlan=0,name=net0,ifname=$VNIC0,macaddr=$mac0 \ -vga std \ -cpu host \ -pidfile /mainpool/vm/freenas/pids/$NAME.pid \ -monitor telnet:localhost:$TLN,server,nowait,nodelay \ -daemonize if [ $? -gt 0 ]; then echo "Failed to start VM" exit fi port=`expr 5900 + $NUM` public_nic=$(dladm show-vnic|grep vnic0|awk '{print $2}') public_ip=$(ifconfig $public_nic|grep inet|awk '{print $2}') echo "Started VM: $NAME" echo "VNC available at: host IP ${public_ip} port ${port}" echo "QEMU Monitor, do: # telnet localhost $TLN. Note: use Control ] to exit monitor before quit! " root at omni:~# The nics are numbered, are the storage devices also numbered? In linux the can "hot add" via virsh. Can't find any docs from linux that are useful for me... Best regards from/Med v?nliga h?lsningar fr?n Johan Kragsterman Capvert From sjorge+ml at blackdot.be Wed Oct 1 17:32:33 2014 From: sjorge+ml at blackdot.be (Jorge Schrauwen) Date: Wed, 01 Oct 2014 19:32:33 +0200 Subject: [OmniOS-discuss] =?utf-8?q?add_second_storage_device_in_KVM=3F?= In-Reply-To: References: Message-ID: Hi Johan, Just add another line like this: -drive file=$HDD2,if=virtio,index=1 \ Were you increment the index value. Does not need to be more complicated than it has to be right :) Regards Jorge On 2014-10-01 19:29, Johan Kragsterman wrote: > Hi! > > Funny, I never thought about this until now: > > How do I add a second storage device in a KVM VM? > > It came up now, when I installed FreeNAS 9.3 beta as a VM, and wanted > to add storage pools. > > I looked at my start script, and of coarse I could do some > experimenting, but perhaps there are also other people on this list > that want to know this too. > > My script looks like this: > > > root at omni:~# cat /usr/bin/freenas.sh > #!/usr/bin/bash > > # on omnios command is /usr/bin/qemu-system-x86_64 > > # configuration > NAME="freenas" > NUM=5 > VNIC0=freenas0 > HDD=/dev/zvol/rdsk/mainpool/vm/freenas/os > CD=/mainpool/nfs/iso/freenas.iso > MEM=8192 > > # don't change below here! > > TLN=`expr 7000 + $NUM` > mac0=`dladm show-vnic -po macaddress $VNIC0` > > /usr/bin/qemu-system-x86_64 \ > -name $NAME \ > -boot cd \ > -enable-kvm \ > -vnc 0.0.0.0:$NUM \ > -smp cores=2,threads=2,sockets=1 \ > -m $MEM \ > -no-hpet \ > -localtime \ > -drive file=$HDD,if=virtio,index=0 \ > -drive file=$CD,media=cdrom,if=ide,index=2 \ > -net nic,vlan=0,name=net0,model=virtio,macaddr=$mac0 \ > -net vnic,vlan=0,name=net0,ifname=$VNIC0,macaddr=$mac0 \ > -vga std \ > -cpu host \ > -pidfile /mainpool/vm/freenas/pids/$NAME.pid \ > -monitor telnet:localhost:$TLN,server,nowait,nodelay \ > -daemonize > > if [ $? -gt 0 ]; then > echo "Failed to start VM" > exit > fi > > port=`expr 5900 + $NUM` > public_nic=$(dladm show-vnic|grep vnic0|awk '{print $2}') > public_ip=$(ifconfig $public_nic|grep inet|awk '{print $2}') > > echo "Started VM: $NAME" > echo "VNC available at: host IP ${public_ip} port ${port}" > echo "QEMU Monitor, do: # telnet localhost $TLN. Note: use Control ] > to exit monitor before quit! > " > root at omni:~# > > > The nics are numbered, are the storage devices also numbered? > > In linux the can "hot add" via virsh. Can't find any docs from linux > that are useful for me... > > > > Best regards from/Med v?nliga h?lsningar fr?n > > Johan Kragsterman > > Capvert > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From danmcd at omniti.com Wed Oct 1 17:32:58 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 1 Oct 2014 13:32:58 -0400 Subject: [OmniOS-discuss] Time zone for Quebec, Canada In-Reply-To: References: Message-ID: On Oct 1, 2014, at 1:28 PM, Simon Boulet wrote: > Hi Dan, > > On Wed, Oct 1, 2014 at 1:09 PM, Dan McDonald wrote: > > I believe those are generated from IANA's timezone database, which we updated for r151012. It looks like there's been a new update post-freeze, but generally, we just compile those and run. > > Here's an up to date list for Canada: http://home.kpn.nl/vanadovv/time/Multizones.html#ca > > I notice Quebec is split in three here, including one zone that does not honor DST. > > Most locations in Quebec do honor DST though, thus my recommendation for renaming "Eastern Time - Ontario - most locations" to "Eastern Time - Ontario & Quebec - most locations". > > Hopefully this can be fixed in the next release. Not a big issue though. You should speak to IANA. They maintain these. Dan From simon at nostalgeek.com Wed Oct 1 17:43:55 2014 From: simon at nostalgeek.com (Simon Boulet) Date: Wed, 1 Oct 2014 13:43:55 -0400 Subject: [OmniOS-discuss] Time zone for Quebec, Canada In-Reply-To: References: Message-ID: On Wed, Oct 1, 2014 at 1:32 PM, Dan McDonald wrote: > > > You should speak to IANA. They maintain these. > > What do you want me to ask them? They seems to have it right. >From here: http://home.kpn.nl/vanadovv/time/Multizones.html#ca You can read: *Qu?bec Eastern time, UTC-5, DST Most of Qu?bec*, West of 63? W. This includes the part of the Minganie region that is East of 63? W and it includes the whole of ?le d'Anticosti. Other sources also confirms this. Most of Quebec honors DST. I don't know how OmniOS pulls the data from IANA, but somehow you're pulling it wrong. Somehow OmniOS seems to think most of Quebec is NOT DST. Agree that's not a huge issue but it does make OmniOS installer a bit less enjoyable. If it's something that can easily be fixed that be great. Thanks Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Wed Oct 1 18:12:57 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 1 Oct 2014 14:12:57 -0400 Subject: [OmniOS-discuss] Bloody is now at r151013 Message-ID: <76EDC42B-49F4-465F-99E8-69ADBCAFBF94@omniti.com> I've reinstalled the bloody repo with new r151013 bits: http://pkg.omniti.com/omnios/bloody/ and put install media in the correct places. This, BTW, is r151013 Oct 1st, or 20141001 in the release media names. Per before, you can only use "pkg update" if you update from something without a broken CherryPy. This include updating from r151010, 012, or bloody/011 prior to mid-August. Old BEs can be your friends here. Happy bleeding! Dan From danmcd at omniti.com Wed Oct 1 19:05:48 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 1 Oct 2014 15:05:48 -0400 Subject: [OmniOS-discuss] ANN: OmniOS r151012 (stable) released Message-ID: We announced this at Illumos Day and on Social Media on Saturday. Here's the YouTube video introducing both Illumos Day and r151012: https://www.youtube.com/watch?v=HRnGZYEBpFg&list=PLH8r-Scm3-2VmZhZ76tFPAhPOG0pvmjdA The relevant bits are at 5:20. . . . OmniOS r151012, the newest Stable release of OmniOS is now available for download and upgrading to. This release offers many enhancements and improvements. It also marks the beginning of this and future major releases of OmniOS having their own IPS repositories. Please read the Release Notes and Upgrade Instructions for full details on these and other changes. * Installation Media: http://omnios.omniti.com/wiki.php/Installation * Release Notes: http://omnios.omniti.com/wiki.php/ReleaseNotes/r151012 * Upgrade Instructions: Use the same ones as http://omnios.omniti.com/wiki.php/Upgrade_r151008_r151010, but substitute in "012" where you see "010". Thanks! Dan McDonald -- OmniOS Engineering p.s. Since the release, there have been many bash patches issued by the bash community. Simply uttering "pkg update" will get you the very latest they have to offer. From lists at mcintyreweb.com Thu Oct 2 03:32:41 2014 From: lists at mcintyreweb.com (Hugh McIntyre) Date: Wed, 01 Oct 2014 20:32:41 -0700 Subject: [OmniOS-discuss] Time zone for Quebec, Canada In-Reply-To: References: Message-ID: <542CC759.4010404@mcintyreweb.com> On 10/01/2014 10:43 AM, Simon Boulet wrote: > On Wed, Oct 1, 2014 at 1:32 PM, Dan McDonald wrote: > >> You should speak to IANA. They maintain these. > > What do you want me to ask them? They seems to have it right. > > From here: http://home.kpn.nl/vanadovv/time/Multizones.html#ca > > You can read: > *Qu?bec Eastern time, UTC-5, DST Most of Qu?bec*, West of 63? W. This > includes the part of the Minganie region that is East of 63? W and it > includes the whole of ?le d'Anticosti. > > Other sources also confirms this. Most of Quebec honors DST. > > I don't know how OmniOS pulls the data from IANA, but somehow you're > pulling it wrong. Somehow OmniOS seems to think most of Quebec is NOT DST. Indeed, from the latest zoneinfo code at iana.org: CA +5125-05707 America/Blanc-Sablon Atlantic Standard Time - Quebec - Lower North Shore CA +4339-07923 America/Toronto Eastern Time - Ontario & Quebec - most locations CA +4901-08816 America/Nipigon Eastern Time - Ontario & Quebec - places that did not observe DST 1967-1973 The latest Illumos sources (and also illumos-omnios) seems to have the same in usr/src/cmd/zic/zone.tab(.txt). Does the installer use a different source file, or is the version of this file in omnios r151012 somehow older? > Agree that's not a huge issue but it does make OmniOS installer a bit > less enjoyable. > If it's something that can easily be fixed that be great. Hugh. From gate03 at landcroft.co.uk Thu Oct 2 08:15:10 2014 From: gate03 at landcroft.co.uk (Michael Mounteney) Date: Thu, 2 Oct 2014 18:15:10 +1000 Subject: [OmniOS-discuss] openldap-server package out of sync? Message-ID: <20141002181510.214d803b@pantry> Hello, I'm on stable. Got all up to date with http://omnios.omniti.com/wiki.php/GeneralAdministration#UpgradingWithNon-GlobalZones then: root at world:/support/Solaris# pkg -R /zone/ldap/root install openldap-server Creating Plan - pkg install: No solution was found to satisfy constraints maintained incorporations: pkg://omnios/entire at 11,5.11-0.151010:20140512T172807Z pkg://omnios/incorporation/jeos/illumos-gate at 11,5.11-0.151010:20140427T150707Z pkg://omnios/consolidation/osnet/osnet-incorporation at 0.5.11,5.11-0.151010:20140428T192608Z pkg://omnios/incorporation/jeos/omnios-userland at 11,5.11-0.151010:20140512T172819Z Plan Creation: dependency error(s) in proposed packages: No suitable version of required package pkg://ms.omniti.com/omniti/network/openldap-server at 2.4.40,5.11-0.151006:20140930T201405Z found: Reject: pkg://ms.omniti.com/omniti/network/openldap-server at 2.4.40,5.11-0.151006:20140930T201405Z Reason: A version for 'incorporate' dependency on pkg:/entire at 11,5.11-0.151006 cannot be found Plan Creation: Errors in installed packages due to proposed changes: No suitable version of installed package pkg://omnios/entire at 11,5.11-0.151010:20140512T172807Z found Reject: pkg://omnios/entire at 11,5.11-0.151010:20140512T172807Z Reason: Excluded by proposed incorporation 'omniti/network/openldap-server' This version is excluded by installed incorporation pkg://omnios/entire at 11,5.11-0.151010:20140512T172807Z Any ideas ? Michael. From lotheac at iki.fi Thu Oct 2 10:56:25 2014 From: lotheac at iki.fi (Lauri Tirkkonen) Date: Thu, 2 Oct 2014 13:56:25 +0300 Subject: [OmniOS-discuss] openldap-server package out of sync? In-Reply-To: <20141002181510.214d803b@pantry> References: <20141002181510.214d803b@pantry> Message-ID: <20141002105625.GC25275@gutsman.lotheac.fi> On Thu, Oct 02 2014 18:15:10 +1000, Michael Mounteney wrote: > No suitable version of required package pkg://ms.omniti.com/omniti/network/openldap-server at 2.4.40,5.11-0.151006:20140930T201405Z found: > Reject: pkg://ms.omniti.com/omniti/network/openldap-server at 2.4.40,5.11-0.151006:20140930T201405Z > Reason: A version for 'incorporate' dependency on pkg:/entire at 11,5.11-0.151006 cannot be found Here's how this reads: the package from ms.omniti.com cannot be installed because it constrains 'entire' to 11,5.11-0.151006. This is because, as far as I know, OmniTI builds their managed services (omniti-ms) packages only for LTS, and they add incorporate dependencies to prevent the packages from being installed on newer releases. > Any ideas ? OpenLDAP is also available in the repo that I maintain (http://pkg.niksula.hut.fi/). While it doesn't constrain entire to 151006 like omniti-ms packages do, the packages have been built in 151006 and I make no guarantees about them working on any release, but of course welcome pull requests ;) All that said, per KYSTY (http://omnios.omniti.com/wiki.php/KYSTY) it might be a good idea to build packages you want yourself to avoid being constrained by someone else's idea of what version you should get. This is pretty easy to do by forking omnios-build and pulling ready-made build scripts from the available repositories. -- Lauri Tirkkonen | +358 50 5341376 | lotheac @ IRCnet From natxo.asenjo at gmail.com Thu Oct 2 17:46:50 2014 From: natxo.asenjo at gmail.com (Natxo Asenjo) Date: Thu, 2 Oct 2014 19:46:50 +0200 Subject: [OmniOS-discuss] centos 7 nfsv4 client nobody problems Message-ID: hi, probably something easy, but I cannot seem to solve it. I have shared using nfs a dataset with these properties on omnios: # zfs get sharenfs tank/nuc01 NAME PROPERTY VALUE SOURCE tank/nuc01 sharenfs rw=@192.168.0.0.0/24,root=@192.168.0.0/24 local so quite liberal you should think ;-) But my centos 7 client always squashes the root files: # ls -l nuc01/ -rw-------. 1 nobody nobody 8591507456 Oct 1 22:43 kdc.qcow2 -rw-r--r--. 1 nobody nobody 0 Oct 2 19:38 test and these are the mount options: (rw,relatime,vers=4.0,rsize=1048576,wsize=1048576,namlen=255,hard,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=192.168.0.11,local_lock=none,addr=192.168.0.172) I have modified /etc/idmapd.conf on the centos host to have the same Domain as the omnios box. Is nfsv3 my only option? Thanks! -- Groeten, natxo -------------- next part -------------- An HTML attachment was scrubbed... URL: From natxo.asenjo at gmail.com Thu Oct 2 18:41:23 2014 From: natxo.asenjo at gmail.com (Natxo Asenjo) Date: Thu, 2 Oct 2014 20:41:23 +0200 Subject: [OmniOS-discuss] centos 7 nfsv4 client nobody problems In-Reply-To: References: Message-ID: On Thu, Oct 2, 2014 at 7:46 PM, Natxo Asenjo wrote: and of course it was every rh admin best friend: selinux # setsebool -P virt_use_nfs on and everything will be fine if you want to use nfs for libvirtd or just turn it off ;-) Sorry for the noise. -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Thu Oct 2 18:45:26 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 2 Oct 2014 14:45:26 -0400 Subject: [OmniOS-discuss] centos 7 nfsv4 client nobody problems In-Reply-To: References: Message-ID: <2967FF79-1896-4DEB-82B1-E55A45892144@omniti.com> On Oct 2, 2014, at 2:41 PM, Natxo Asenjo wrote: > > > On Thu, Oct 2, 2014 at 7:46 PM, Natxo Asenjo wrote: > > and of course it was every rh admin best friend: selinux > > # setsebool -P virt_use_nfs on > > and everything will be fine if you want to use nfs for libvirtd > > or just turn it off ;-) > > Sorry for the noise. That wasn't noise. That's a problem you solved publically that now is in the archives in case someone ELSE sees this same problem. NEVER feel sorry for that kind of noise. Dan From danmcd at omniti.com Thu Oct 2 19:04:39 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 2 Oct 2014 15:04:39 -0400 Subject: [OmniOS-discuss] ATTENTION: End of Support Life for OmniOS r151008 References: <0E9D0A37-8295-4791-A698-B300FB7B6B4F@omniti.com> Message-ID: Hello OmniOS users! With the release of r151012, support for OmniOS r151008 is now terminated. This is in accordance with the OmniOS release cycle shown here: http://omnios.omniti.com/wiki.php/ReleaseCycle . Anyone with questions about their r151008 problems will be told, without question, to upgrade to r151010 or r151012. We officially support an upgrade from r151008 to r151010. Experiments with a direct jump from r151008 to r151012 have proven promising, but as of now, officially we do not support such a direct upgrade (we will, however, not be as harsh about such an upgrade as we will be with continued r151008 questions). Furthermore, we will be REMOVING r151008 packages from the http://pkg.omniti.com/omnios/release/ repository within the next 2-4 weeks. The release of r151008 caused many headaches to users who wished to remain with r151006 (our Long-Term Support). Removing the 008 packages SHOULD make life easier for those still on the LTS release. Please speak up now if you have any concerns. I will issue another warning when we get closer to actual removal. Thank you, Dan McDonald - OmniOS Engineering p.s. beadm(1M) is your friend. Creating backup BEs in case of mistakes is easy. ZFS means they don't take up a lot of space and happen quickly. From danmcd at omniti.com Thu Oct 2 21:41:30 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 2 Oct 2014 17:41:30 -0400 Subject: [OmniOS-discuss] Fwd: [discuss] SmartOS/Illumos meetup in Berlin (2014-10-08) References: <3E643A40-2C9F-41FF-AE82-B4F5B76F6152@core.io> Message-ID: <3C633045-4631-4222-9489-DDF75A2D2D5D@omniti.com> Figured this should be forwarded here as well for anyone nearby or traveling there. Dan Sent from my iPhone (typos, autocorrect, and all) Begin forwarded message: > From: "Sebastian Wiedenroth via illumos-discuss" > Date: October 2, 2014 at 5:30:46 PM EDT > To: discuss at lists.illumos.org > Subject: [discuss] SmartOS/Illumos meetup in Berlin (2014-10-08) > Reply-To: discuss at lists.illumos.org, "Sebastian Wiedenroth" > > Hi everyone, > > we?re organizing a SmartOS Meetup in Berlin next week. > Since there is probably a large overlap in topics with other Illumos distributions I'm also posting to this list. > Illumos users of all distros are invited! :-) > > Time: Wednesday 2014-10-08 19:00 (CEST) > > The idea was to get to meet the local community, order some pizza and maybe do some short unconference style talks for topics that gather interest - or just have a general discussion. > As this is the first time for us doing such an event we?re very open to suggestions/feedback regarding planning and format. > > Exact details can be found on the meetup page: http://www.meetup.com/SmartOS-D-A-CH-UG/events/208543672/ > Please sign up there so we know how many people to expect. > > If you have any questions don?t hesitate to email me or ping me on irc (wiedi on freenode). > > Best regards, > Sebastian > > > ------------------------------------------- > illumos-discuss > Archives: https://www.listbox.com/member/archive/182180/=now > RSS Feed: https://www.listbox.com/member/archive/rss/182180/21175487-64bb3106 > Modify Your Subscription: https://www.listbox.com/member/?member_id=21175487&id_secret=21175487-26add75e > Powered by Listbox: http://www.listbox.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From gate03 at landcroft.co.uk Thu Oct 2 22:38:10 2014 From: gate03 at landcroft.co.uk (Michael Mounteney) Date: Fri, 3 Oct 2014 08:38:10 +1000 Subject: [OmniOS-discuss] switching from Stable to LTS Message-ID: <39bcb14c201f7d1cacc45f29f6ceebfa.squirrel@landcroft.co.uk> Thanks for the answers to my question about installing openldap-server. As I would prefer to limit the number of repositories from which I'm taking packages, I'm considering switching to LTS so I can take packages from OmniTI. I'm sure I saw somewhere how to switch from one release to another but can't find the instructions now. It seems to me that the process would be something like: clone the current BE to something like omnios-LTS. use pkg -R /rpool/omnios-LTS er_something to switch the release of omnios-LTS. use pkg -R /rpool/omnios-LTS update reboot into BE omnios-LTS but that won't fix the zones. I wonder if a more simple answer is to have one zone just for openldap, on LTS, and the rest of the system on Stable, if that's possible. Obviously I could just reinstall but that's painful. Any other ideas? Thanks, Michael. From danmcd at omniti.com Thu Oct 2 22:48:29 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 2 Oct 2014 18:48:29 -0400 Subject: [OmniOS-discuss] switching from Stable to LTS In-Reply-To: <39bcb14c201f7d1cacc45f29f6ceebfa.squirrel@landcroft.co.uk> References: <39bcb14c201f7d1cacc45f29f6ceebfa.squirrel@landcroft.co.uk> Message-ID: <9B51D75F-2F00-4A0C-994A-D99AC09E003B@omniti.com> Going backwards is VERY difficult. IPS assumes progress is forward, and penalizes you accordingly. (cf the recent CherryPy incident in bloody). Sorry I can't help more, Dan Sent from my iPhone (typos, autocorrect, and all) > On Oct 2, 2014, at 6:38 PM, Michael Mounteney wrote: > > Thanks for the answers to my question about installing openldap-server. > As I would prefer to limit the number of repositories from which I'm > taking packages, I'm considering switching to LTS so I can take packages > from OmniTI. > > I'm sure I saw somewhere how to switch from one release to another but > can't find the instructions now. It seems to me that the process would be > something like: > > clone the current BE to something like omnios-LTS. > use pkg -R /rpool/omnios-LTS er_something to switch the release of > omnios-LTS. > use pkg -R /rpool/omnios-LTS update > reboot into BE omnios-LTS > > but that won't fix the zones. I wonder if a more simple answer is to have > one zone just for openldap, on LTS, and the rest of the system on Stable, > if that's possible. > > Obviously I could just reinstall but that's painful. > > Any other ideas? Thanks, Michael. > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From mir at miras.org Thu Oct 2 23:11:31 2014 From: mir at miras.org (Michael Rasmussen) Date: Fri, 3 Oct 2014 01:11:31 +0200 Subject: [OmniOS-discuss] switching from Stable to LTS In-Reply-To: <39bcb14c201f7d1cacc45f29f6ceebfa.squirrel@landcroft.co.uk> References: <39bcb14c201f7d1cacc45f29f6ceebfa.squirrel@landcroft.co.uk> Message-ID: <20141003011131.5e81500b@sleipner.datanom.net> > but that won't fix the zones. I wonder if a more simple answer is to have > one zone just for openldap, on LTS, and the rest of the system on Stable, > if that's possible. > > Obviously I could just reinstall but that's painful. > > Any other ideas? Thanks, Michael. I think the generel consensus is that unless you are very very very knowledgeable in the darkest corners of Illumos, having patience above and beyond the call of duty, have loads of spare time, and only have a server populated with data of absolut nonsense you should never even think of downgrading your server(s). -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: > Where in the US is Linus? He was in the "Promise Land". -- David S. Miller -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From danmcd at omniti.com Fri Oct 3 00:21:41 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 2 Oct 2014 20:21:41 -0400 Subject: [OmniOS-discuss] switching from Stable to LTS In-Reply-To: <20141003011131.5e81500b@sleipner.datanom.net> References: <39bcb14c201f7d1cacc45f29f6ceebfa.squirrel@landcroft.co.uk> <20141003011131.5e81500b@sleipner.datanom.net> Message-ID: On Oct 2, 2014, at 7:11 PM, Michael Rasmussen wrote: > > I think the generel consensus is that unless you are very very very > knowledgeable in the darkest corners of Illumos, IPS, not illumos. IPS, in spite of having its metadata in illumos-gate, is not part of illumos itself. > having patience above > and beyond the call of duty, have loads of spare time, and only have a > server populated with data of absolut nonsense you should never even > think of downgrading your server(s). Dan From magnus at yonderway.com Fri Oct 3 03:52:14 2014 From: magnus at yonderway.com (Magnus) Date: Thu, 2 Oct 2014 23:52:14 -0400 Subject: [OmniOS-discuss] Jenkins CI on OmniOS? (NPE) Message-ID: <356698EA-D360-49F9-90A1-F5F422FEE087@yonderway.com> So I?ve just done this same thing successfully on a number of other platforms (that I don?t wish to run on), including a Macbook, a Debian laptop, a SmartOS zone? but when I try to do this inside of a zone on OmniOS, I get a Java stack trace that doesn?t make a whole lot of sense to me? and Google isn?t being helpful with this one. I?m hoping to write a blog about hosting CI on OmniOS, complete with getting it running under SMF, etc. but this one piece is being stubborn. Thoughts? $ java -jar jenkins.war Running from: /export/home/jenkins/jenkins.war webroot: $user.home/.jenkins Oct 03, 2014 3:28:19 AM winstone.Logger logInternal INFO: Beginning extraction from war file Oct 03, 2014 3:28:20 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: jetty-8.y.z-SNAPSHOT Oct 03, 2014 3:28:22 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: NO JSP Support for , did not find org.apache.jasper.servlet.JspServlet Jenkins home directory: /export/home/jenkins/.jenkins found at: $user.home/.jenkins Oct 03, 2014 3:28:23 AM hudson.util.BootFailure publish SEVERE: Failed to initialize Jenkins hudson.util.AWTProblem: java.lang.NullPointerException at hudson.WebAppMain.contextInitialized(WebAppMain.java:182) at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) at org.eclipse.jetty.server.Server.doStart(Server.java:282) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at winstone.Launcher.(Launcher.java:154) at winstone.Launcher.main(Launcher.java:354) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:606) at Main._main(Main.java:293) at Main.main(Main.java:98) Caused by: java.lang.NullPointerException at sun.awt.X11FontManager.getDefaultPlatformFont(X11FontManager.java:779) at sun.font.SunFontManager$2.run(SunFontManager.java:433) at java.security.AccessController.doPrivileged(Native Method) at sun.font.SunFontManager.(SunFontManager.java:376) at sun.awt.X11FontManager.(X11FontManager.java:32) at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:57) at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) at java.lang.reflect.Constructor.newInstance(Constructor.java:526) at java.lang.Class.newInstance(Class.java:374) at sun.font.FontManagerFactory$1.run(FontManagerFactory.java:83) at java.security.AccessController.doPrivileged(Native Method) at sun.font.FontManagerFactory.getInstance(FontManagerFactory.java:74) at java.awt.Font.getFont2D(Font.java:490) at java.awt.Font.getFamily(Font.java:1219) at java.awt.Font.getFamily_NoClientCode(Font.java:1193) at java.awt.Font.getFamily(Font.java:1185) at java.awt.Font.toString(Font.java:1682) at hudson.util.ChartUtil.(ChartUtil.java:229) at hudson.WebAppMain.contextInitialized(WebAppMain.java:181) ... 19 more Oct 03, 2014 3:28:23 AM org.eclipse.jetty.util.log.JavaUtilLog warn WARNING: Failed startup of context w.{,file:/export/home/jenkins/.jenkins/war/},/export/home/jenkins/.jenkins/war java.lang.NullPointerException at jenkins.util.groovy.GroovyHookScript.run(GroovyHookScript.java:63) at hudson.util.BootFailure.publish(BootFailure.java:43) at hudson.WebAppMain.contextInitialized(WebAppMain.java:244) at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) at org.eclipse.jetty.server.Server.doStart(Server.java:282) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at winstone.Launcher.(Launcher.java:154) at winstone.Launcher.main(Launcher.java:354) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:606) at Main._main(Main.java:293) at Main.main(Main.java:98) Oct 03, 2014 3:28:23 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: Started SelectChannelConnector at 0.0.0.0:8080 Oct 03, 2014 3:28:23 AM winstone.Logger logInternal INFO: Winstone Servlet Engine v2.0 running: controlPort=disabled *********** If I try the same thing with an Oracle JRE, I get a different error: $ /opt/jre1.8.0_20/bin/java -jar jenkins.war Running from: /export/home/jenkins/jenkins.war webroot: $user.home/.jenkins Oct 03, 2014 3:45:00 AM winstone.Logger logInternal INFO: Beginning extraction from war file Oct 03, 2014 3:45:01 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: jetty-8.y.z-SNAPSHOT Oct 03, 2014 3:45:03 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: NO JSP Support for , did not find org.apache.jasper.servlet.JspServlet Jenkins home directory: /export/home/jenkins/.jenkins found at: $user.home/.jenkins Oct 03, 2014 3:45:03 AM hudson.util.BootFailure publish SEVERE: Failed to initialize Jenkins hudson.util.AWTProblem: java.lang.UnsatisfiedLinkError: /opt/jre1.8.0_20/lib/amd64/libfontmanager.so: ld.so.1: java: fatal: libXext.so.0: open failed: No such file or directory at hudson.WebAppMain.contextInitialized(WebAppMain.java:182) at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) at org.eclipse.jetty.server.Server.doStart(Server.java:282) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at winstone.Launcher.(Launcher.java:154) at winstone.Launcher.main(Launcher.java:354) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:483) at Main._main(Main.java:293) at Main.main(Main.java:98) Caused by: java.lang.UnsatisfiedLinkError: /opt/jre1.8.0_20/lib/amd64/libfontmanager.so: ld.so.1: java: fatal: libXext.so.0: open failed: No such file or directory at java.lang.ClassLoader$NativeLibrary.load(Native Method) at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1929) at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1835) at java.lang.Runtime.loadLibrary0(Runtime.java:870) at java.lang.System.loadLibrary(System.java:1119) at sun.font.FontManagerNativeLibrary$1.run(FontManagerNativeLibrary.java:61) at java.security.AccessController.doPrivileged(Native Method) at sun.font.FontManagerNativeLibrary.(FontManagerNativeLibrary.java:32) at sun.font.SunFontManager$1.run(SunFontManager.java:339) at java.security.AccessController.doPrivileged(Native Method) at sun.font.SunFontManager.(SunFontManager.java:335) at java.lang.Class.forName0(Native Method) at java.lang.Class.forName(Class.java:340) at sun.font.FontManagerFactory$1.run(FontManagerFactory.java:82) at java.security.AccessController.doPrivileged(Native Method) at sun.font.FontManagerFactory.getInstance(FontManagerFactory.java:74) at java.awt.Font.getFont2D(Font.java:491) at java.awt.Font.getFamily(Font.java:1220) at java.awt.Font.getFamily_NoClientCode(Font.java:1194) at java.awt.Font.getFamily(Font.java:1186) at java.awt.Font.toString(Font.java:1683) at hudson.util.ChartUtil.(ChartUtil.java:229) at hudson.WebAppMain.contextInitialized(WebAppMain.java:181) ... 19 more Oct 03, 2014 3:45:04 AM org.eclipse.jetty.util.log.JavaUtilLog warn WARNING: Failed startup of context w.{,file:/export/home/jenkins/.jenkins/war/},/export/home/jenkins/.jenkins/war java.lang.NullPointerException at jenkins.util.groovy.GroovyHookScript.run(GroovyHookScript.java:63) at hudson.util.BootFailure.publish(BootFailure.java:43) at hudson.WebAppMain.contextInitialized(WebAppMain.java:244) at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) at org.eclipse.jetty.server.Server.doStart(Server.java:282) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at winstone.Launcher.(Launcher.java:154) at winstone.Launcher.main(Launcher.java:354) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:483) at Main._main(Main.java:293) at Main.main(Main.java:98) Oct 03, 2014 3:51:05 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: Started SelectChannelConnector at 0.0.0.0:8080 Oct 03, 2014 3:51:05 AM winstone.Logger logInternal INFO: Winstone Servlet Engine v2.0 running: controlPort=disabled ^COct 03, 2014 3:51:13 AM winstone.Logger logInternal INFO: JVM is terminating. Shutting down Winstone From gearboxes at outlook.com Fri Oct 3 04:51:10 2014 From: gearboxes at outlook.com (Machine Man) Date: Fri, 3 Oct 2014 00:51:10 -0400 Subject: [OmniOS-discuss] KVM disk performance r151012 Message-ID: After upgrading disk performance dropped from avg. 60MB/s to 17MB/s when copying over the network to a windows vm.Booting back to pre-upgrade and performance is restored Would this be resolved if the pool is upgraded? -------------- next part -------------- An HTML attachment was scrubbed... URL: From tobi at oetiker.ch Fri Oct 3 05:02:05 2014 From: tobi at oetiker.ch (Tobi Oetiker) Date: Fri, 3 Oct 2014 07:02:05 +0200 Subject: [OmniOS-discuss] KVM disk performance r151012 In-Reply-To: References: Message-ID: <546FCD5D-A3F9-438D-AA18-22A2F0FB8BC2@oetiker.ch> > On 03.10.2014, at 06:51, Machine Man wrote: > > After upgrading disk performance dropped from avg. 60MB/s to 17MB/s when copying over the network to a windows vm. > Booting back to pre-upgrade and performance is restored > > Would this be resolved if the pool is upgraded? > are you sure it is the disk and not the network? cheers tobi > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From ben at fluffy.co.uk Fri Oct 3 08:01:45 2014 From: ben at fluffy.co.uk (Ben Summers) Date: Fri, 3 Oct 2014 09:01:45 +0100 Subject: [OmniOS-discuss] Jenkins CI on OmniOS? (NPE) In-Reply-To: <356698EA-D360-49F9-90A1-F5F422FEE087@yonderway.com> References: <356698EA-D360-49F9-90A1-F5F422FEE087@yonderway.com> Message-ID: <21FD24CE-6691-48E6-9E15-10AF28438575@fluffy.co.uk> On 3 Oct 2014, at 04:52, Magnus wrote: > So I?ve just done this same thing successfully on a number of other platforms (that I don?t wish to run on), including a Macbook, a Debian laptop, a SmartOS zone? but when I try to do this inside of a zone on OmniOS, I get a Java stack trace that doesn?t make a whole lot of sense to me? and Google isn?t being helpful with this one. I?m hoping to write a blog about hosting CI on OmniOS, complete with getting it running under SMF, etc. but this one piece is being stubborn. > > Thoughts? OmniOS is a server OS, and doesn't include X or any of the related UI components. Try adding -Djava.awt.headless=true to the java command line, which may stop it attempting to load these things. Although if Jenkins actually requires them, it will just change the exception. Ben > > $ java -jar jenkins.war > Running from: /export/home/jenkins/jenkins.war > webroot: $user.home/.jenkins > Oct 03, 2014 3:28:19 AM winstone.Logger logInternal > INFO: Beginning extraction from war file > Oct 03, 2014 3:28:20 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: jetty-8.y.z-SNAPSHOT > Oct 03, 2014 3:28:22 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: NO JSP Support for , did not find org.apache.jasper.servlet.JspServlet > Jenkins home directory: /export/home/jenkins/.jenkins found at: $user.home/.jenkins > Oct 03, 2014 3:28:23 AM hudson.util.BootFailure publish > SEVERE: Failed to initialize Jenkins > hudson.util.AWTProblem: java.lang.NullPointerException > at hudson.WebAppMain.contextInitialized(WebAppMain.java:182) > at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) > at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) > at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) > at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) > at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) > at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) > at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) > at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) > at org.eclipse.jetty.server.Server.doStart(Server.java:282) > at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at winstone.Launcher.(Launcher.java:154) > at winstone.Launcher.main(Launcher.java:354) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) > at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) > at java.lang.reflect.Method.invoke(Method.java:606) > at Main._main(Main.java:293) > at Main.main(Main.java:98) > Caused by: java.lang.NullPointerException > at sun.awt.X11FontManager.getDefaultPlatformFont(X11FontManager.java:779) > at sun.font.SunFontManager$2.run(SunFontManager.java:433) > at java.security.AccessController.doPrivileged(Native Method) > at sun.font.SunFontManager.(SunFontManager.java:376) > at sun.awt.X11FontManager.(X11FontManager.java:32) > at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) > at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:57) > at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) > at java.lang.reflect.Constructor.newInstance(Constructor.java:526) > at java.lang.Class.newInstance(Class.java:374) > at sun.font.FontManagerFactory$1.run(FontManagerFactory.java:83) > at java.security.AccessController.doPrivileged(Native Method) > at sun.font.FontManagerFactory.getInstance(FontManagerFactory.java:74) > at java.awt.Font.getFont2D(Font.java:490) > at java.awt.Font.getFamily(Font.java:1219) > at java.awt.Font.getFamily_NoClientCode(Font.java:1193) > at java.awt.Font.getFamily(Font.java:1185) > at java.awt.Font.toString(Font.java:1682) > at hudson.util.ChartUtil.(ChartUtil.java:229) > at hudson.WebAppMain.contextInitialized(WebAppMain.java:181) > ... 19 more > > Oct 03, 2014 3:28:23 AM org.eclipse.jetty.util.log.JavaUtilLog warn > WARNING: Failed startup of context w.{,file:/export/home/jenkins/.jenkins/war/},/export/home/jenkins/.jenkins/war > java.lang.NullPointerException > at jenkins.util.groovy.GroovyHookScript.run(GroovyHookScript.java:63) > at hudson.util.BootFailure.publish(BootFailure.java:43) > at hudson.WebAppMain.contextInitialized(WebAppMain.java:244) > at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) > at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) > at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) > at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) > at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) > at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) > at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) > at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) > at org.eclipse.jetty.server.Server.doStart(Server.java:282) > at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at winstone.Launcher.(Launcher.java:154) > at winstone.Launcher.main(Launcher.java:354) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) > at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) > at java.lang.reflect.Method.invoke(Method.java:606) > at Main._main(Main.java:293) > at Main.main(Main.java:98) > > Oct 03, 2014 3:28:23 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: Started SelectChannelConnector at 0.0.0.0:8080 > Oct 03, 2014 3:28:23 AM winstone.Logger logInternal > INFO: Winstone Servlet Engine v2.0 running: controlPort=disabled > > *********** > > If I try the same thing with an Oracle JRE, I get a different error: > > $ /opt/jre1.8.0_20/bin/java -jar jenkins.war > Running from: /export/home/jenkins/jenkins.war > webroot: $user.home/.jenkins > Oct 03, 2014 3:45:00 AM winstone.Logger logInternal > INFO: Beginning extraction from war file > Oct 03, 2014 3:45:01 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: jetty-8.y.z-SNAPSHOT > Oct 03, 2014 3:45:03 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: NO JSP Support for , did not find org.apache.jasper.servlet.JspServlet > Jenkins home directory: /export/home/jenkins/.jenkins found at: $user.home/.jenkins > Oct 03, 2014 3:45:03 AM hudson.util.BootFailure publish > SEVERE: Failed to initialize Jenkins > hudson.util.AWTProblem: java.lang.UnsatisfiedLinkError: /opt/jre1.8.0_20/lib/amd64/libfontmanager.so: ld.so.1: java: fatal: libXext.so.0: open failed: No such file or directory > at hudson.WebAppMain.contextInitialized(WebAppMain.java:182) > at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) > at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) > at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) > at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) > at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) > at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) > at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) > at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) > at org.eclipse.jetty.server.Server.doStart(Server.java:282) > at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at winstone.Launcher.(Launcher.java:154) > at winstone.Launcher.main(Launcher.java:354) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) > at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) > at java.lang.reflect.Method.invoke(Method.java:483) > at Main._main(Main.java:293) > at Main.main(Main.java:98) > Caused by: java.lang.UnsatisfiedLinkError: /opt/jre1.8.0_20/lib/amd64/libfontmanager.so: ld.so.1: java: fatal: libXext.so.0: open failed: No such file or directory > at java.lang.ClassLoader$NativeLibrary.load(Native Method) > at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1929) > at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1835) > at java.lang.Runtime.loadLibrary0(Runtime.java:870) > at java.lang.System.loadLibrary(System.java:1119) > at sun.font.FontManagerNativeLibrary$1.run(FontManagerNativeLibrary.java:61) > at java.security.AccessController.doPrivileged(Native Method) > at sun.font.FontManagerNativeLibrary.(FontManagerNativeLibrary.java:32) > at sun.font.SunFontManager$1.run(SunFontManager.java:339) > at java.security.AccessController.doPrivileged(Native Method) > at sun.font.SunFontManager.(SunFontManager.java:335) > at java.lang.Class.forName0(Native Method) > at java.lang.Class.forName(Class.java:340) > at sun.font.FontManagerFactory$1.run(FontManagerFactory.java:82) > at java.security.AccessController.doPrivileged(Native Method) > at sun.font.FontManagerFactory.getInstance(FontManagerFactory.java:74) > at java.awt.Font.getFont2D(Font.java:491) > at java.awt.Font.getFamily(Font.java:1220) > at java.awt.Font.getFamily_NoClientCode(Font.java:1194) > at java.awt.Font.getFamily(Font.java:1186) > at java.awt.Font.toString(Font.java:1683) > at hudson.util.ChartUtil.(ChartUtil.java:229) > at hudson.WebAppMain.contextInitialized(WebAppMain.java:181) > ... 19 more > > Oct 03, 2014 3:45:04 AM org.eclipse.jetty.util.log.JavaUtilLog warn > WARNING: Failed startup of context w.{,file:/export/home/jenkins/.jenkins/war/},/export/home/jenkins/.jenkins/war > java.lang.NullPointerException > at jenkins.util.groovy.GroovyHookScript.run(GroovyHookScript.java:63) > at hudson.util.BootFailure.publish(BootFailure.java:43) > at hudson.WebAppMain.contextInitialized(WebAppMain.java:244) > at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) > at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) > at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) > at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) > at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) > at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) > at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) > at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) > at org.eclipse.jetty.server.Server.doStart(Server.java:282) > at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at winstone.Launcher.(Launcher.java:154) > at winstone.Launcher.main(Launcher.java:354) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) > at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) > at java.lang.reflect.Method.invoke(Method.java:483) > at Main._main(Main.java:293) > at Main.main(Main.java:98) > > Oct 03, 2014 3:51:05 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: Started SelectChannelConnector at 0.0.0.0:8080 > Oct 03, 2014 3:51:05 AM winstone.Logger logInternal > INFO: Winstone Servlet Engine v2.0 running: controlPort=disabled > ^COct 03, 2014 3:51:13 AM winstone.Logger logInternal > INFO: JVM is terminating. Shutting down Winstone > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss -- http://bens.me.uk From dan at syneto.net Fri Oct 3 08:09:58 2014 From: dan at syneto.net (Dan Vatca) Date: Fri, 3 Oct 2014 11:09:58 +0300 Subject: [OmniOS-discuss] Jenkins CI on OmniOS? (NPE) In-Reply-To: <356698EA-D360-49F9-90A1-F5F422FEE087@yonderway.com> References: <356698EA-D360-49F9-90A1-F5F422FEE087@yonderway.com> Message-ID: Hi Magnus, The user you are installing jenkins with (called jenkins) has configured his home directory as /export/home/jenkins, and that directory does not exist. Create a filesystem for it (or just a directory) and then change the user's home directory to that directory using usermod. Also change the owner of that directory to be jenkins (using chown). Hope this helps, Dan. On Fri, Oct 3, 2014 at 6:52 AM, Magnus wrote: > So I?ve just done this same thing successfully on a number of other > platforms (that I don?t wish to run on), including a Macbook, a Debian > laptop, a SmartOS zone? but when I try to do this inside of a zone on > OmniOS, I get a Java stack trace that doesn?t make a whole lot of sense to > me? and Google isn?t being helpful with this one. I?m hoping to write a > blog about hosting CI on OmniOS, complete with getting it running under > SMF, etc. but this one piece is being stubborn. > > Thoughts? > > $ java -jar jenkins.war > Running from: /export/home/jenkins/jenkins.war > webroot: $user.home/.jenkins > Oct 03, 2014 3:28:19 AM winstone.Logger logInternal > INFO: Beginning extraction from war file > Oct 03, 2014 3:28:20 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: jetty-8.y.z-SNAPSHOT > Oct 03, 2014 3:28:22 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: NO JSP Support for , did not find > org.apache.jasper.servlet.JspServlet > Jenkins home directory: /export/home/jenkins/.jenkins found at: > $user.home/.jenkins > Oct 03, 2014 3:28:23 AM hudson.util.BootFailure publish > SEVERE: Failed to initialize Jenkins > hudson.util.AWTProblem: java.lang.NullPointerException > at hudson.WebAppMain.contextInitialized(WebAppMain.java:182) > at > org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) > at > org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) > at > org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) > at > org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) > at > org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) > at > org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) > at > org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) > at > org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at > org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) > at org.eclipse.jetty.server.Server.doStart(Server.java:282) > at > org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at winstone.Launcher.(Launcher.java:154) > at winstone.Launcher.main(Launcher.java:354) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at > sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) > at > sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) > at java.lang.reflect.Method.invoke(Method.java:606) > at Main._main(Main.java:293) > at Main.main(Main.java:98) > Caused by: java.lang.NullPointerException > at > sun.awt.X11FontManager.getDefaultPlatformFont(X11FontManager.java:779) > at sun.font.SunFontManager$2.run(SunFontManager.java:433) > at java.security.AccessController.doPrivileged(Native Method) > at sun.font.SunFontManager.(SunFontManager.java:376) > at sun.awt.X11FontManager.(X11FontManager.java:32) > at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native > Method) > at > sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:57) > at > sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) > at java.lang.reflect.Constructor.newInstance(Constructor.java:526) > at java.lang.Class.newInstance(Class.java:374) > at sun.font.FontManagerFactory$1.run(FontManagerFactory.java:83) > at java.security.AccessController.doPrivileged(Native Method) > at > sun.font.FontManagerFactory.getInstance(FontManagerFactory.java:74) > at java.awt.Font.getFont2D(Font.java:490) > at java.awt.Font.getFamily(Font.java:1219) > at java.awt.Font.getFamily_NoClientCode(Font.java:1193) > at java.awt.Font.getFamily(Font.java:1185) > at java.awt.Font.toString(Font.java:1682) > at hudson.util.ChartUtil.(ChartUtil.java:229) > at hudson.WebAppMain.contextInitialized(WebAppMain.java:181) > ... 19 more > > Oct 03, 2014 3:28:23 AM org.eclipse.jetty.util.log.JavaUtilLog warn > WARNING: Failed startup of context > w.{,file:/export/home/jenkins/.jenkins/war/},/export/home/jenkins/.jenkins/war > java.lang.NullPointerException > at > jenkins.util.groovy.GroovyHookScript.run(GroovyHookScript.java:63) > at hudson.util.BootFailure.publish(BootFailure.java:43) > at hudson.WebAppMain.contextInitialized(WebAppMain.java:244) > at > org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) > at > org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) > at > org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) > at > org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) > at > org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) > at > org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) > at > org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) > at > org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at > org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) > at org.eclipse.jetty.server.Server.doStart(Server.java:282) > at > org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at winstone.Launcher.(Launcher.java:154) > at winstone.Launcher.main(Launcher.java:354) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at > sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) > at > sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) > at java.lang.reflect.Method.invoke(Method.java:606) > at Main._main(Main.java:293) > at Main.main(Main.java:98) > > Oct 03, 2014 3:28:23 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: Started SelectChannelConnector at 0.0.0.0:8080 > Oct 03, 2014 3:28:23 AM winstone.Logger logInternal > INFO: Winstone Servlet Engine v2.0 running: controlPort=disabled > > *********** > > If I try the same thing with an Oracle JRE, I get a different error: > > $ /opt/jre1.8.0_20/bin/java -jar jenkins.war > Running from: /export/home/jenkins/jenkins.war > webroot: $user.home/.jenkins > Oct 03, 2014 3:45:00 AM winstone.Logger logInternal > INFO: Beginning extraction from war file > Oct 03, 2014 3:45:01 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: jetty-8.y.z-SNAPSHOT > Oct 03, 2014 3:45:03 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: NO JSP Support for , did not find > org.apache.jasper.servlet.JspServlet > Jenkins home directory: /export/home/jenkins/.jenkins found at: > $user.home/.jenkins > Oct 03, 2014 3:45:03 AM hudson.util.BootFailure publish > SEVERE: Failed to initialize Jenkins > hudson.util.AWTProblem: java.lang.UnsatisfiedLinkError: > /opt/jre1.8.0_20/lib/amd64/libfontmanager.so: ld.so.1: java: fatal: > libXext.so.0: open failed: No such file or directory > at hudson.WebAppMain.contextInitialized(WebAppMain.java:182) > at > org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) > at > org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) > at > org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) > at > org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) > at > org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) > at > org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) > at > org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) > at > org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at > org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) > at org.eclipse.jetty.server.Server.doStart(Server.java:282) > at > org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at winstone.Launcher.(Launcher.java:154) > at winstone.Launcher.main(Launcher.java:354) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at > sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) > at > sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) > at java.lang.reflect.Method.invoke(Method.java:483) > at Main._main(Main.java:293) > at Main.main(Main.java:98) > Caused by: java.lang.UnsatisfiedLinkError: > /opt/jre1.8.0_20/lib/amd64/libfontmanager.so: ld.so.1: java: fatal: > libXext.so.0: open failed: No such file or directory > at java.lang.ClassLoader$NativeLibrary.load(Native Method) > at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1929) > at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1835) > at java.lang.Runtime.loadLibrary0(Runtime.java:870) > at java.lang.System.loadLibrary(System.java:1119) > at > sun.font.FontManagerNativeLibrary$1.run(FontManagerNativeLibrary.java:61) > at java.security.AccessController.doPrivileged(Native Method) > at > sun.font.FontManagerNativeLibrary.(FontManagerNativeLibrary.java:32) > at sun.font.SunFontManager$1.run(SunFontManager.java:339) > at java.security.AccessController.doPrivileged(Native Method) > at sun.font.SunFontManager.(SunFontManager.java:335) > at java.lang.Class.forName0(Native Method) > at java.lang.Class.forName(Class.java:340) > at sun.font.FontManagerFactory$1.run(FontManagerFactory.java:82) > at java.security.AccessController.doPrivileged(Native Method) > at > sun.font.FontManagerFactory.getInstance(FontManagerFactory.java:74) > at java.awt.Font.getFont2D(Font.java:491) > at java.awt.Font.getFamily(Font.java:1220) > at java.awt.Font.getFamily_NoClientCode(Font.java:1194) > at java.awt.Font.getFamily(Font.java:1186) > at java.awt.Font.toString(Font.java:1683) > at hudson.util.ChartUtil.(ChartUtil.java:229) > at hudson.WebAppMain.contextInitialized(WebAppMain.java:181) > ... 19 more > > Oct 03, 2014 3:45:04 AM org.eclipse.jetty.util.log.JavaUtilLog warn > WARNING: Failed startup of context > w.{,file:/export/home/jenkins/.jenkins/war/},/export/home/jenkins/.jenkins/war > java.lang.NullPointerException > at > jenkins.util.groovy.GroovyHookScript.run(GroovyHookScript.java:63) > at hudson.util.BootFailure.publish(BootFailure.java:43) > at hudson.WebAppMain.contextInitialized(WebAppMain.java:244) > at > org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) > at > org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) > at > org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) > at > org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) > at > org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) > at > org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) > at > org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) > at > org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at > org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) > at org.eclipse.jetty.server.Server.doStart(Server.java:282) > at > org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) > at winstone.Launcher.(Launcher.java:154) > at winstone.Launcher.main(Launcher.java:354) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at > sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) > at > sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) > at java.lang.reflect.Method.invoke(Method.java:483) > at Main._main(Main.java:293) > at Main.main(Main.java:98) > > Oct 03, 2014 3:51:05 AM org.eclipse.jetty.util.log.JavaUtilLog info > INFO: Started SelectChannelConnector at 0.0.0.0:8080 > Oct 03, 2014 3:51:05 AM winstone.Logger logInternal > INFO: Winstone Servlet Engine v2.0 running: controlPort=disabled > ^COct 03, 2014 3:51:13 AM winstone.Logger logInternal > INFO: JVM is terminating. Shutting down Winstone > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sjorge+ml at blackdot.be Fri Oct 3 08:59:43 2014 From: sjorge+ml at blackdot.be (Jorge Schrauwen) Date: Fri, 03 Oct 2014 10:59:43 +0200 Subject: [OmniOS-discuss] KVM disk performance r151012 In-Reply-To: References: Message-ID: <072017d3ce0ba6769b555d36be724f98@blackdot.be> To your first question, upgrading your zpool will probably not fix it. If the vm is on your rpool, your old BE might also become unbootable (due to feature flag changes!) so watch out. Actually this reminded me of '[OmniOS-discuss] OmniOS r151012 vioblk not working with ISO' When I fell back to ide mode it would still sort of hang for hours. Switching back to 010 did make it boot on ide, still taking ages so I attributed it to OmniOS on OmniOS+KVM was just a bad idea and I found an alternative. But that detour aside... do you have the windows virtio drivers installed? If not, try installing them and switching your nic to virtio. Jorge On 2014-10-03 06:51, Machine Man wrote: > After upgrading disk performance dropped from avg. 60MB/s to 17MB/s > when copying over the network to a windows vm. > Booting back to pre-upgrade and performance is restored > > Would this be resolved if the pool is upgraded? > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss [1] Links: ------ [1] http://lists.omniti.com/mailman/listinfo/omnios-discuss From magnus at yonderway.com Fri Oct 3 11:51:46 2014 From: magnus at yonderway.com (Magnus) Date: Fri, 3 Oct 2014 07:51:46 -0400 Subject: [OmniOS-discuss] Jenkins CI on OmniOS? (NPE) In-Reply-To: <21FD24CE-6691-48E6-9E15-10AF28438575@fluffy.co.uk> References: <356698EA-D360-49F9-90A1-F5F422FEE087@yonderway.com> <21FD24CE-6691-48E6-9E15-10AF28438575@fluffy.co.uk> Message-ID: On Oct 3, 2014, at 4:01 AM, Ben Summers wrote: > > OmniOS is a server OS, and doesn't include X or any of the related UI components. I don?t believe SmartOS ships with any of that stuff, either, but it worked over there. > > Try adding > > -Djava.awt.headless=true > > to the java command line, which may stop it attempting to load these things. Although if Jenkins actually requires them, it will just change the exception. jenkins at jenkins:/export/home/jenkins$ java -Djava.awt.headless=true -jar jenkins.war Running from: /export/home/jenkins/jenkins.war webroot: $user.home/.jenkins Oct 03, 2014 11:48:35 AM winstone.Logger logInternal INFO: Beginning extraction from war file Oct 03, 2014 11:48:35 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: jetty-8.y.z-SNAPSHOT Oct 03, 2014 11:48:37 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: NO JSP Support for , did not find org.apache.jasper.servlet.JspServlet Jenkins home directory: /export/home/jenkins/.jenkins found at: $user.home/.jenkins Oct 03, 2014 11:48:38 AM hudson.util.BootFailure publish SEVERE: Failed to initialize Jenkins hudson.util.AWTProblem: java.lang.NullPointerException at hudson.WebAppMain.contextInitialized(WebAppMain.java:182) at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) at org.eclipse.jetty.server.Server.doStart(Server.java:282) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at winstone.Launcher.(Launcher.java:154) at winstone.Launcher.main(Launcher.java:354) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:606) at Main._main(Main.java:293) at Main.main(Main.java:98) Caused by: java.lang.NullPointerException at sun.awt.X11FontManager.getDefaultPlatformFont(X11FontManager.java:779) at sun.font.SunFontManager$2.run(SunFontManager.java:433) at java.security.AccessController.doPrivileged(Native Method) at sun.font.SunFontManager.(SunFontManager.java:376) at sun.awt.X11FontManager.(X11FontManager.java:32) at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:57) at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) at java.lang.reflect.Constructor.newInstance(Constructor.java:526) at java.lang.Class.newInstance(Class.java:374) at sun.font.FontManagerFactory$1.run(FontManagerFactory.java:83) at java.security.AccessController.doPrivileged(Native Method) at sun.font.FontManagerFactory.getInstance(FontManagerFactory.java:74) at java.awt.Font.getFont2D(Font.java:490) at java.awt.Font.getFamily(Font.java:1219) at java.awt.Font.getFamily_NoClientCode(Font.java:1193) at java.awt.Font.getFamily(Font.java:1185) at java.awt.Font.toString(Font.java:1682) at hudson.util.ChartUtil.(ChartUtil.java:229) at hudson.WebAppMain.contextInitialized(WebAppMain.java:181) ... 19 more Oct 03, 2014 11:48:38 AM org.eclipse.jetty.util.log.JavaUtilLog warn WARNING: Failed startup of context w.{,file:/export/home/jenkins/.jenkins/war/},/export/home/jenkins/.jenkins/war java.lang.NullPointerException at jenkins.util.groovy.GroovyHookScript.run(GroovyHookScript.java:63) at hudson.util.BootFailure.publish(BootFailure.java:43) at hudson.WebAppMain.contextInitialized(WebAppMain.java:244) at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) at org.eclipse.jetty.server.Server.doStart(Server.java:282) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at winstone.Launcher.(Launcher.java:154) at winstone.Launcher.main(Launcher.java:354) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:606) at Main._main(Main.java:293) at Main.main(Main.java:98) Oct 03, 2014 11:48:38 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: Started SelectChannelConnector at 0.0.0.0:8080 Oct 03, 2014 11:48:38 AM winstone.Logger logInternal INFO: Winstone Servlet Engine v2.0 running: controlPort=disabled From magnus at yonderway.com Fri Oct 3 11:54:51 2014 From: magnus at yonderway.com (Magnus) Date: Fri, 3 Oct 2014 07:54:51 -0400 Subject: [OmniOS-discuss] Jenkins CI on OmniOS? (NPE) In-Reply-To: References: <356698EA-D360-49F9-90A1-F5F422FEE087@yonderway.com> Message-ID: On Oct 3, 2014, at 4:09 AM, Dan Vatca wrote: > > > The user you are installing jenkins with (called jenkins) has configured his home directory as /export/home/jenkins, and that directory does not exist. Create a filesystem for it (or just a directory) and then change the user's home directory to that directory using usermod. Also change the owner of that directory to be jenkins (using chown). > Hope this helps, Hi Dan, On the surface that might sound true. But? jenkins at jenkins:/export/home/jenkins$ pwd /export/home/jenkins -and- jenkins at jenkins:/export/home/jenkins$ cd .. jenkins at jenkins:/export/home$ ls -la | grep jenkins drwx------ 4 jenkins other 14 Oct 3 03:45 jenkins Strange, huh? -M From dan at syneto.net Fri Oct 3 12:11:03 2014 From: dan at syneto.net (Dan Vatca) Date: Fri, 3 Oct 2014 15:11:03 +0300 Subject: [OmniOS-discuss] Jenkins CI on OmniOS? (NPE) In-Reply-To: References: <356698EA-D360-49F9-90A1-F5F422FEE087@yonderway.com> Message-ID: Strange indeed. This might be related to Jenkins scripts expecting /bin/sh to be bash, while on illumos it is not. I've seen build/installation scripts in the past (especially coming from the linux world) that assume this and fail. Dan. On Fri, Oct 3, 2014 at 2:54 PM, Magnus wrote: > > On Oct 3, 2014, at 4:09 AM, Dan Vatca wrote: > > > > > > The user you are installing jenkins with (called jenkins) has configured > his home directory as /export/home/jenkins, and that directory does not > exist. Create a filesystem for it (or just a directory) and then change the > user's home directory to that directory using usermod. Also change the > owner of that directory to be jenkins (using chown). > > Hope this helps, > > Hi Dan, > > On the surface that might sound true. But? > > jenkins at jenkins:/export/home/jenkins$ pwd > /export/home/jenkins > > -and- > > jenkins at jenkins:/export/home/jenkins$ cd .. > jenkins at jenkins:/export/home$ ls -la | grep jenkins > drwx------ 4 jenkins other 14 Oct 3 03:45 jenkins > > Strange, huh? > > -M > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: From fabio at fabiorabelo.wiki.br Fri Oct 3 13:25:35 2014 From: fabio at fabiorabelo.wiki.br (=?UTF-8?Q?F=C3=A1bio_Rabelo?=) Date: Fri, 3 Oct 2014 10:25:35 -0300 Subject: [OmniOS-discuss] how many HBAs my system have ? Message-ID: How can I check how many HBA controlres do I have connected in a system, if this system are in a ( very ) remote location, and I just have SSH available ? prtconf -v returns too many noise ! prtconv -v |grep RAID returns only one line ... There is an easy way ? F?bio Rabelo From groups at tierarzt-mueller.de Fri Oct 3 13:29:26 2014 From: groups at tierarzt-mueller.de (Alexander Lesle) Date: Fri, 3 Oct 2014 15:29:26 +0200 Subject: [OmniOS-discuss] feature request Message-ID: <18461750.20141003152926@tierarzt-mueller.de> Hello Dan McDonald and List, in my mind this was a great proposal what you done at Sep, 02. That's the way to build a great OmniOS near to the users. Here my request for the next release: It would be convenient if open-vm-tools were installed in the new release. http://sourceforge.net/projects/open-vm-tools/files/ Its very convenient and helpful when you install OmniOS new the drivers are available. Since ESXi 5.5 U2 VMware supports now open-vm-tools. http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=2073803 The second way is to put it in the repository. Thanks. On September, 02 2014, 20:22 wrote in [1]: > I'm now starting to wind up the r151012 build process. You'll see > me dump updates here occasionally. > What I'd like to know now is: > Any updates you need/want in r151012? > If you've been keeping up with my bloody announcements, everything > you see there will be landing in r151012. This includes HW goodies > like LSI 3008-based 12G SAS (albeit not at optimal performance yet), > for example. Once this week's bloody update goes out, there will be > a lot of updates to both userspace and in the kernel. > I'd like to hear if anyone here has suggestions. I cannot > guarantee I will take 'em, but I can at least explain why-not if I don't. > Thanks, > Dan McD. - OmniOS Engineering -- Best Regards Alexander Oktober, 03 2014 ........ [1] mid:E1B54D4D-59E7-4E32-94DA-22E22DBCA015 at omniti.com ........ From mbi at losstech.de Fri Oct 3 13:32:18 2014 From: mbi at losstech.de (Bielicki, Michal) Date: Fri, 3 Oct 2014 13:32:18 +0000 Subject: [OmniOS-discuss] how many HBAs my system have ? In-Reply-To: References: Message-ID: <288CCB8B-5559-4F33-A66F-520B8F525933@losstech.de> sasinfo is your friend. Kind regards / Gr??e aus dem hohen Norden Michal Bielicki [cid:1C01820A-2F8C-422C-9089-72CFBE601F8E at fritz.box][cid:9DCD0525-2332-419B-AA48-F175C80366D4 at fritz.box] Aktuelle Publikationen: www.losstech-blog.de losstech GmbH Hauptstra?e 24, DE-25355 Lutzhorn mailto:mbi at losstech.de +49 (0)4123 929900 (fon) +49 (0)4123 929911 (fax) Amtsgericht Pinneberg, HRB 5737 USt-Id: DE 227 8636 32 Hamburger Sparkasse, KTO 1217 149 150, BLZ 200 505 50 Sparkasse S?dholstein, KTO 151 228 56, BLZ 230 510 30 This message and any attachment is losstech proprietary and may be privileged or otherwise protected from disclosure. If you are not the intended recipient, please phone or email the sender and delete this message and any attachment from your system. If you are not the intended recipient you are not allowed to copy this message or attachment or disclose the contents to any other person and have to destroy all copies of the original message. On 03 Oct 2014, at 15:25, F?bio Rabelo > wrote: How can I check how many HBA controlres do I have connected in a system, if this system are in a ( very ) remote location, and I just have SSH available ? prtconf -v returns too many noise ! prtconv -v |grep RAID returns only one line ... There is an easy way ? F?bio Rabelo _______________________________________________ OmniOS-discuss mailing list OmniOS-discuss at lists.omniti.com http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 10803 bytes Desc: image001.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.jpg Type: image/jpeg Size: 3874 bytes Desc: image002.jpg URL: From fabio at fabiorabelo.wiki.br Fri Oct 3 13:46:19 2014 From: fabio at fabiorabelo.wiki.br (=?UTF-8?Q?F=C3=A1bio_Rabelo?=) Date: Fri, 3 Oct 2014 10:46:19 -0300 Subject: [OmniOS-discuss] how many HBAs my system have ? In-Reply-To: <288CCB8B-5559-4F33-A66F-520B8F525933@losstech.de> References: <288CCB8B-5559-4F33-A66F-520B8F525933@losstech.de> Message-ID: Thanks for the tip ... Well, it would be my friend if it was present in my system : sasinfo: command not found F?bio Rabelo 2014-10-03 10:32 GMT-03:00 Bielicki, Michal : > sasinfo is your friend. > > Kind regards / > Gr??e aus dem hohen Norden > > Michal Bielicki > > > > > > > > Aktuelle Publikationen: > www.losstech-blog.de > > > > losstech GmbH > Hauptstra?e 24, DE-25355 Lutzhorn > > mailto:mbi at losstech.de > +49 (0)4123 929900 (fon) > +49 (0)4123 929911 (fax) > > > Amtsgericht Pinneberg, HRB 5737 > USt-Id: DE 227 8636 32 > > Hamburger Sparkasse, KTO 1217 149 150, BLZ 200 505 50 > Sparkasse S?dholstein, KTO 151 228 56, BLZ 230 510 30 > > *This message and any attachment is losstech proprietary and may be > privileged or otherwise protected from disclosure.* > *If you are not the intended recipient, please phone or email the sender > and delete this message and any attachment from your system.* > *If you are not the intended recipient you are not allowed to copy this > message or attachment or disclose the contents to any other person and have > to destroy all copies of the original message.* > > On 03 Oct 2014, at 15:25, F?bio Rabelo wrote: > > How can I check how many HBA controlres do I have connected in a > system, if this system are in a ( very ) remote location, and I just > have SSH available ? > > prtconf -v returns too many noise ! > > prtconv -v |grep RAID returns only one line ... > > There is an easy way ? > > > F?bio Rabelo > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 10803 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.jpg Type: image/jpeg Size: 3874 bytes Desc: not available URL: From esproul at omniti.com Fri Oct 3 14:13:11 2014 From: esproul at omniti.com (Eric Sproul) Date: Fri, 3 Oct 2014 10:13:11 -0400 Subject: [OmniOS-discuss] Jenkins CI on OmniOS? (NPE) In-Reply-To: References: <356698EA-D360-49F9-90A1-F5F422FEE087@yonderway.com> Message-ID: Magnus, As Theo mentioned on Twitter, Circonus runs Jenkins in a zone on OmniOS (which reminds me, I need to switch my list subscription...) We use the Oracle Java JDK for Solaris, which we have in an IPS package, but you can install it the "traditional" way too by just extracting the archives into whatever directory you want. $ getent passwd jenkins jenkins:x:10098:1:Jenkins CI User:/jenkins:/bin/bash We put jenkins's home directory on its own ZFS dataset for backup purposes. We set its default shell to bash for the reason Dan mentioned. Our SMF manifest looks like this: https://gist.github.com/esproul/2a0f9c9aeab84ba70620 Pretty self-explanatory, but a couple of notes: the net_privaddr privilege isn't strictly necessary here because we're binding to an unprivileged port, but it's there for completeness since we set up the methods in such a way that you could run multiple instances of Jenkins, some of them perhaps on low-numbered ports. We use a non-default process model (startd/duration property) because the process doesn't daemonize, and we ignore errors because Jenkins can restart itself for updates and such. See svc.startd(1M). Hope this helps, Eric From admdwrf at gmail.com Fri Oct 3 14:14:59 2014 From: admdwrf at gmail.com (=?UTF-8?B?RnLDqWTDqXJpYyBBbGl4?=) Date: Fri, 3 Oct 2014 16:14:59 +0200 Subject: [OmniOS-discuss] Jenkins CI on OmniOS? (NPE) Message-ID: It run well with pkgsrc. It contains all dependencies. curl -s http://pkgsrc.joyent.com/packages/SmartOS/bootstrap/bootstrap-2014Q2-x86_64.tar.gz | gzcat | (cd /; sudo tar -xpf -) PATH="/usr/gnu/bin:/usr/bin:/usr/sbin:/sbin:/opt/local/bin:/opt/local/sbin" pkgin install openjdk7-1.7.60nb3 /opt/local/java/openjdk7/bin/java -jar jenkins.war fralix at zdoctcmsfrb1:~/app2$ /opt/local/java/openjdk7/bin/java -jar jenkins.war Running from: /export/home/fralix/app2/jenkins.war webroot: $user.home/.jenkins Oct 03, 2014 12:45:25 PM winstone.Logger logInternal INFO: Beginning extraction from war file Oct 03, 2014 12:45:25 PM org.eclipse.jetty.util.log.JavaUtilLog info INFO: jetty-8.y.z-SNAPSHOT Oct 03, 2014 12:45:27 PM org.eclipse.jetty.util.log.JavaUtilLog info INFO: NO JSP Support for , did not find org.apache.jasper.servlet.JspServlet Jenkins home directory: /export/home/fralix/.jenkins found at: $user.home/.jenkins Oct 03, 2014 12:45:28 PM org.eclipse.jetty.util.log.JavaUtilLog info INFO: Started SelectChannelConnector at 0.0.0.0:8080 Oct 03, 2014 12:45:28 PM winstone.Logger logInternal INFO: Winstone Servlet Engine v2.0 running: controlPort=disabled Oct 03, 2014 12:45:28 PM jenkins.InitReactorRunner$1 onAttained INFO: Started initialization Oct 03, 2014 12:45:48 PM jenkins.InitReactorRunner$1 onAttained INFO: Listed all plugins Oct 03, 2014 12:45:48 PM jenkins.InitReactorRunner$1 onAttained INFO: Prepared all plugins Oct 03, 2014 12:45:48 PM jenkins.InitReactorRunner$1 onAttained INFO: Started all plugins Oct 03, 2014 12:45:48 PM jenkins.InitReactorRunner$1 onAttained INFO: Augmented all extensions Oct 03, 2014 12:45:48 PM jenkins.InitReactorRunner$1 onAttained INFO: Loaded all jobs Oct 03, 2014 12:45:51 PM org.jenkinsci.main.modules.sshd.SSHD start INFO: Started SSHD at port 59567 Oct 03, 2014 12:45:51 PM jenkins.InitReactorRunner$1 onAttained INFO: Completed initialization Oct 03, 2014 12:45:51 PM hudson.WebAppMain$3 run INFO: Jenkins is fully up and running On Oct 3, 2014, at 4:01 AM, Ben Summers > wrote: > >* OmniOS is a server OS, and doesn't include X or any of the related UI components. * I don?t believe SmartOS ships with any of that stuff, either, but it worked over there. > >* Try adding *> >* -Djava.awt.headless=true *> >* to the java command line, which may stop it attempting to load these things. Although if Jenkins actually requires them, it will just change the exception. * jenkins at jenkins :/export/home/jenkins$ java -Djava.awt.headless=true -jar jenkins.war Running from: /export/home/jenkins/jenkins.war webroot: $user.home/.jenkins Oct 03, 2014 11:48:35 AM winstone.Logger logInternal INFO: Beginning extraction from war file Oct 03, 2014 11:48:35 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: jetty-8.y.z-SNAPSHOT Oct 03, 2014 11:48:37 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: NO JSP Support for , did not find org.apache.jasper.servlet.JspServlet Jenkins home directory: /export/home/jenkins/.jenkins found at: $user.home/.jenkins Oct 03, 2014 11:48:38 AM hudson.util.BootFailure publish SEVERE: Failed to initialize Jenkins hudson.util.AWTProblem: java.lang.NullPointerException at hudson.WebAppMain.contextInitialized(WebAppMain.java:182) at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) at org.eclipse.jetty.server.Server.doStart(Server.java:282) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at winstone.Launcher.(Launcher.java:154) at winstone.Launcher.main(Launcher.java:354) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:606) at Main._main(Main.java:293) at Main.main(Main.java:98) Caused by: java.lang.NullPointerException at sun.awt.X11FontManager.getDefaultPlatformFont(X11FontManager.java:779) at sun.font.SunFontManager$2.run(SunFontManager.java:433) at java.security.AccessController.doPrivileged(Native Method) at sun.font.SunFontManager.(SunFontManager.java:376) at sun.awt.X11FontManager.(X11FontManager.java:32) at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:57) at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45) at java.lang.reflect.Constructor.newInstance(Constructor.java:526) at java.lang.Class.newInstance(Class.java:374) at sun.font.FontManagerFactory$1.run(FontManagerFactory.java:83) at java.security.AccessController.doPrivileged(Native Method) at sun.font.FontManagerFactory.getInstance(FontManagerFactory.java:74) at java.awt.Font.getFont2D(Font.java:490) at java.awt.Font.getFamily(Font.java:1219) at java.awt.Font.getFamily_NoClientCode(Font.java:1193) at java.awt.Font.getFamily(Font.java:1185) at java.awt.Font.toString(Font.java:1682) at hudson.util.ChartUtil.(ChartUtil.java:229) at hudson.WebAppMain.contextInitialized(WebAppMain.java:181) ... 19 more Oct 03, 2014 11:48:38 AM org.eclipse.jetty.util.log.JavaUtilLog warn WARNING: Failed startup of context w.{,file:/export/home/jenkins/.jenkins/war/},/export/home/jenkins/.jenkins/war java.lang.NullPointerException at jenkins.util.groovy.GroovyHookScript.run(GroovyHookScript.java:63) at hudson.util.BootFailure.publish(BootFailure.java:43) at hudson.WebAppMain.contextInitialized(WebAppMain.java:244) at org.eclipse.jetty.server.handler.ContextHandler.callContextInitialized(ContextHandler.java:782) at org.eclipse.jetty.servlet.ServletContextHandler.callContextInitialized(ServletContextHandler.java:424) at org.eclipse.jetty.server.handler.ContextHandler.startContext(ContextHandler.java:774) at org.eclipse.jetty.servlet.ServletContextHandler.startContext(ServletContextHandler.java:249) at org.eclipse.jetty.webapp.WebAppContext.startContext(WebAppContext.java:1242) at org.eclipse.jetty.server.handler.ContextHandler.doStart(ContextHandler.java:717) at org.eclipse.jetty.webapp.WebAppContext.doStart(WebAppContext.java:494) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at org.eclipse.jetty.server.handler.HandlerWrapper.doStart(HandlerWrapper.java:95) at org.eclipse.jetty.server.Server.doStart(Server.java:282) at org.eclipse.jetty.util.component.AbstractLifeCycle.start(AbstractLifeCycle.java:64) at winstone.Launcher.(Launcher.java:154) at winstone.Launcher.main(Launcher.java:354) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:606) at Main._main(Main.java:293) at Main.main(Main.java:98) Oct 03, 2014 11:48:38 AM org.eclipse.jetty.util.log.JavaUtilLog info INFO: Started SelectChannelConnector at 0.0.0.0 :8080 Oct 03, 2014 11:48:38 AM winstone.Logger logInternal INFO: Winstone Servlet Engine v2.0 running: controlPort=disabled -------------- next part -------------- An HTML attachment was scrubbed... URL: From tom-omnios-discuss at tom.bn-ulm.de Fri Oct 3 14:14:37 2014 From: tom-omnios-discuss at tom.bn-ulm.de (Thomas Wagner) Date: Fri, 3 Oct 2014 16:14:37 +0200 Subject: [OmniOS-discuss] how many HBAs my system have ? In-Reply-To: References: Message-ID: <20141003141436.GU24835@wagner-net.com> I like the display of "cfgadm". cfgadm Ap_Id Type Receptacle Occupant Condition sata2/0::dsk/c2t0d0 disk connected configured ok sata2/1::dsk/c2t1d0 disk connected configured ok sata2/2::dsk/c2t2d0 disk connected configured ok sata2/3::dsk/c2t3d0 cd/dvd connected configured ok sata2/4::dsk/c2t4d0 disk connected configured ok sata2/5::dsk/c2t5d0 disk connected configured ok In my system the controller is the onboard SATA, numbered as "sata2" with 6 ports. Note: Sometimes a controller might be present, but no driver attached. Then you don't see that on the list. The cfgadm manpage has examples how controllers are displayed (e.g. just Ap_Id is "c0"). To configure a controller you can run e.g. cfgadm -c configure c1 (controller numbers are incremented, getting a symlink in /dev/cfg/ and registered in /etc/path_to_inst.) So in the example you would have two, the first controllert already linked to /dev/cfg/c0 and the second controller being unconfigured, getting now the /dev/cfg/c1 link. Else, the best source is still "/usr/sbin/prtconf -vp | ggrep -2 model", even if there is too much information. It just never lies, as it displays all the PCI informations. Regards Thomas On Fri, Oct 03, 2014 at 10:25:35AM -0300, F??bio Rabelo wrote: > How can I check how many HBA controlres do I have connected in a > system, if this system are in a ( very ) remote location, and I just > have SSH available ? > > prtconf -v returns too many noise ! > > prtconv -v |grep RAID returns only one line ... > > There is an easy way ? > > > F??bio Rabelo > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -- -- Thomas Wagner ------------------------------------------------------------------------ Service rund um UNIX(TM), Wagner Network Services, Thomas Wagner Solaris(TM), Linux(TM) Eschenweg 21, 89174 Altheim, Germany Novell(TM), Windows(TM) TEL: +49-731-9807799, FAX: +49-731-9807711 Telekommunikation, LAN, MOBILE/CELL: +49-171-6135989 Internet-Service, Elektronik EMAIL: wagner at wagner-net.com From sjorge+ml at blackdot.be Fri Oct 3 14:17:39 2014 From: sjorge+ml at blackdot.be (Jorge Schrauwen) Date: Fri, 03 Oct 2014 16:17:39 +0200 Subject: [OmniOS-discuss] feature request In-Reply-To: <18461750.20141003152926@tierarzt-mueller.de> References: <18461750.20141003152926@tierarzt-mueller.de> Message-ID: <032735e728f325ab9642f592c1d8b095@blackdot.be> +1 on this... also as just discussed on #illumos... also bundling vmxnet/pvscsi would be nice as mentioned by patdk-wk. (And look as to why vioblk is not working, but that is on my extended todo list) Jorge On 2014-10-03 15:29, Alexander Lesle wrote: > Hello Dan McDonald and List, > > in my mind this was a great proposal what you done at Sep, 02. > That's the way to build a great OmniOS near to the users. > > Here my request for the next release: > It would be convenient if open-vm-tools were installed in the new > release. http://sourceforge.net/projects/open-vm-tools/files/ > Its very convenient and helpful when you install OmniOS new the > drivers are available. > Since ESXi 5.5 U2 VMware supports now open-vm-tools. > http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=2073803 > > The second way is to put it in the repository. > > Thanks. > > On September, 02 2014, 20:22 wrote in [1]: > >> I'm now starting to wind up the r151012 build process. You'll see >> me dump updates here occasionally. > >> What I'd like to know now is: > >> Any updates you need/want in r151012? > >> If you've been keeping up with my bloody announcements, everything >> you see there will be landing in r151012. This includes HW goodies >> like LSI 3008-based 12G SAS (albeit not at optimal performance yet), >> for example. Once this week's bloody update goes out, there will be >> a lot of updates to both userspace and in the kernel. > >> I'd like to hear if anyone here has suggestions. I cannot >> guarantee I will take 'em, but I can at least explain why-not if I >> don't. > >> Thanks, >> Dan McD. - OmniOS Engineering From sjorge+ml at blackdot.be Fri Oct 3 14:22:07 2014 From: sjorge+ml at blackdot.be (Jorge Schrauwen) Date: Fri, 03 Oct 2014 16:22:07 +0200 Subject: [OmniOS-discuss] =?utf-8?q?how_many_HBAs_my_system_have_=3F?= In-Reply-To: References: <288CCB8B-5559-4F33-A66F-520B8F525933@losstech.de> Message-ID: The 'pkg' command is your friend :) pkg search sasinfo, returns a package called sasinfo. pfexec pkg install sasinfo, would install it. After this sasinfo should just work. --- ~ sjorge On 2014-10-03 15:46, F?bio Rabelo wrote: > Thanks for the tip ... > > Well, it would be my friend if it was present in my system : > > sasinfo: command not found > > F?bio Rabelo > > 2014-10-03 10:32 GMT-03:00 Bielicki, Michal : > > sasinfo is your friend. > > Kind regards / > Gr??e aus dem hohen Norden > > Michal Bielicki > > [1] > > Aktuelle Publikationen: [1]www.losstech-blog.de [2] > > losstech GmbH > Hauptstra?e 24, DE-25355 Lutzhorn > > mailto:mbi at losstech.de > +49 (0)4123 929900 [3] (fon) > +49 (0)4123 929911 [4] (fax) > > Amtsgericht Pinneberg, HRB 5737 > USt-Id: DE 227 8636 32 > > Hamburger Sparkasse, KTO 1217 149 150, BLZ 200 505 50 > Sparkasse S?dholstein, KTO 151 228 56, BLZ 230 510 30 > > _This message and any attachment is losstech proprietary and may be > privileged or otherwise protected from disclosure._ > _If you are not the intended recipient, please phone or email the > sender and delete this message and any attachment from your system._ > _If you are not the intended recipient you are not allowed to copy this > message or attachment or disclose the contents to any other person and > have to destroy all copies of the original message._ > > On 03 Oct 2014, at 15:25, F?bio Rabelo > wrote: > > How can I check how many HBA controlres do I have connected in a > system, if this system are in a ( very ) remote location, and I just > have SSH available ? > > prtconf -v returns too many noise ! > > prtconv -v |grep RAID returns only one line ... > > There is an easy way ? > > F?bio Rabelo _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss [5] _______________________________________________ OmniOS-discuss mailing list OmniOS-discuss at lists.omniti.com http://lists.omniti.com/mailman/listinfo/omnios-discuss [5] Links: ------ [1] http://www.innovationspreis-it.de/ [2] http://www.losstech-blog.de [3] tel:%2B49%20%280%294123%20929900 [4] tel:%2B49%20%280%294123%20929911 [5] http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 10803 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.jpg Type: image/jpeg Size: 3874 bytes Desc: not available URL: From dan at syneto.net Fri Oct 3 14:30:04 2014 From: dan at syneto.net (Dan Vatca) Date: Fri, 3 Oct 2014 17:30:04 +0300 Subject: [OmniOS-discuss] how many HBAs my system have ? In-Reply-To: References: Message-ID: I think sasinfo might not be installed by default. Try 'sudo pkg install sasinfo'. It is part of the illumos kernel, so it must be present in omnios. What version are you running? On Fri, Oct 3, 2014 at 4:25 PM, F?bio Rabelo wrote: > How can I check how many HBA controlres do I have connected in a > system, if this system are in a ( very ) remote location, and I just > have SSH available ? > > prtconf -v returns too many noise ! > > prtconv -v |grep RAID returns only one line ... > > There is an easy way ? > > > F?bio Rabelo > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Fri Oct 3 15:22:24 2014 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 3 Oct 2014 11:22:24 -0400 Subject: [OmniOS-discuss] feature request In-Reply-To: <18461750.20141003152926@tierarzt-mueller.de> References: <18461750.20141003152926@tierarzt-mueller.de> Message-ID: On Oct 3, 2014, at 9:29 AM, Alexander Lesle wrote: > Hello Dan McDonald and List, > > in my mind this was a great proposal what you done at Sep, 02. > That's the way to build a great OmniOS near to the users. > > Here my request for the next release: > It would be convenient if open-vm-tools were installed in the new > release. http://sourceforge.net/projects/open-vm-tools/files/ Actually, this work has been done. I dropped it on the ground: https://github.com/omniti-labs/omnios-build/pull/39 I will be merging this pull requests now in the master branch, and seeing how OOod chews on it. I know there's other work to be done for OmniOS-as-a-guest as well. Thanks, DAn From fabio at fabiorabelo.wiki.br Fri Oct 3 16:54:32 2014 From: fabio at fabiorabelo.wiki.br (=?UTF-8?Q?F=C3=A1bio_Rabelo?=) Date: Fri, 3 Oct 2014 13:54:32 -0300 Subject: [OmniOS-discuss] how many HBAs my system have ? In-Reply-To: References: Message-ID: pkg search sasinfo returns nothing ... pkg install sasinfo returns "no matching version of the system ... can be installed" Same with pfexec pkg install sasinfo cfgadm shows the boot disk, and the ZIL disk, but nothing in the HBAs ... But I know there are 16 hard disks connected to this system . F?bio Rabelo 2014-10-03 11:30 GMT-03:00 Dan Vatca : > I think sasinfo might not be installed by default. Try 'sudo pkg install > sasinfo'. It is part of the illumos kernel, so it must be present in omnios. > What version are you running? > > > > On Fri, Oct 3, 2014 at 4:25 PM, F?bio Rabelo > wrote: >> >> How can I check how many HBA controlres do I have connected in a >> system, if this system are in a ( very ) remote location, and I just >> have SSH available ? >> >> prtconf -v returns too many noise ! >> >> prtconv -v |grep RAID returns only one line ... >> >> There is an easy way ? >> >> >> F?bio Rabelo >> _______________________________________________ >> OmniOS-discuss mailing list >> OmniOS-discuss at lists.omniti.com >> http://lists.omniti.com/mailman/listinfo/omnios-discuss > > From richard.elling at richardelling.com Fri Oct 3 17:26:58 2014 From: richard.elling at richardelling.com (Richard Elling) Date: Fri, 3 Oct 2014 10:26:58 -0700 Subject: [OmniOS-discuss] how many HBAs my system have ? In-Reply-To: References: Message-ID: On Oct 3, 2014, at 6:25 AM, F?bio Rabelo wrote: > How can I check how many HBA controlres do I have connected in a > system, if this system are in a ( very ) remote location, and I just > have SSH available ? > > prtconf -v returns too many noise ! > > prtconv -v |grep RAID returns only one line ... > > There is an easy way ? As previously mentioned, sasinfo shows the number of SAS HBAs and their connectivity to enclosures and ultimately disks. Most systems also have one or more SATA HBAs, which use a different driver and needs a different tool. If you have a RAID HBA, then it could be using the mr_sas driver, with a different tool. "prtconf -D" shows the device tree with driver attachments, which is sometimes reliably scriptable. The usual driver names are mpt_sas, mr_sas, mpt, ahci, pci-ide -- richard From mir at miras.org Fri Oct 3 17:46:32 2014 From: mir at miras.org (Michael Rasmussen) Date: Fri, 3 Oct 2014 19:46:32 +0200 Subject: [OmniOS-discuss] how many HBAs my system have ? In-Reply-To: References: Message-ID: <20141003194632.7b3c1d2d@sleipner.datanom.net> On Fri, 3 Oct 2014 13:54:32 -0300 F?bio Rabelo wrote: > pkg search sasinfo returns nothing ... > > pkg install sasinfo returns "no matching version of the system ... > can be installed" > > Same with pfexec pkg install sasinfo > > cfgadm shows the boot disk, and the ZIL disk, but nothing in the HBAs ... > > But I know there are 16 hard disks connected to this system . > > > F?bio Rabelo > > 2014-10-03 11:30 GMT-03:00 Dan Vatca : > > I think sasinfo might not be installed by default. Try 'sudo pkg install > > sasinfo'. It is part of the illumos kernel, so it must be present in omnios. > > What version are you running? > > > > > > > > On Fri, Oct 3, 2014 at 4:25 PM, F?bio Rabelo > > wrote: > >> > >> How can I check how many HBA controlres do I have connected in a > >> system, if this system are in a ( very ) remote location, and I just > >> have SSH available ? > >> > >> prtconf -v returns too many noise ! > >> > >> prtconv -v |grep RAID returns only one line ... > >> > >> There is an easy way ? > >> Is your HBA LSI based? Then checkout lsiutils from the omni repo. -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: Never promise more than you can perform. -- Publilius Syrus -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From mbi at losstech.de Fri Oct 3 22:20:22 2014 From: mbi at losstech.de (Bielicki, Michal) Date: Fri, 3 Oct 2014 22:20:22 +0000 Subject: [OmniOS-discuss] how many HBAs my system have ? In-Reply-To: References: Message-ID: <40B8675C-B518-4968-BE74-0DD7D1BE69C0@losstech.de> Fresh ominous stable install: root at omnios-omnios:/root# pkg refresh root at omnios-omnios:/root# pkg search sasinfo INDEX ACTION VALUE PACKAGE basename file usr/sbin/sasinfo pkg:/system/storage/sasinfo at 0.5.11-0.151012 pkg.fmri set omnios/system/storage/sasinfo pkg:/system/storage/sasinfo at 0.5.11-0.151012 root at omnios-omnios:/root# pkg publisher PUBLISHER TYPE STATUS URI omnios origin online http://pkg.omniti.com/omnios/r151012/ root at omnios-omnios:/root# not really sure what you have installed there that it can?t find a corresponding sassing pkg ? Kind regards / Gr??e aus dem hohen Norden Michal Bielicki [cid:1C01820A-2F8C-422C-9089-72CFBE601F8E at fritz.box][cid:9DCD0525-2332-419B-AA48-F175C80366D4 at fritz.box] Aktuelle Publikationen: www.losstech-blog.de losstech GmbH Hauptstra?e 24, DE-25355 Lutzhorn mailto:mbi at losstech.de +49 (0)4123 929900 (fon) +49 (0)4123 929911 (fax) Amtsgericht Pinneberg, HRB 5737 USt-Id: DE 227 8636 32 Hamburger Sparkasse, KTO 1217 149 150, BLZ 200 505 50 Sparkasse S?dholstein, KTO 151 228 56, BLZ 230 510 30 This message and any attachment is losstech proprietary and may be privileged or otherwise protected from disclosure. If you are not the intended recipient, please phone or email the sender and delete this message and any attachment from your system. If you are not the intended recipient you are not allowed to copy this message or attachment or disclose the contents to any other person and have to destroy all copies of the original message. On 3 pa? 2014, at 18:54, F?bio Rabelo > wrote: pkg search sasinfo returns nothing ... pkg install sasinfo returns "no matching version of the system ... can be installed" Same with pfexec pkg install sasinfo cfgadm shows the boot disk, and the ZIL disk, but nothing in the HBAs ... But I know there are 16 hard disks connected to this system . F?bio Rabelo 2014-10-03 11:30 GMT-03:00 Dan Vatca >: I think sasinfo might not be installed by default. Try 'sudo pkg install sasinfo'. It is part of the illumos kernel, so it must be present in omnios. What version are you running? On Fri, Oct 3, 2014 at 4:25 PM, F?bio Rabelo > wrote: How can I check how many HBA controlres do I have connected in a system, if this system are in a ( very ) remote location, and I just have SSH available ? prtconf -v returns too many noise ! prtconv -v |grep RAID returns only one line ... There is an easy way ? F?bio Rabelo _______________________________________________ OmniOS-discuss mailing list OmniOS-discuss at lists.omniti.com http://lists.omniti.com/mailman/listinfo/omnios-discuss _______________________________________________ OmniOS-discuss mailing list OmniOS-discuss at lists.omniti.com http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 10803 bytes Desc: image001.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.jpg Type: image/jpeg Size: 3874 bytes Desc: image002.jpg URL: From admdwrf at gmail.com Sat Oct 4 09:59:11 2014 From: admdwrf at gmail.com (=?UTF-8?B?RnLDqWTDqXJpYyBBbGl4?=) Date: Sat, 4 Oct 2014 11:59:11 +0200 Subject: [OmniOS-discuss] Omnios under Google Compute Engine: virtio scsi Message-ID: I would like install OmniOS under GCE instances. But problem, GCE use only virtio scsi and not blk ;( I found this link: https://github.com/trisk/illumos-gate/tree/vioscsi-fk1/usr/src/uts/common/io/vioscsi/ Someone already try to use it under OmniOS ? Good weekend, Fred -------------- next part -------------- An HTML attachment was scrubbed... URL: From gearboxes at outlook.com Sun Oct 5 02:26:38 2014 From: gearboxes at outlook.com (Machine Man) Date: Sat, 4 Oct 2014 22:26:38 -0400 Subject: [OmniOS-discuss] KVM disk performance r151012 In-Reply-To: <546FCD5D-A3F9-438D-AA18-22A2F0FB8BC2@oetiker.ch> References: , <546FCD5D-A3F9-438D-AA18-22A2F0FB8BC2@oetiker.ch> Message-ID: It is not disk. There is hardly any IO. I did a number of tests and I hardly see 10MB/s. Appears to be network related. Not sure if anyone else is experiencing this. For now I will stay on the previous release due to lack of time to troubleshoot. Subject: Re: [OmniOS-discuss] KVM disk performance r151012 From: tobi at oetiker.ch Date: Fri, 3 Oct 2014 07:02:05 +0200 CC: omnios-discuss at lists.omniti.com To: gearboxes at outlook.com On 03.10.2014, at 06:51, Machine Man wrote: After upgrading disk performance dropped from avg. 60MB/s to 17MB/s when copying over the network to a windows vm.Booting back to pre-upgrade and performance is restored Would this be resolved if the pool is upgraded? are you sure it is the disk and not the network? cheers tobi _______________________________________________ OmniOS-discuss mailing list OmniOS-discuss at lists.omniti.com http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at marzocchi.net Sun Oct 5 11:56:24 2014 From: lists at marzocchi.net (Olaf Marzocchi) Date: Sun, 5 Oct 2014 13:56:24 +0200 Subject: [OmniOS-discuss] Building CalendarServer fails for GSSAPI Message-ID: <1E904E11-62D9-40CF-B283-485A0946B7FE@marzocchi.net> Dear all, I submitted the same request some weeks ago to the CalendarServer but I got no answer, so I try with OmniOS, where I also expect to find the knowledge necessary to solve this issue. I am preparing a server based on OmniOS and I would like to offer CalDAV/CardDAV to the users. I chose calendarserver because the users use iOS, OS X, Android. I know there are other options like DAViCal, but for the time being I tried to go with Apple?s implementation. I checked the repository locally and then I run ?python setup.py?. Everything gets compiled well after putting the GNU tar in the path before the one provided by OmniOS, except PyKerberos. It is a library written to avoid including the whole Python Kerberos libraries (or so I read in the READMEs). This is the output: ------ ~/CalendarServer/CalendarServer-5.2 $ ./run -s Using built libevent. Using built memcached. Using built PostgreSQL. Using built OpenLDAP. Using built libffi. Using system version of setuptools. Building Zope Interface... Using system version of pyOpenSSL. Building PyKerberos... gcc: error: /usr/bin/krb5-config:: No such file or directory gcc: error: Unknown: No such file or directory gcc: error: option: No such file or directory gcc: error: `gssapi': No such file or directory gcc: error: use: No such file or directory gcc: error: `--help': No such file or directory gcc: error: for: No such file or directory gcc: error: usage: No such file or directory gcc: error: unrecognized command line option '--' error: command 'gcc' failed with exit status 1 ~/CalendarServer/CalendarServer-5.2 $ cd ../PyKerberos ~/CalendarServer/PyKerberos $ python setup.py build running build running build_ext building 'kerberos' extension gcc -m64 -fno-strict-aliasing -std=c99 -m64 -DNDEBUG -g -O3 -Wall -Wstrict-prototypes -fPIC -I/usr/include/python2.6 -c src/kerberos.c -o build/temp.solaris-2.11-i86pc-2.6/src/kerberos.o /usr/bin/krb5-config: Unknown option `gssapi' -- use `--help' for usage gcc: error: /usr/bin/krb5-config:: No such file or directory gcc: error: Unknown: No such file or directory gcc: error: option: No such file or directory gcc: error: `gssapi': No such file or directory gcc: error: use: No such file or directory gcc: error: `--help': No such file or directory gcc: error: for: No such file or directory gcc: error: usage: No such file or directory gcc: error: unrecognized command line option '--' error: command 'gcc' failed with exit status 1 ~/src/CalendarServer/PyKerberos $ krb5-config --all --libs --cflags Version: Solaris Kerberos (based on MIT Kerberos 5 release 1.6.3) Vendor: Sun Microsystems, Inc. Prefix: /usr Exec_prefix: /usr -I/usr/include/kerberosv5 -L/usr/lib -R/usr/lib -lkrb5 ------ It is also important to mention that the build script uses the commandline "krb5-config --libs gssapi" to get the path for the GSSAPI libs. However the gssapi packages ARE installed (I never tested Kerberos though): ------ ~/src/CalendarServer/PyKerberos $ pkg search gssapi INDEX ACTION VALUE PACKAGE pkg.summary set GSSAPI CONFIG V2 pkg:/service/security/gss at 0.5.11-0.151010 pkg.summary set GSSAPI V2 pkg:/system/library/security/gss at 0.5.11-0.151010 pkg.summary set kernel GSSAPI V2 pkg:/system/kernel/security/gss at 0.5.11-0.151010 basename dir usr/include/gssapi pkg:/system/header at 0.5.11-0.151010 ~/src/CalendarServer/PyKerberos $ pkg info pkg:/service/security/gss at 0.5.11-0.151010 pkg:/system/library/security/gss at 0.5.11-0.151010 pkg:/system/kernel/security/gss at 0.5.11-0.151010 pkg:/system/header at 0.5.11-0.151010 Name: service/security/gss Summary: GSSAPI CONFIG V2 Description: Generic Security Service Application Program Interface, Version 2 - config Category: System/Security State: Installed Publisher: omnios Version: 0.5.11 Build Release: 5.11 Branch: 0.151010 Packaging Date: Mon Apr 28 19:28:56 2014 Size: 17.53 kB FMRI: pkg://omnios/service/security/gss at 0.5.11,5.11-0.151010:20140428T192856Z Name: system/header Summary: SunOS Header Files Description: SunOS C/C++ header files for general development of software Category: System/Core State: Installed Publisher: omnios Version: 0.5.11 Build Release: 5.11 Branch: 0.151010 Packaging Date: Mon Apr 28 19:29:16 2014 Size: 12.19 MB FMRI: pkg://omnios/system/header at 0.5.11,5.11-0.151010:20140428T192916Z Name: system/kernel/security/gss Summary: kernel GSSAPI V2 Description: Generic Security Service Application Program Interface, Version 2 - kernel Category: System/Security State: Installed Publisher: omnios Version: 0.5.11 Build Release: 5.11 Branch: 0.151010 Packaging Date: Mon Apr 28 19:29:21 2014 Size: 335.22 kB FMRI: pkg://omnios/system/kernel/security/gss at 0.5.11,5.11-0.151010:20140428T192921Z Name: system/library/security/gss Summary: GSSAPI V2 Description: Generic Security Service Application Program Interface, Version 2 - user Category: System/Security State: Installed Publisher: omnios Version: 0.5.11 Build Release: 5.11 Branch: 0.151010 Packaging Date: Mon Apr 28 19:29:26 2014 Size: 687.41 kB FMRI: pkg://omnios/system/library/security/gss at 0.5.11,5.11-0.151010:20140428T192926Z ------ Could anyone help me with the issue? I tried some changes like providing directly the path with the libraries but I was not able to solve the problem. Thanks Olaf Marzocchi -------------- next part -------------- An HTML attachment was scrubbed... URL: From mir at miras.org Sun Oct 5 12:19:34 2014 From: mir at miras.org (Michael Rasmussen) Date: Sun, 5 Oct 2014 14:19:34 +0200 Subject: [OmniOS-discuss] Building CalendarServer fails for GSSAPI In-Reply-To: <1E904E11-62D9-40CF-B283-485A0946B7FE@marzocchi.net> References: <1E904E11-62D9-40CF-B283-485A0946B7FE@marzocchi.net> Message-ID: <20141005141934.5d99857a@sleipner.datanom.net> On Sun, 5 Oct 2014 13:56:24 +0200 Olaf Marzocchi wrote: > I am preparing a server based on OmniOS and I would like to offer CalDAV/CardDAV to the users. I chose calendarserver because the users use iOS, OS X, Android. I know there are other options like DAViCal, but for the time being I tried to go with Apple?s implementation. > If you fetch davical from git it works with iOS <= 7. I have no idea about iOS 8. OSX and Android works out of the box. If you do try davical you should search for caldavzap and cardavzap which is webapps developed primary against davical. > > gcc: error: /usr/bin/krb5-config:: No such file or directory > gcc: error: Unknown: No such file or directory > gcc: error: option: No such file or directory > gcc: error: `gssapi': No such file or directory > gcc: error: use: No such file or directory > gcc: error: `--help': No such file or directory > gcc: error: for: No such file or directory > gcc: error: usage: No such file or directory > gcc: error: unrecognized command line option '--' > error: command 'gcc' failed with exit status 1 > You are missing kerberos libraries $ pkg search kerberos INDEX ACTION VALUE PACKAGE pkg.description set Kerberos V5 Master KDC (user) pkg:/system/security/kerberos-5 at 0.5.11-0.151012 pkg.description set Kerberos version 5 support pkg:/service/security/kerberos-5 at 0.5.11-0.151012 pkg.summary set Kerberos V5 Master KDC (user) pkg:/system/security/kerberos-5 at 0.5.11-0.151012 pkg.summary set Kerberos version 5 support pkg:/service/security/kerberos-5 at 0.5.11-0.151012 Whether these are sufficient I don't know. -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: You are a bundle of energy, always on the go. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From fabio at fabiorabelo.wiki.br Mon Oct 6 11:46:02 2014 From: fabio at fabiorabelo.wiki.br (=?UTF-8?Q?F=C3=A1bio_Rabelo?=) Date: Mon, 6 Oct 2014 08:46:02 -0300 Subject: [OmniOS-discuss] replace disk with difrferent model/size Message-ID: Hi to all I have a system with failed hard disk . When I try to replace it, after 3 to 5 seconds "thinking" the system returns a msg like geometry mismatch error So, how can I replace a failed disk with another model or size ? This system uses 2 TB Hard disks, at first, I've installed 4 TB one, after this geometry error I get a try with 2 TB model, same result . F?bio Rabelo From chip at innovates.com Mon Oct 6 14:41:29 2014 From: chip at innovates.com (Schweiss, Chip) Date: Mon, 6 Oct 2014 09:41:29 -0500 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start Message-ID: I've started the process of moving my systems to r151012. The first production system I did this on nlockmgr repeatedly went to maintenance mode. Several attempts to clear it failed. I've restarted on the previous boot environment and it starts fine on r151010. I think this is related to https://www.illumos.org/issues/4518, but I'm not sure if this is a new bug. This system has only 3 zfs folders, all NFS shared. Unfortunately this being a production system, I cannot dedicate much downtime to diagnosing this problem. I mounted the r151012 be, to a temporary mount point so I could examine logs more closely. Unfortunately, not much detail there: root at hcp-iops1:/tmp/omnios-r151012/var/svc/log# cat network-nfs-nlockmgr\:default.log .... ...old lines deleted .... [ Oct 6 07:05:26 Disabled. ] [ Oct 6 07:05:48 Enabled. ] [ Oct 6 07:05:48 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:06:17 Method "start" exited with status 1. ] [ Oct 6 07:06:17 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:06:18 Method "start" exited with status 0. ] [ Oct 6 07:14:06 Disabled. ] [ Oct 6 07:14:35 Enabled. ] [ Oct 6 07:14:35 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:15:00 Method "start" exited with status 1. ] [ Oct 6 07:15:00 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:15:25 Method "start" exited with status 1. ] [ Oct 6 07:15:25 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:15:50 Method "start" exited with status 1. ] [ Oct 6 07:18:11 Leaving maintenance because clear requested. ] [ Oct 6 07:18:11 Enabled. ] [ Oct 6 07:18:11 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:18:36 Method "start" exited with status 1. ] [ Oct 6 07:18:45 Leaving maintenance because clear requested. ] [ Oct 6 07:18:45 Enabled. ] [ Oct 6 07:18:45 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:19:10 Method "start" exited with status 1. ] [ Oct 6 07:19:17 Leaving maintenance because clear requested. ] [ Oct 6 07:19:17 Enabled. ] [ Oct 6 07:19:17 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:19:42 Method "start" exited with status 1. ] [ Oct 6 07:20:09 Leaving maintenance because clear requested. ] [ Oct 6 07:20:09 Enabled. ] [ Oct 6 07:20:09 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:20:34 Method "start" exited with status 1. ] [ Oct 6 07:21:23 Disabled. ] [ Oct 6 07:22:36 Enabled. ] [ Oct 6 07:22:36 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:23:06 Method "start" exited with status 1. ] [ Oct 6 07:23:06 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:23:31 Method "start" exited with status 1. ] [ Oct 6 07:23:31 Executing start method ("/lib/svc/method/nlockmgr"). ] [ Oct 6 07:23:36 Method "start" exited with status 0. ] Anyone else seeing this in r151012? Any tips on collecting better information on this would be appreciated. -Chip -------------- next part -------------- An HTML attachment was scrubbed... URL: From chip at innovates.com Mon Oct 6 14:48:20 2014 From: chip at innovates.com (Schweiss, Chip) Date: Mon, 6 Oct 2014 09:48:20 -0500 Subject: [OmniOS-discuss] replace disk with difrferent model/size In-Reply-To: References: Message-ID: Looks like you are trying to add a 4K disk to pool with ashift=9. If the disk is a 512e disk you can force it to ashift=9 by adding an entry in sd.conf. This will degrade write performance. It's best to move the whole pool to ashift=12 (4K sectors) or find a 512b native disk. -Chip On Mon, Oct 6, 2014 at 6:46 AM, F?bio Rabelo wrote: > Hi to all > > I have a system with failed hard disk . > > When I try to replace it, after 3 to 5 seconds "thinking" the system > returns a msg like > > geometry mismatch error > > So, how can I replace a failed disk with another model or size ? > > This system uses 2 TB Hard disks, at first, I've installed 4 TB one, > after this geometry error I get a try with 2 TB model, same result . > > > F?bio Rabelo > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Mon Oct 6 14:59:24 2014 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 6 Oct 2014 10:59:24 -0400 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: References: Message-ID: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> On Oct 6, 2014, at 10:41 AM, Schweiss, Chip wrote: > > Anyone else seeing this in r151012? > > Any tips on collecting better information on this would be appreciated. I saw this in once in 012, but not as persistently as you have. It cleared up for me with a single reboot, but that may have just been because I got lucky w.r.t. statd. I take it "svcadm disable nlockmgr ; svcadm enable nlockmgr" doesn't help? And as for output, I've seen in /var/adm/messages what you have, per the cited illumos bug (4518). Dan From chip at innovates.com Mon Oct 6 15:56:08 2014 From: chip at innovates.com (Schweiss, Chip) Date: Mon, 6 Oct 2014 10:56:08 -0500 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> References: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> Message-ID: On Mon, Oct 6, 2014 at 9:59 AM, Dan McDonald wrote: > > On Oct 6, 2014, at 10:41 AM, Schweiss, Chip wrote: > > > > > Anyone else seeing this in r151012? > > > > Any tips on collecting better information on this would be appreciated. > > I saw this in once in 012, but not as persistently as you have. It > cleared up for me with a single reboot, but that may have just been because > I got lucky w.r.t. statd. > > I take it "svcadm disable nlockmgr ; svcadm enable nlockmgr" doesn't help? > I didn't think to try that. Tried rebooting, but that didn't help. I was already past my maintenance window, so I resorted to backing out to r151010. > > And as for output, I've seen in /var/adm/messages what you have, per the > cited illumos bug (4518). > I wasn't aware of any new NFS changes that may have caused this to become exaggerated in r151012. But, it's not clear on what is triggering this bug either. On my larger pool systems with lots of NFS exports, almost every reboot causes this and a 'svcadm clear' fixes it. Is there something different that a disable and enable does? I can likely get a small window in the next few days to give it a try. -Chip -------------- next part -------------- An HTML attachment was scrubbed... URL: From fabio at fabiorabelo.wiki.br Mon Oct 6 18:56:21 2014 From: fabio at fabiorabelo.wiki.br (=?UTF-8?Q?F=C3=A1bio_Rabelo?=) Date: Mon, 6 Oct 2014 15:56:21 -0300 Subject: [OmniOS-discuss] replace disk with difrferent model/size In-Reply-To: References: Message-ID: Thanks a lot ! This user intent to replace the entire pool, so it is the best to change it . But I have no idea what I have to do so ... Point me to a doc with this info and I will read and apply . Many thanks again .... F?bio Rabelo 2014-10-06 11:48 GMT-03:00 Schweiss, Chip : > Looks like you are trying to add a 4K disk to pool with ashift=9. > > If the disk is a 512e disk you can force it to ashift=9 by adding an entry > in sd.conf. This will degrade write performance. It's best to move the > whole pool to ashift=12 (4K sectors) or find a 512b native disk. > > -Chip > > On Mon, Oct 6, 2014 at 6:46 AM, F?bio Rabelo > wrote: > >> Hi to all >> >> I have a system with failed hard disk . >> >> When I try to replace it, after 3 to 5 seconds "thinking" the system >> returns a msg like >> >> geometry mismatch error >> >> So, how can I replace a failed disk with another model or size ? >> >> This system uses 2 TB Hard disks, at first, I've installed 4 TB one, >> after this geometry error I get a try with 2 TB model, same result . >> >> >> F?bio Rabelo >> _______________________________________________ >> OmniOS-discuss mailing list >> OmniOS-discuss at lists.omniti.com >> http://lists.omniti.com/mailman/listinfo/omnios-discuss >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Kevin.Swab at ColoState.EDU Mon Oct 6 18:58:58 2014 From: Kevin.Swab at ColoState.EDU (Kevin Swab) Date: Mon, 06 Oct 2014 12:58:58 -0600 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: References: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> Message-ID: <5432E672.2080805@ColoState.EDU> I had this same problem after upgrading a system to r151012. nlockmgr failed to start because "svc:/network/nfs/status:default" was disabled. I enabled that service, then nlockmgr was happy. I've done 3 other upgrades to r151012 and none of them had a problem with nlockmgr... Kevin On 10/06/2014 09:56 AM, Schweiss, Chip wrote: > > > On Mon, Oct 6, 2014 at 9:59 AM, Dan McDonald > wrote: > > > On Oct 6, 2014, at 10:41 AM, Schweiss, Chip > wrote: > > > > > Anyone else seeing this in r151012? > > > > Any tips on collecting better information on this would be appreciated. > > I saw this in once in 012, but not as persistently as you have. It > cleared up for me with a single reboot, but that may have just been > because I got lucky w.r.t. statd. > > I take it "svcadm disable nlockmgr ; svcadm enable nlockmgr" doesn't > help? > > > I didn't think to try that. Tried rebooting, but that didn't help. I > was already past my maintenance window, so I resorted to backing out to > r151010. > > > > > And as for output, I've seen in /var/adm/messages what you have, per > the cited illumos bug (4518). > > > > I wasn't aware of any new NFS changes that may have caused this to > become exaggerated in r151012. But, it's not clear on what is > triggering this bug either. On my larger pool systems with lots of NFS > exports, almost every reboot causes this and a 'svcadm clear' fixes it. > > Is there something different that a disable and enable does? I can > likely get a small window in the next few days to give it a try. > > -Chip > > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -- ------------------------------------------------------------------- Kevin Swab UNIX Systems Administrator ACNS Colorado State University Phone: (970)491-6572 Email: Kevin.Swab at ColoState.EDU GPG Fingerprint: 7026 3F66 A970 67BD 6F17 8EB8 8A7D 142F 2392 791C From danmcd at omniti.com Tue Oct 7 19:40:52 2014 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 7 Oct 2014 15:40:52 -0400 Subject: [OmniOS-discuss] Anyone using Sun DHCP *server* still? Message-ID: <4DDB817D-8148-4A42-91FE-44C9F4C5BE20@omniti.com> Note: This is the server, not the client. I ask because: 1.) I've put in ISC DHCP server support into r151012 in anticipation of... 2.)... the impending removal of the Sun DHCP server upstream. This removal is now under RTI review (means it's coming soon unless we find something horribly broken about it). The current bloody cycle is leading up to r151014, which is not only the next stable, but is ALSO the next long-term support release. At this rate, it's likely to NOT have Sun DHCP. If you have any objections to the removal of the Sun DHCP server, speak now on this list. It's coming anyway, but if you're unduly affected by it, speaking now will allow folks to help out in any migration. FYI, Dan From rt at steait.net Tue Oct 7 23:00:47 2014 From: rt at steait.net (Rune Tipsmark) Date: Tue, 7 Oct 2014 23:00:47 +0000 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files Message-ID: <3dde7c3ae1994829b588d3df09855cf0@EX1301.steait.net> hi guys, wondering if someone might know why my pool is still allocated 1.45T after I removed the files on the LU's provisioned onto that pool. pool: pool02 state: ONLINE scan: scrub in progress since Wed Oct 8 02:31:41 2014 29.1G scanned out of 1.45T at 52.4M/s, 7h54m to go 0 repaired, 1.96% done config: NAME STATE READ WRITE CKSUM pool02 ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c4t5000C500570858EFd0 ONLINE 0 0 0 c4t5000C50057085A6Bd0 ONLINE 0 0 0 mirror-1 ONLINE 0 0 0 c4t5000C500570870D3d0 ONLINE 0 0 0 c4t5000C50057089753d0 ONLINE 0 0 0 logs c11d0 ONLINE 0 0 0 cache c13d0 ONLINE 0 0 0 c15d0 ONLINE 0 0 0 Supporting 3 LU's mounted in Vmware, I formatted the LUNs again and no change... Do I really need to delete the pool, then re-create it? br, Rune -------------- next part -------------- An HTML attachment was scrubbed... URL: From sjorge+ml at blackdot.be Tue Oct 7 23:04:00 2014 From: sjorge+ml at blackdot.be (Jorge Schrauwen) Date: Wed, 08 Oct 2014 01:04:00 +0200 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <3dde7c3ae1994829b588d3df09855cf0@EX1301.steait.net> References: <3dde7c3ae1994829b588d3df09855cf0@EX1301.steait.net> Message-ID: <380eec4d6c3d75dc5fb55d43945e6221@blackdot.be> A few things come to mind, e.g. snapshots, reservations,... On 2014-10-08 01:00, Rune Tipsmark wrote: > hi guys, wondering if someone might know why my pool is still allocated 1.45T after I removed the files on the LU's provisioned onto that pool. > > pool: pool02 > state: ONLINE > scan: scrub in progress since Wed Oct 8 02:31:41 2014 > 29.1G scanned out of 1.45T at 52.4M/s, 7h54m to go > 0 repaired, 1.96% done > config: > > NAME STATE READ WRITE CKSUM > pool02 ONLINE 0 0 0 > mirror-0 ONLINE 0 0 0 > c4t5000C500570858EFd0 ONLINE 0 0 0 > c4t5000C50057085A6Bd0 ONLINE 0 0 0 > mirror-1 ONLINE 0 0 0 > c4t5000C500570870D3d0 ONLINE 0 0 0 > c4t5000C50057089753d0 ONLINE 0 0 0 > logs > c11d0 ONLINE 0 0 0 > cache > c13d0 ONLINE 0 0 0 > c15d0 ONLINE 0 0 0 > > Supporting 3 LU's mounted in Vmware, I formatted the LUNs again and no change... > > Do I really need to delete the pool, then re-create it? > > br, > Rune > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss [1] Links: ------ [1] http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From minikola at gmail.com Wed Oct 8 14:26:12 2014 From: minikola at gmail.com (Nikola M.) Date: Wed, 08 Oct 2014 16:26:12 +0200 Subject: [OmniOS-discuss] Anyone using Sun DHCP *server* still? In-Reply-To: <4DDB817D-8148-4A42-91FE-44C9F4C5BE20@omniti.com> References: <4DDB817D-8148-4A42-91FE-44C9F4C5BE20@omniti.com> Message-ID: <54354984.8060105@gmail.com> On 10/ 7/14 09:40 PM, Dan McDonald wrote: > Note: This is the server, not the client. > > I ask because: > > 1.) I've put in ISC DHCP server support into r151012 in anticipation of... > > 2.)... the impending removal of the Sun DHCP server upstream. This removal is now under RTI review (means it's coming soon unless we find something horribly broken about it). > > The current bloody cycle is leading up to r151014, which is not only the next stable, but is ALSO the next long-term support release. At this rate, it's likely to NOT have Sun DHCP. > > If you have any objections to the removal of the Sun DHCP server, speak now on this list. It's coming anyway, but if you're unduly affected by it, speaking now will allow folks to help out in any migration. > Hi, what are procedures in OmniOS, Btw, when some packages are removed, replaced etc? Do you issue recommendations for migrating configurations, alternative packages to use and how to make them work and does distribution documentation change to reflect the change. (past certain release number)? Thanks. BTW From danmcd at omniti.com Wed Oct 8 14:38:55 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 8 Oct 2014 10:38:55 -0400 Subject: [OmniOS-discuss] Anyone using Sun DHCP *server* still? In-Reply-To: <54354984.8060105@gmail.com> References: <4DDB817D-8148-4A42-91FE-44C9F4C5BE20@omniti.com> <54354984.8060105@gmail.com> Message-ID: <16712547-F363-4210-87FE-9A67F148D94D@omniti.com> On Oct 8, 2014, at 10:26 AM, Nikola M. wrote: > Hi, what are procedures in OmniOS, Btw, when some packages are removed, replaced etc? We're in brand-new territory here, so to be honest, we don't have any procedures as of yet. This bloody cycle (r151013) gives us a chance to work it out. It's why I'm *asking*. > Do you issue recommendations for migrating configurations, alternative packages to use and how to make them work and does distribution documentation change to reflect the change. (past certain release number)? The illumos community has been under the impression that very few, if any, people use the Sun DHCP. I'm asking here to confirm/deny that hunch. Are you migrating from Sun DHCP to ISC DHCP? Dan From filip.marvan at aira.cz Wed Oct 8 14:53:23 2014 From: filip.marvan at aira.cz (Filip Marvan) Date: Wed, 8 Oct 2014 16:53:23 +0200 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files Message-ID: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2B7D@AIRA-SRV.aira.local> Hello, maybe you will have to enable compression on that ZVOL, and rewrite that disk with zeroes from client. Because your pool propably don't even know from vmware client, that you delete your data. Filip -------------------------------------------------- Date: Tue, 7 Oct 2014 23:00:47 +0000 From: Rune Tipsmark To: omnios-discuss Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files Message-ID: <3dde7c3ae1994829b588d3df09855cf0 at EX1301.steait.net> Content-Type: text/plain; charset="us-ascii" hi guys, wondering if someone might know why my pool is still allocated 1.45T after I removed the files on the LU's provisioned onto that pool. pool: pool02 state: ONLINE scan: scrub in progress since Wed Oct 8 02:31:41 2014 29.1G scanned out of 1.45T at 52.4M/s, 7h54m to go 0 repaired, 1.96% done config: NAME STATE READ WRITE CKSUM pool02 ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c4t5000C500570858EFd0 ONLINE 0 0 0 c4t5000C50057085A6Bd0 ONLINE 0 0 0 mirror-1 ONLINE 0 0 0 c4t5000C500570870D3d0 ONLINE 0 0 0 c4t5000C50057089753d0 ONLINE 0 0 0 logs c11d0 ONLINE 0 0 0 cache c13d0 ONLINE 0 0 0 c15d0 ONLINE 0 0 0 Supporting 3 LU's mounted in Vmware, I formatted the LUNs again and no change... Do I really need to delete the pool, then re-create it? br, Rune -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6247 bytes Desc: not available URL: From alexandre.lecuyer at gmail.com Wed Oct 8 16:08:44 2014 From: alexandre.lecuyer at gmail.com (Alexandre Lecuyer) Date: Wed, 8 Oct 2014 18:08:44 +0200 Subject: [OmniOS-discuss] itadm delete-tpg hang Message-ID: I have seen this happen twice : "/usr/sbin/itadm delete-tpg " hangs. The stack : [ ffffff002d2c9930 _resume_from_idle+0xf4() ] ffffff002d2c9960 swtch+0x141() ffffff002d2c99a0 cv_wait+0x70(ffffff1724512ba8, ffffff1724512ba0) ffffff002d2c99e0 idm_refcnt_wait_ref+0x7b(ffffff1724512b88) ffffff002d2c9a20 iscsit_config_destroy_tpgs+0x4d(ffffff002d2c9a90) ffffff002d2c9ae0 iscsit_config_merge+0xfb(ffffff486c690c70) ffffff002d2c9c80 iscsit_drv_ioctl+0x3d6(10800000000, 1, 8047cf8, 100403, ffffff295594dc48, ffffff002d2c9e68) ffffff002d2c9cc0 cdev_ioctl+0x39(10800000000, 1, 8047cf8, 100403, ffffff295594dc48, ffffff002d2c9e68) ffffff002d2c9d10 spec_ioctl+0x60(ffffff2847bbca00, 1, 8047cf8, 100403, ffffff295594dc48, ffffff002d2c9e68, 0) ffffff002d2c9da0 fop_ioctl+0x55(ffffff2847bbca00, 1, 8047cf8, 100403, ffffff295594dc48, ffffff002d2c9e68, 0) ffffff002d2c9ec0 ioctl+0x9b(4, 1, 8047cf8) ffffff002d2c9f10 sys_syscall32+0xff() idm_refcnt_wait wants refcount to be 0, but that doesn't happen. > ffffff1724512b88::print -t idm_refcnt_t idm_refcnt_t { int ir_refcnt = 0x1 [..] I found only two places where ir_refcnt is decremented : idm_refcnt_rele idm_refcnt_rele_and_destroy They seem likely to be called from iscsit_config_merge_tpg or iscsit_portal_offline But at this point I don't really know what to look for. Has anyone already seen this ? cheers, alex -------------- next part -------------- An HTML attachment was scrubbed... URL: From bbickerton at omniti.com Wed Oct 8 17:51:15 2014 From: bbickerton at omniti.com (Brian Bickerton) Date: Wed, 8 Oct 2014 13:51:15 -0400 Subject: [OmniOS-discuss] New EC2 AMIs for r151006 and r151012 Message-ID: Hi OmniOS EC2 users! A new r151006 AMI has been created (AMI name "OmniOS r151006_067") to fix the cert issues with "pkg update" as well as to pull in the latest packages. Additionally, a new r151012 AMI has been created (AMI name "OmniOS r151012"). They are both available in all regions. The AMI IDs and instructions for finding them are in the OmniOS cloud installation wiki: http://omnios.omniti.com/wiki.php/Installation#IntheCloud Brian Bickerton Web Engineer OmniTI Computer Consulting, Inc. -------------- next part -------------- An HTML attachment was scrubbed... URL: From keith at paskett.org Wed Oct 8 21:59:18 2014 From: keith at paskett.org (Keith Paskett) Date: Wed, 8 Oct 2014 15:59:18 -0600 Subject: [OmniOS-discuss] GCC 4.7 on OmniOS r151010 Message-ID: Seems like this should be easy, but has actually been quite infuriating. How do I install GCC 4.7 on r151010? No matter what I try, pkg search only shows me GCC 4.8. ---- Keith Paskett -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 203 bytes Desc: Message signed with OpenPGP using GPGMail URL: From magnus at yonderway.com Thu Oct 9 03:06:39 2014 From: magnus at yonderway.com (Magnus Hedemark) Date: Wed, 8 Oct 2014 23:06:39 -0400 Subject: [OmniOS-discuss] Anyone using Sun DHCP *server* still? In-Reply-To: <16712547-F363-4210-87FE-9A67F148D94D@omniti.com> References: <4DDB817D-8148-4A42-91FE-44C9F4C5BE20@omniti.com> <54354984.8060105@gmail.com> <16712547-F363-4210-87FE-9A67F148D94D@omniti.com> Message-ID: On Oct 8, 2014, at 10:38 AM, Dan McDonald wrote: > > > We're in brand-new territory here, so to be honest, we don't have any procedures as of yet. This bloody cycle (r151013) gives us a chance to work it out. It's why I'm *asking*. Just some unsolicited thoughts on how to navigate the new territory? Maybe announce deprecated features in release notes, with a plan of intent for when a deprecated feature will be removed? That?s probably a more reliable way to get word out than the mailing list, so that people don?t get caught by surprise when they upgrade. I?m not a user of this feature. No skin in that game. But this might be a good opportunity to figure out how to EOL legacy features in a way that users can plan around. -M From magnus at yonderway.com Thu Oct 9 03:12:08 2014 From: magnus at yonderway.com (Magnus Hedemark) Date: Wed, 8 Oct 2014 23:12:08 -0400 Subject: [OmniOS-discuss] GCC 4.7 on OmniOS r151010 In-Reply-To: References: Message-ID: <99523FA2-7F90-4B79-A2E4-9CA6F19078C9@yonderway.com> On Oct 8, 2014, at 5:59 PM, Keith Paskett wrote: > Seems like this should be easy, but has actually been quite infuriating. > > How do I install GCC 4.7 on r151010? No matter what I try, pkg search only shows me GCC 4.8. You?re not crazy. developer/gcc44 developer/gcc48 That?s all that seems to be available right now. -M From rt at steait.net Thu Oct 9 10:02:51 2014 From: rt at steait.net (Rune Tipsmark) Date: Thu, 9 Oct 2014 10:02:51 +0000 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2B7D@AIRA-SRV.aira.local> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2B7D@AIRA-SRV.aira.local> Message-ID: <84b71eba9b5f43828f944cddb42f0160@EX1301.steait.net> Thanks for the input Filip, I think I found the answer though... no VAAI in OmniOS/ZFS VAAI provides amongst other features: Thin Provisioning in ESXi 5.x and later hosts, which allows the ESXi host to tell the array when the space previously occupied by a virtual machine (whether it is deleted or migrated to another datastore) can be reclaimed on thin provisioned LUNs. Next questions is, will VAAI be supported in OmniOS/ZFS Br, Rune -----Original Message----- From: OmniOS-discuss [mailto:omnios-discuss-bounces at lists.omniti.com] On Behalf Of Filip Marvan Sent: Wednesday, October 08, 2014 7:53 AM To: omnios-discuss at lists.omniti.com Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files Hello, maybe you will have to enable compression on that ZVOL, and rewrite that disk with zeroes from client. Because your pool propably don't even know from vmware client, that you delete your data. Filip -------------------------------------------------- Date: Tue, 7 Oct 2014 23:00:47 +0000 From: Rune Tipsmark To: omnios-discuss Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files Message-ID: <3dde7c3ae1994829b588d3df09855cf0 at EX1301.steait.net> Content-Type: text/plain; charset="us-ascii" hi guys, wondering if someone might know why my pool is still allocated 1.45T after I removed the files on the LU's provisioned onto that pool. pool: pool02 state: ONLINE scan: scrub in progress since Wed Oct 8 02:31:41 2014 29.1G scanned out of 1.45T at 52.4M/s, 7h54m to go 0 repaired, 1.96% done config: NAME STATE READ WRITE CKSUM pool02 ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c4t5000C500570858EFd0 ONLINE 0 0 0 c4t5000C50057085A6Bd0 ONLINE 0 0 0 mirror-1 ONLINE 0 0 0 c4t5000C500570870D3d0 ONLINE 0 0 0 c4t5000C50057089753d0 ONLINE 0 0 0 logs c11d0 ONLINE 0 0 0 cache c13d0 ONLINE 0 0 0 c15d0 ONLINE 0 0 0 Supporting 3 LU's mounted in Vmware, I formatted the LUNs again and no change... Do I really need to delete the pool, then re-create it? br, Rune From filip.marvan at aira.cz Thu Oct 9 10:21:01 2014 From: filip.marvan at aira.cz (Filip Marvan) Date: Thu, 9 Oct 2014 12:21:01 +0200 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files Message-ID: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> Hello, yes, VAAI could be probably the answer for vmware, and for example NexentaStor have some VAAI support as I know, but there were many problems with that (based on posts from users on Nexenta's forum) like freezing under heavy load and so on. I gues, it will be very complicated to bring VAAI support to OmniOS. Filip >------------------------------------- >Date: Thu, 9 Oct 2014 10:02:51 +0000 >From: Rune Tipsmark >To: Filip Marvan , > "omnios-discuss at lists.omniti.com" >Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after > removing all files >Message-ID: <84b71eba9b5f43828f944cddb42f0160 at EX1301.steait.net> >Content-Type: text/plain; charset="us-ascii" > >Thanks for the input Filip, > >I think I found the answer though... no VAAI in OmniOS/ZFS > >VAAI provides amongst other features: > >Thin Provisioning in ESXi 5.x and later hosts, which allows the ESXi host to tell the array when the >space previously occupied by a virtual machine (whether it is deleted or migrated to another >datastore) can be reclaimed on thin provisioned LUNs. > >Next questions is, will VAAI be supported in OmniOS/ZFS > >Br, >Rune From rt at steait.net Thu Oct 9 10:39:21 2014 From: rt at steait.net (Rune Tipsmark) Date: Thu, 9 Oct 2014 10:39:21 +0000 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> Message-ID: Yeah, I searched and found a few threads about it, seems like it won't happen anytime soon. I would actually pay for that feature in OmniOS. Rune -----Original Message----- From: Filip Marvan [mailto:filip.marvan at aira.cz] Sent: Thursday, October 09, 2014 3:21 AM To: omnios-discuss at lists.omniti.com Cc: Rune Tipsmark Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files Hello, yes, VAAI could be probably the answer for vmware, and for example NexentaStor have some VAAI support as I know, but there were many problems with that (based on posts from users on Nexenta's forum) like freezing under heavy load and so on. I gues, it will be very complicated to bring VAAI support to OmniOS. Filip >------------------------------------- >Date: Thu, 9 Oct 2014 10:02:51 +0000 >From: Rune Tipsmark >To: Filip Marvan , > "omnios-discuss at lists.omniti.com" >Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after > removing all files >Message-ID: <84b71eba9b5f43828f944cddb42f0160 at EX1301.steait.net> >Content-Type: text/plain; charset="us-ascii" > >Thanks for the input Filip, > >I think I found the answer though... no VAAI in OmniOS/ZFS > >VAAI provides amongst other features: > >Thin Provisioning in ESXi 5.x and later hosts, which allows the ESXi host to tell the array when the >space previously occupied by a virtual machine (whether it is deleted or migrated to another >datastore) can be reclaimed on thin provisioned LUNs. > >Next questions is, will VAAI be supported in OmniOS/ZFS > >Br, >Rune From richard.elling at richardelling.com Thu Oct 9 15:21:10 2014 From: richard.elling at richardelling.com (Richard Elling) Date: Thu, 9 Oct 2014 08:21:10 -0700 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> Message-ID: <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> DanMcD will know for sure, but vols do support SCSI UNMAP over comstar. The "missing" support is for ZFS to issue SCSI UNMAP commands to the disks. -- richard On Oct 9, 2014, at 3:39 AM, Rune Tipsmark wrote: > Yeah, I searched and found a few threads about it, seems like it won't happen anytime soon. > > I would actually pay for that feature in OmniOS. > > Rune > > -----Original Message----- > From: Filip Marvan [mailto:filip.marvan at aira.cz] > Sent: Thursday, October 09, 2014 3:21 AM > To: omnios-discuss at lists.omniti.com > Cc: Rune Tipsmark > Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files > > Hello, > > yes, VAAI could be probably the answer for vmware, and for example NexentaStor have some VAAI support as I know, but there were many problems with that (based on posts from users on Nexenta's forum) like freezing under heavy load and so on. > > I gues, it will be very complicated to bring VAAI support to OmniOS. > > Filip > >> ------------------------------------- >> Date: Thu, 9 Oct 2014 10:02:51 +0000 >> From: Rune Tipsmark >> To: Filip Marvan , >> "omnios-discuss at lists.omniti.com" >> Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after >> removing all files >> Message-ID: <84b71eba9b5f43828f944cddb42f0160 at EX1301.steait.net> >> Content-Type: text/plain; charset="us-ascii" >> >> Thanks for the input Filip, >> >> I think I found the answer though... no VAAI in OmniOS/ZFS >> >> VAAI provides amongst other features: >> >> Thin Provisioning in ESXi 5.x and later hosts, which allows the ESXi host to tell the array when the >space previously occupied by a virtual machine (whether it is deleted or migrated to another >datastore) can be reclaimed on thin provisioned LUNs. >> >> Next questions is, will VAAI be supported in OmniOS/ZFS >> >> Br, >> Rune > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From danmcd at omniti.com Thu Oct 9 15:21:38 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 9 Oct 2014 11:21:38 -0400 Subject: [OmniOS-discuss] GCC 4.7 on OmniOS r151010 In-Reply-To: <99523FA2-7F90-4B79-A2E4-9CA6F19078C9@yonderway.com> References: <99523FA2-7F90-4B79-A2E4-9CA6F19078C9@yonderway.com> Message-ID: <63AFCD5A-AF0A-4749-8980-19D5D9C67950@omniti.com> On Oct 8, 2014, at 11:12 PM, Magnus Hedemark wrote: > On Oct 8, 2014, at 5:59 PM, Keith Paskett wrote: > >> Seems like this should be easy, but has actually been quite infuriating. >> >> How do I install GCC 4.7 on r151010? No matter what I try, pkg search only shows me GCC 4.8. > > You?re not crazy. > > developer/gcc44 > developer/gcc48 > > That?s all that seems to be available right now. With the advent of per-release repos, things we discontinue from old releases are no longer available. This was the case for gcc47. Dan From danmcd at omniti.com Thu Oct 9 15:46:25 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 9 Oct 2014 11:46:25 -0400 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> Message-ID: <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> On Oct 9, 2014, at 11:21 AM, Richard Elling wrote: > DanMcD will know for sure, but vols do support SCSI UNMAP over comstar. AND there are improvements to this for ZFS compliments of Delphix. > The "missing" support is for ZFS to issue SCSI UNMAP commands to the disks. That is true. Dan From rt at steait.net Thu Oct 9 17:51:49 2014 From: rt at steait.net (Rune Tipsmark) Date: Thu, 9 Oct 2014 17:51:49 +0000 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> Message-ID: <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> I am happy to test in our environment, we have some semi-live demo servers (10+ virtual servers) that are being used as in a real production environment with load on them etc. If some VAAI works, should I be able to see it in VMware? Does it make any difference I use SRP to connect the LU's? My VAAI status for each LUN is as follows: naa.600144f0908abf5d00005391079d0008 VAAI Plugin Name: ATS Status: unsupported Clone Status: unsupported Zero Status: supported Delete Status: unsupported Br, Rune -----Original Message----- From: Dan McDonald [mailto:danmcd at omniti.com] Sent: Thursday, October 09, 2014 8:46 AM To: Richard Elling Cc: Rune Tipsmark; Filip Marvan; omnios-discuss at lists.omniti.com Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files On Oct 9, 2014, at 11:21 AM, Richard Elling wrote: > DanMcD will know for sure, but vols do support SCSI UNMAP over comstar. AND there are improvements to this for ZFS compliments of Delphix. > The "missing" support is for ZFS to issue SCSI UNMAP commands to the disks. That is true. Dan From danmcd at omniti.com Thu Oct 9 18:11:14 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 9 Oct 2014 14:11:14 -0400 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> Message-ID: <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> On Oct 9, 2014, at 1:51 PM, Rune Tipsmark wrote: > I am happy to test in our environment, we have some semi-live demo servers (10+ virtual servers) that are being used as in a real production environment with load on them etc. > > If some VAAI works, should I be able to see it in VMware? As "hardware assisted erase", I believe. > Does it make any difference I use SRP to connect the LU's? My VMware is weak. > My VAAI status for each LUN is as follows: > > naa.600144f0908abf5d00005391079d0008 > VAAI Plugin Name: > ATS Status: unsupported Corresponds to ATS, which is in Nexenta only for now. > Clone Status: unsupported Corresponds to XCOPY, which is in Nexenta only for now. > Zero Status: supported Corresponds to WRITE_SAME, which is in all illumos distros. > Delete Status: unsupported Corresponds to UNMAP, which is in all illumos distros. I'm surprised "delete status" isn't marked as supported, but I do seem to recall a recent push disabled UNMAP (delete status) by default. Let me look... ... Yeah... there's a global "zvol_unmap_enabled" now. It *should* be true by default, though. Which OmniOS revision are you on? Dan From rt at steait.net Thu Oct 9 18:25:51 2014 From: rt at steait.net (Rune Tipsmark) Date: Thu, 9 Oct 2014 18:25:51 +0000 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> Message-ID: <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> On OmniOS v11 r151010 -----Original Message----- From: Dan McDonald [mailto:danmcd at omniti.com] Sent: Thursday, October 09, 2014 11:11 AM To: Rune Tipsmark Cc: Richard Elling; Filip Marvan; omnios-discuss at lists.omniti.com Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files On Oct 9, 2014, at 1:51 PM, Rune Tipsmark wrote: > I am happy to test in our environment, we have some semi-live demo servers (10+ virtual servers) that are being used as in a real production environment with load on them etc. > > If some VAAI works, should I be able to see it in VMware? As "hardware assisted erase", I believe. > Does it make any difference I use SRP to connect the LU's? My VMware is weak. > My VAAI status for each LUN is as follows: > > naa.600144f0908abf5d00005391079d0008 > VAAI Plugin Name: > ATS Status: unsupported Corresponds to ATS, which is in Nexenta only for now. > Clone Status: unsupported Corresponds to XCOPY, which is in Nexenta only for now. > Zero Status: supported Corresponds to WRITE_SAME, which is in all illumos distros. > Delete Status: unsupported Corresponds to UNMAP, which is in all illumos distros. I'm surprised "delete status" isn't marked as supported, but I do seem to recall a recent push disabled UNMAP (delete status) by default. Let me look... ... Yeah... there's a global "zvol_unmap_enabled" now. It *should* be true by default, though. Which OmniOS revision are you on? Dan From danmcd at omniti.com Thu Oct 9 18:36:02 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 9 Oct 2014 14:36:02 -0400 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> Message-ID: <3599C645-8A66-49E0-94A4-A110CB5092D5@omniti.com> On Oct 9, 2014, at 2:25 PM, Rune Tipsmark wrote: > On OmniOS v11 r151010 And it turns out that global switch arrived past the *r151012* freeze, never mind the 010 one. In theory, the "HW zeroing" should be available. I'm not sure off the top of my head why it's not available in practice. Dan From danmcd at omniti.com Thu Oct 9 18:36:41 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 9 Oct 2014 14:36:41 -0400 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <3599C645-8A66-49E0-94A4-A110CB5092D5@omniti.com> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> <3599C645-8A66-49E0-94A4-A110CB5092D5@omniti.com> Message-ID: <3AD8817E-34CB-4CE3-A916-8A29382F85D3@omniti.com> I meant to say: In theory, the "HW Delete" should be available. I'm not sure off the top of my head why it's not available in practice. Dan From rt at steait.net Thu Oct 9 18:38:31 2014 From: rt at steait.net (Rune Tipsmark) Date: Thu, 9 Oct 2014 18:38:31 +0000 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <3AD8817E-34CB-4CE3-A916-8A29382F85D3@omniti.com> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> <3599C645-8A66-49E0-94A4-A110CB5092D5@omniti.com> <3AD8817E-34CB-4CE3-A916-8A29382F85D3@omniti.com> Message-ID: So if I just upgrade to latest it should be supported? Rune -----Original Message----- From: Dan McDonald [mailto:danmcd at omniti.com] Sent: Thursday, October 09, 2014 11:37 AM To: Rune Tipsmark Cc: Richard Elling; Filip Marvan; omnios-discuss at lists.omniti.com Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files I meant to say: In theory, the "HW Delete" should be available. I'm not sure off the top of my head why it's not available in practice. Dan From danmcd at omniti.com Thu Oct 9 18:51:20 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 9 Oct 2014 14:51:20 -0400 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> <3599C645-8A66-49E0-94A4-A110CB5092D5@omniti.com> <3AD8817E-34CB-4CE3-A916-8A29382F85D3@omniti.com> Message-ID: On Oct 9, 2014, at 2:38 PM, Rune Tipsmark wrote: > So if I just upgrade to latest it should be supported? It should be available in r151010! That's why I'm surprised. Dan From minikola at gmail.com Thu Oct 9 21:26:02 2014 From: minikola at gmail.com (Nikola M.) Date: Thu, 09 Oct 2014 23:26:02 +0200 Subject: [OmniOS-discuss] KVM disk performance r151012 In-Reply-To: <072017d3ce0ba6769b555d36be724f98@blackdot.be> References: <072017d3ce0ba6769b555d36be724f98@blackdot.be> Message-ID: <5436FD6A.6060200@gmail.com> On 10/ 3/14 10:59 AM, Jorge Schrauwen wrote: > But that detour aside... do you have the windows virtio drivers > installed? > If not, try installing them and switching your nic to virtio. Mentioning Virtio, illumos itself still does not have it's own virtio drivers, to run inside KVM? Anyone knows if that changed in meantime? It would be fine place for contribution... From sjorge+ml at blackdot.be Thu Oct 9 21:31:50 2014 From: sjorge+ml at blackdot.be (Jorge Schrauwen) Date: Thu, 09 Oct 2014 23:31:50 +0200 Subject: [OmniOS-discuss] KVM disk performance r151012 In-Reply-To: <5436FD6A.6060200@gmail.com> References: <072017d3ce0ba6769b555d36be724f98@blackdot.be> <5436FD6A.6060200@gmail.com> Message-ID: <4c4beacba7966d7af1373fce89c6c256@blackdot.be> vioblk got included in the latest release. However for some reason both caiman (installer) and kayak (network installer) do recognized virtio disk. The driver is correctly loaded. (SmartOS does not suffer form this, the driver code is the same) Some debugging still needs to happen, I give it a shot but failed to find what was broken. There is no virtio-net driver however. If all goes well bloody will have open-vm-tools soon so at least under vmware thing should work smoother soon. Regards Jorge On 2014-10-09 23:26, Nikola M. wrote: > On 10/ 3/14 10:59 AM, Jorge Schrauwen wrote: >> But that detour aside... do you have the windows virtio drivers >> installed? >> If not, try installing them and switching your nic to virtio. > Mentioning Virtio, illumos itself still does not have it's own virtio > drivers, to run inside KVM? > Anyone knows if that changed in meantime? > It would be fine place for contribution... > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From rt at steait.net Thu Oct 9 22:33:18 2014 From: rt at steait.net (Rune Tipsmark) Date: Thu, 9 Oct 2014 22:33:18 +0000 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> <3599C645-8A66-49E0-94A4-A110CB5092D5@omniti.com> <3AD8817E-34CB-4CE3-A916-8A29382F85D3@omniti.com> Message-ID: <52148cc65f0e41c2a538e5e66241970e@EX1301.steait.net> Is there a command I can run to check? Rune -----Original Message----- From: Dan McDonald [mailto:danmcd at omniti.com] Sent: Thursday, October 09, 2014 11:51 AM To: Rune Tipsmark Cc: omnios-discuss Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files On Oct 9, 2014, at 2:38 PM, Rune Tipsmark wrote: > So if I just upgrade to latest it should be supported? It should be available in r151010! That's why I'm surprised. Dan From rt at steait.net Thu Oct 9 23:58:15 2014 From: rt at steait.net (Rune Tipsmark) Date: Thu, 9 Oct 2014 23:58:15 +0000 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <52148cc65f0e41c2a538e5e66241970e@EX1301.steait.net> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> <3599C645-8A66-49E0-94A4-A110CB5092D5@omniti.com> <3AD8817E-34CB-4CE3-A916-8A29382F85D3@omniti.com> <52148cc65f0e41c2a538e5e66241970e@EX1301.steait.net> Message-ID: <127df2661d744d70b22377bd854de648@EX1301.steait.net> Just updated to latest version r151012 Still same... I checked for vdev settings, is there another place I can check? root at zfs10:/root# echo "::zfs_params" | mdb -k | grep vdev zfs_vdev_max_active = 0x3e8 zfs_vdev_sync_read_min_active = 0xa zfs_vdev_sync_read_max_active = 0xa zfs_vdev_sync_write_min_active = 0xa zfs_vdev_sync_write_max_active = 0xa zfs_vdev_async_read_min_active = 0x1 zfs_vdev_async_read_max_active = 0x3 zfs_vdev_async_write_min_active = 0x1 zfs_vdev_async_write_max_active = 0xa zfs_vdev_scrub_min_active = 0x1 zfs_vdev_scrub_max_active = 0x2 zfs_vdev_async_write_active_min_dirty_percent = 0x1e zfs_vdev_async_write_active_max_dirty_percent = 0x3c mdb: variable reference_tracking_enable not found: unknown symbol name mdb: variable reference_history not found: unknown symbol name zfs_vdev_cache_max = 0x4000 zfs_vdev_cache_size = 0x0 zfs_vdev_cache_bshift = 0x10 vdev_mirror_shift = 0x15 zfs_vdev_aggregation_limit = 0x20000 Rune -----Original Message----- From: OmniOS-discuss [mailto:omnios-discuss-bounces at lists.omniti.com] On Behalf Of Rune Tipsmark Sent: Thursday, October 09, 2014 3:33 PM To: Dan McDonald Cc: omnios-discuss Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files Is there a command I can run to check? Rune -----Original Message----- From: Dan McDonald [mailto:danmcd at omniti.com] Sent: Thursday, October 09, 2014 11:51 AM To: Rune Tipsmark Cc: omnios-discuss Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files On Oct 9, 2014, at 2:38 PM, Rune Tipsmark wrote: > So if I just upgrade to latest it should be supported? It should be available in r151010! That's why I'm surprised. Dan _______________________________________________ OmniOS-discuss mailing list OmniOS-discuss at lists.omniti.com http://lists.omniti.com/mailman/listinfo/omnios-discuss From chip at innovates.com Fri Oct 10 02:23:04 2014 From: chip at innovates.com (Schweiss, Chip) Date: Thu, 9 Oct 2014 21:23:04 -0500 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: <5432E672.2080805@ColoState.EDU> References: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> <5432E672.2080805@ColoState.EDU> Message-ID: On Mon, Oct 6, 2014 at 1:58 PM, Kevin Swab wrote: > I had this same problem after upgrading a system to r151012. nlockmgr > failed to start because "svc:/network/nfs/status:default" was disabled. > I enabled that service, then nlockmgr was happy. > > Just tried my 2nd system. r151010 nlockmgr starts after clearing maintenance mode. r151012 it will not start at all. nfs/status was enabled and online. The commonality I see on the two systems I have tried is they are both part of an HA cluster. So they don't import the pool at boot, but RSF-1 imports it with cache mapped to a different location. nlockmgr is becoming a real show stopper. -Chip > I've done 3 other upgrades to r151012 and none of them had a problem > with nlockmgr... > > Kevin > > On 10/06/2014 09:56 AM, Schweiss, Chip wrote: > > > > > > On Mon, Oct 6, 2014 at 9:59 AM, Dan McDonald > > wrote: > > > > > > On Oct 6, 2014, at 10:41 AM, Schweiss, Chip > > wrote: > > > > > > > > Anyone else seeing this in r151012? > > > > > > Any tips on collecting better information on this would be > appreciated. > > > > I saw this in once in 012, but not as persistently as you have. It > > cleared up for me with a single reboot, but that may have just been > > because I got lucky w.r.t. statd. > > > > I take it "svcadm disable nlockmgr ; svcadm enable nlockmgr" doesn't > > help? > > > > > > I didn't think to try that. Tried rebooting, but that didn't help. I > > was already past my maintenance window, so I resorted to backing out to > > r151010. > > > > > > > > > > And as for output, I've seen in /var/adm/messages what you have, per > > the cited illumos bug (4518). > > > > > > > > I wasn't aware of any new NFS changes that may have caused this to > > become exaggerated in r151012. But, it's not clear on what is > > triggering this bug either. On my larger pool systems with lots of NFS > > exports, almost every reboot causes this and a 'svcadm clear' fixes it. > > > > Is there something different that a disable and enable does? I can > > likely get a small window in the next few days to give it a try. > > > > -Chip > > > > > > _______________________________________________ > > OmniOS-discuss mailing list > > OmniOS-discuss at lists.omniti.com > > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > > > -- > ------------------------------------------------------------------- > Kevin Swab UNIX Systems Administrator > ACNS Colorado State University > Phone: (970)491-6572 Email: Kevin.Swab at ColoState.EDU > GPG Fingerprint: 7026 3F66 A970 67BD 6F17 8EB8 8A7D 142F 2392 791C > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Fri Oct 10 02:54:33 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 9 Oct 2014 22:54:33 -0400 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: References: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> <5432E672.2080805@ColoState.EDU> Message-ID: <5949A3B6-E21F-4FC5-9F31-0AAA91EFCBA2@omniti.com> On Oct 9, 2014, at 10:23 PM, Schweiss, Chip wrote: > On Mon, Oct 6, 2014 at 1:58 PM, Kevin Swab wrote: > I had this same problem after upgrading a system to r151012. nlockmgr > failed to start because "svc:/network/nfs/status:default" was disabled. > I enabled that service, then nlockmgr was happy. > > > Just tried my 2nd system. r151010 nlockmgr starts after clearing maintenance mode. r151012 it will not start at all. nfs/status was enabled and online. > > The commonality I see on the two systems I have tried is they are both part of an HA cluster. So they don't import the pool at boot, but RSF-1 imports it with cache mapped to a different location. That could be something HA Inc. needs to further test. We don't directly support RSF-1, after all. > nlockmgr is becoming a real show stopper. svcadm disable nlockmgr nfs/status svcadm enable nfs/status svcadm enable nlockmgr You may wish to discuss this on illumos as well, I'm not sure who all else is seeing this save me one time, and you seemingly a lot of times. Dan From danmcd at omniti.com Fri Oct 10 02:55:09 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 9 Oct 2014 22:55:09 -0400 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: References: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> <5432E672.2080805@ColoState.EDU> Message-ID: On Oct 9, 2014, at 10:23 PM, Schweiss, Chip wrote: > > The commonality I see on the two systems I have tried is they are both part of an HA cluster. So they don't import the pool at boot, but RSF-1 imports it with cache mapped to a different location. > > nlockmgr is becoming a real show stopper. Your dmesg(1M) output would be useful as well. Dan From chip at innovates.com Fri Oct 10 13:15:24 2014 From: chip at innovates.com (Schweiss, Chip) Date: Fri, 10 Oct 2014 08:15:24 -0500 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: <5949A3B6-E21F-4FC5-9F31-0AAA91EFCBA2@omniti.com> References: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> <5432E672.2080805@ColoState.EDU> <5949A3B6-E21F-4FC5-9F31-0AAA91EFCBA2@omniti.com> Message-ID: On Thu, Oct 9, 2014 at 9:54 PM, Dan McDonald wrote: > > On Oct 9, 2014, at 10:23 PM, Schweiss, Chip wrote: > > > Just tried my 2nd system. r151010 nlockmgr starts after clearing > maintenance mode. r151012 it will not start at all. nfs/status was > enabled and online. > > > > The commonality I see on the two systems I have tried is they are both > part of an HA cluster. So they don't import the pool at boot, but RSF-1 > imports it with cache mapped to a different location. > > That could be something HA Inc. needs to further test. We don't directly > support RSF-1, after all. > > I there really isn't anything different than an auto imported pool. I'm suspecting using an alternate cache location my be triggering something else to go wrong in the nlockmgr. Here's the command RSF-1 uses to import the pool: zpool import -c /opt/HAC/RSF-1/etc/volume-cache/nrgpool.cache -o cachefile=/opt/HAC/RSF-1/etc/v olume-cache/nrgpool.cache-live -o failmode=panic nrgpool After the pool import it puts the ip addresses back and is done. That happens in less than 1 second. In the mean time NFS services auto start and nlockmgr starts spinning. > > nlockmgr is becoming a real show stopper. > > svcadm disable nlockmgr nfs/status > svcadm enable nfs/status > svcadm enable nlockmgr > > You may wish to discuss this on illumos as well, I'm not sure who all else > is seeing this save me one time, and you seemingly a lot of times. > I did that this time, no joy. Today I'm working on a virtual setup with HA to see if I can get this reproduced on r151012. I thought this nlockmgr propblem was related to lots of nfs exports until, I ran into this on my SSD pool. It used to be able to fail over in about 3-5 seconds. It takes nlockmgr now sits in a spinning state for a few minutes and fails every time. A clear of the maintenance mode, brings it back nearly instantly. This is on r151010. On r151012 it fails every time. Hopefully I can reproduce and I'll start a new thread copying Illumos too. -Chip -------------- next part -------------- An HTML attachment was scrubbed... URL: From richard.elling at richardelling.com Fri Oct 10 14:26:46 2014 From: richard.elling at richardelling.com (Richard Elling) Date: Fri, 10 Oct 2014 07:26:46 -0700 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: References: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> <5432E672.2080805@ColoState.EDU> <5949A3B6-E21F-4FC5-9F31-0AAA91EFCBA2@omniti.com> Message-ID: > On Oct 10, 2014, at 6:15 AM, "Schweiss, Chip" wrote: > > >> On Thu, Oct 9, 2014 at 9:54 PM, Dan McDonald wrote: >> >> On Oct 9, 2014, at 10:23 PM, Schweiss, Chip wrote: >> >> > Just tried my 2nd system. r151010 nlockmgr starts after clearing maintenance mode. r151012 it will not start at all. nfs/status was enabled and online. >> > >> > The commonality I see on the two systems I have tried is they are both part of an HA cluster. So they don't import the pool at boot, but RSF-1 imports it with cache mapped to a different location. >> >> That could be something HA Inc. needs to further test. We don't directly support RSF-1, after all. > > I there really isn't anything different than an auto imported pool. I'm suspecting using an alternate cache location my be triggering something else to go wrong in the nlockmgr. no, these are totally separate subsystems. RSF-1 imports the pool. NFS sharing is started by the zpool command, in userland, via sharing after the dataset is mounted. You can do the same procedure manually... no magic pixie dust needed. > > Here's the command RSF-1 uses to import the pool: > zpool import -c /opt/HAC/RSF-1/etc/volume-cache/nrgpool.cache -o cachefile=/opt/HAC/RSF-1/etc/v > olume-cache/nrgpool.cache-live -o failmode=panic nrgpool > > After the pool import it puts the ip addresses back and is done. That happens in less than 1 second. > > In the mean time NFS services auto start and nlockmgr starts spinning. Perhaps share doesn't properly start all of the services? Does it work ok if you manually "svcadm enable" all of the NFS services? -- richard > > >> > nlockmgr is becoming a real show stopper. >> >> svcadm disable nlockmgr nfs/status >> svcadm enable nfs/status >> svcadm enable nlockmgr >> >> You may wish to discuss this on illumos as well, I'm not sure who all else is seeing this save me one time, and you seemingly a lot of times. > > I did that this time, no joy. Today I'm working on a virtual setup with HA to see if I can get this reproduced on r151012. > > I thought this nlockmgr propblem was related to lots of nfs exports until, I ran into this on my SSD pool. It used to be able to fail over in about 3-5 seconds. It takes nlockmgr now sits in a spinning state for a few minutes and fails every time. A clear of the maintenance mode, brings it back nearly instantly. This is on r151010. On r151012 it fails every time. > > Hopefully I can reproduce and I'll start a new thread copying Illumos too. > > -Chip > > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From richard.elling at richardelling.com Fri Oct 10 17:00:47 2014 From: richard.elling at richardelling.com (Richard Elling) Date: Fri, 10 Oct 2014 10:00:47 -0700 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <127df2661d744d70b22377bd854de648@EX1301.steait.net> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> <3599C645-8A66-49E0-94A4-A110CB5092D5@omniti.com> <3AD8817E-34CB-4CE3-A916-8A29382F85D3@omniti.com> <52148cc65f0e41c2a538e5e66241970e@EX1301.steait.net> <127df2661d744d70b22377bd854de648@EX1301.steait.net> Message-ID: <19800DA1-528D-4442-8326-63BB97B2B700@richardelling.com> On Oct 9, 2014, at 4:58 PM, Rune Tipsmark wrote: > Just updated to latest version r151012 > > Still same... I checked for vdev settings, is there another place I can check? It won't be a ZFS feature. On the initiator, use something like sg3_utils thusly: [root at congo ~]# sg_opcodes /dev/rdsk/c0t5000C50030117C3Bd0 SEAGATE ST800FM0043 0005 Peripheral device type: disk Opcode Service CDB Name (hex) action(h) size ----------------------------------------------- 00 6 Test Unit Ready 01 6 Rezero Unit 03 6 Request Sense 04 6 Format Unit 07 6 Reassign Blocks 08 6 Read(6) 0a 6 Write(6) 0b 6 Seek(6) 12 6 Inquiry 15 6 Mode select(6) 16 6 Reserve(6) 17 6 Release(6) 1a 6 Mode sense(6) 1b 6 Start stop unit 1c 6 Receive diagnostic results 1d 6 Send diagnostic 25 10 Read capacity(10) 28 10 Read(10) 2a 10 Write(10) 2b 10 Seek(10) 2e 10 Write and verify(10) 2f 10 Verify(10) 35 10 Synchronize cache(10) 37 10 Read defect data(10) 3b 0 10 Write buffer, combined header and data [or multiple modes] 3b 2 10 Write buffer, data 3b 4 10 Write buffer, download microcode and activate 3b 5 10 Write buffer, download microcode, save, and activate 3b 6 10 Write buffer, download microcode with offsets and activate 3b 7 10 Write buffer, download microcode with offsets, save, and activate 3b a 10 Write buffer, write data to echo buffer 3b d 10 Write buffer, download microcode with offsets, select activation events, save and defer activate 3b e 10 Write buffer, download microcode with offsets, save and defer activate 3b f 10 Write buffer, activate deferred microcode 3b 1a 10 Write buffer, enable expander comms protocol and echo buffer 3b 1c 10 Write buffer, download application client error history 3c 0 10 Read buffer, combined header and data [or multiple modes] 3c 2 10 Read buffer, data 3c 3 10 Read buffer, descriptor 3c a 10 Read buffer, read data from echo buffer 3c b 10 Read buffer, echo buffer descriptor 3c 1c 10 Read buffer, error history 3e 10 Read long(10) 3f 10 Write long(10) 41 10 Write same(10) 42 10 Unmap 48 2 10 Sanitize, block erase 48 1f 10 Sanitize, exit failure mode 4c 10 Log select 4d 10 Log sense 55 10 Mode select(10) 56 10 Reserve(10) 57 10 Release(10) 5a 10 Mode sense(10) 5e 0 10 Persistent reserve in, read keys 5e 1 10 Persistent reserve in, read reservation 5e 2 10 Persistent reserve in, report capabilities 5e 3 10 Persistent reserve in, read full status 5f 0 10 Persistent reserve out, register 5f 1 10 Persistent reserve out, reserve 5f 2 10 Persistent reserve out, release 5f 3 10 Persistent reserve out, clear 5f 4 10 Persistent reserve out, preempt 5f 5 10 Persistent reserve out, preempt and abort 5f 6 10 Persistent reserve out, register and ignore existing key 5f 7 10 Persistent reserve out, register and move 7f 9 32 Read(32) 7f a 32 Verify(32) 7f b 32 Write(32) 7f c 32 Write an verify(32) 7f d 32 Write same(32) 88 16 Read(16) 8a 16 Write(16) 8e 16 Write and verify(16) 8f 16 Verify(16) 91 16 Synchronize cache(16) 93 16 Write same(16) 9e 10 16 Read capacity(16) 9e 11 16 Read long(16) 9f 11 16 Write long(16) a0 12 Report luns a3 5 12 Report identifying information a3 c 12 Report supported operation codes a3 d 12 Report supported task management functions a4 6 12 Set identifying information a4 f 12 Set timestamp b7 12 Read defect data(12) e0 10 Vendor specific [0xe0] e1 10 Vendor specific [0xe1] e2 10 Vendor specific [0xe2] e6 10 Vendor specific [0xe6] f7 10 Vendor specific [0xf7] I'd try it for iSCSI, but since I no longer use iSCSI (yea AoE!) I can't test :-P -- richard > > root at zfs10:/root# echo "::zfs_params" | mdb -k | grep vdev > zfs_vdev_max_active = 0x3e8 > zfs_vdev_sync_read_min_active = 0xa > zfs_vdev_sync_read_max_active = 0xa > zfs_vdev_sync_write_min_active = 0xa > zfs_vdev_sync_write_max_active = 0xa > zfs_vdev_async_read_min_active = 0x1 > zfs_vdev_async_read_max_active = 0x3 > zfs_vdev_async_write_min_active = 0x1 > zfs_vdev_async_write_max_active = 0xa > zfs_vdev_scrub_min_active = 0x1 > zfs_vdev_scrub_max_active = 0x2 > zfs_vdev_async_write_active_min_dirty_percent = 0x1e > zfs_vdev_async_write_active_max_dirty_percent = 0x3c > mdb: variable reference_tracking_enable not found: unknown symbol name > mdb: variable reference_history not found: unknown symbol name > zfs_vdev_cache_max = 0x4000 > zfs_vdev_cache_size = 0x0 > zfs_vdev_cache_bshift = 0x10 > vdev_mirror_shift = 0x15 > zfs_vdev_aggregation_limit = 0x20000 > > Rune > > -----Original Message----- > From: OmniOS-discuss [mailto:omnios-discuss-bounces at lists.omniti.com] On Behalf Of Rune Tipsmark > Sent: Thursday, October 09, 2014 3:33 PM > To: Dan McDonald > Cc: omnios-discuss > Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files > > Is there a command I can run to check? > > Rune > > -----Original Message----- > From: Dan McDonald [mailto:danmcd at omniti.com] > Sent: Thursday, October 09, 2014 11:51 AM > To: Rune Tipsmark > Cc: omnios-discuss > Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files > > > On Oct 9, 2014, at 2:38 PM, Rune Tipsmark wrote: > >> So if I just upgrade to latest it should be supported? > > It should be available in r151010! That's why I'm surprised. > > Dan > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From chip at innovates.com Fri Oct 10 19:09:46 2014 From: chip at innovates.com (Schweiss, Chip) Date: Fri, 10 Oct 2014 14:09:46 -0500 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: References: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> <5432E672.2080805@ColoState.EDU> <5949A3B6-E21F-4FC5-9F31-0AAA91EFCBA2@omniti.com> Message-ID: Apparently something common in my OmniOS setup is triggering this. I have no idea what yet, and I'm feeling green at digging through this issue. On one of my VMs for doing script development, I exported the data pool planing to test importing it with a different cache location and the problem immediately happened. Now I cannot get nlockmgr to start at all on this VM. I tried disabling all nfs services and re-enabling. Still failing with /usr/lib/nfs/lockd[862]: [ID 491006 daemon.error] Cannot establish NLM service over : I/O error. Exiting root at ZFSsendTest1:/root# svcs -a|grep nfs disabled 13:47:05 svc:/network/nfs/log:default disabled 13:47:11 svc:/network/nfs/rquota:default disabled 13:55:05 svc:/network/nfs/server:default disabled 13:55:32 svc:/network/nfs/nlockmgr:default disabled 13:55:32 svc:/network/nfs/mapid:default disabled 13:55:32 svc:/network/nfs/status:default disabled 13:55:32 svc:/network/nfs/client:default disabled 13:55:57 svc:/network/nfs/cbd:default root at ZFSsendTest1:/root# svcadm enable svc:/network/nfs/status:default svc:/network/nfs/cbd:default svc:/network/nfs/mapid:default svc:/network/nfs/server:default svc:/network/nfs/nlockmgr:default root at ZFSsendTest1:/root# svcs -a|grep nfs disabled 13:47:05 svc:/network/nfs/log:default disabled 13:47:11 svc:/network/nfs/rquota:default disabled 13:55:32 svc:/network/nfs/client:default online 13:56:56 svc:/network/nfs/status:default online 13:56:56 svc:/network/nfs/cbd:default online 13:56:56 svc:/network/nfs/mapid:default offline 13:56:56 svc:/network/nfs/server:default offline* 13:56:56 svc:/network/nfs/nlockmgr:default root at ZFSsendTest1:/root# svcs -a|grep nfs disabled 13:47:05 svc:/network/nfs/log:default disabled 13:47:11 svc:/network/nfs/rquota:default disabled 13:55:32 svc:/network/nfs/client:default online 13:56:56 svc:/network/nfs/status:default online 13:56:56 svc:/network/nfs/cbd:default online 13:56:56 svc:/network/nfs/mapid:default offline 13:56:56 svc:/network/nfs/server:default maintenance 13:58:11 svc:/network/nfs/nlockmgr:default This VM has never had RSF-1 on it, so that definitely isn't the trigger. This VM has never exhibited this problem before today. It has been rebooted many times. I wonder if the problem is triggered by exporting a pool with NFS exports that have active client connections. That is always the case on my production systems. This VM has one NFS client that was connected when I exported the pool. Now nlockmgr dies and goes to maintenance mode regardless if I import the data pool or not. Any advice on where to dig for better diagnosis of this would be helpful. If any developers would like to get access to this VM I'd be happy to arrange that too. -Chip On Fri, Oct 10, 2014 at 9:26 AM, Richard Elling < richard.elling at richardelling.com> wrote: > > On Oct 10, 2014, at 6:15 AM, "Schweiss, Chip" wrote: > > > On Thu, Oct 9, 2014 at 9:54 PM, Dan McDonald wrote: > >> >> On Oct 9, 2014, at 10:23 PM, Schweiss, Chip wrote: >> >> > Just tried my 2nd system. r151010 nlockmgr starts after clearing >> maintenance mode. r151012 it will not start at all. nfs/status was >> enabled and online. >> > >> > The commonality I see on the two systems I have tried is they are both >> part of an HA cluster. So they don't import the pool at boot, but RSF-1 >> imports it with cache mapped to a different location. >> >> That could be something HA Inc. needs to further test. We don't directly >> support RSF-1, after all. >> >> > I there really isn't anything different than an auto imported pool. I'm > suspecting using an alternate cache location my be triggering something > else to go wrong in the nlockmgr. > > > no, these are totally separate subsystems. RSF-1 imports the pool. NFS > sharing is started by the zpool command, in userland, via sharing after the > dataset is mounted. You can do the same procedure manually... no magic > pixie dust needed. > > > Here's the command RSF-1 uses to import the pool: > zpool import -c /opt/HAC/RSF-1/etc/volume-cache/nrgpool.cache -o > cachefile=/opt/HAC/RSF-1/etc/v > olume-cache/nrgpool.cache-live -o failmode=panic nrgpool > > After the pool import it puts the ip addresses back and is done. That > happens in less than 1 second. > > In the mean time NFS services auto start and nlockmgr starts spinning. > > > Perhaps share doesn't properly start all of the services? Does it work ok > if you manually "svcadm enable" all of the NFS services? > > > -- richard > > > > >> > nlockmgr is becoming a real show stopper. >> >> svcadm disable nlockmgr nfs/status >> svcadm enable nfs/status >> svcadm enable nlockmgr >> >> You may wish to discuss this on illumos as well, I'm not sure who all >> else is seeing this save me one time, and you seemingly a lot of times. >> > > I did that this time, no joy. Today I'm working on a virtual setup with > HA to see if I can get this reproduced on r151012. > > I thought this nlockmgr propblem was related to lots of nfs exports until, > I ran into this on my SSD pool. It used to be able to fail over in about > 3-5 seconds. It takes nlockmgr now sits in a spinning state for a few > minutes and fails every time. A clear of the maintenance mode, brings it > back nearly instantly. This is on r151010. On r151012 it fails every > time. > > Hopefully I can reproduce and I'll start a new thread copying Illumos too. > > -Chip > > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Fri Oct 10 20:24:54 2014 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 10 Oct 2014 16:24:54 -0400 Subject: [OmniOS-discuss] What do you *actually* want when you want to build illumos-gate? Message-ID: <5EE18079-52B1-420A-B2B3-6C8DFAD88DF0@omniti.com> Pardon the long subject line. Some people have asked, "Why can't I build illumos-gate on OmniOS?" It's a legitimate question, and I'm in the middle of an attempt to say, "Well, now you can." As I'm doing this, however, I want to make sure expectations for you (and for me) are properly set. So my subject-line question stands. Are you interested because: - I want to make sure my changes compile cleanly. - I want to make sure my changes lint cleanly. - I want to generate replacement binaries (e.g. /kernel/drv/amd64/e1000g). - I want to generate packages to ONU on my OmniOS box. - More? I can answer now with my take on the above. - I want to make sure my changes compile. I believe I have this working now. It will require some .env file hacking on an OmniOS build box, but I will include a sample when I upstream. - I want to make sure my changes lint cleanly. Not yet, but there's a set of changes ALREADY in OmniOS (like the already in use java ones for this project) that may need to show up in illumos-gate as well. - I want to generate replacement binaries. You should be able to do this. And if you build small (per http://kebesays.blogspot.com/2011/03/for-illumos-newbies-on-developing-small.html), you probably already could more a good chunk of binaries. - I want to generate packages to ONU on my OmniOS box. This will be REALLY hard. If you really want to do this, I'd just use illumos-omnios. - More? I'm all ears for now. Before I go any further along, I want to make sure I'm solving people's problems, or that they are, indeed, solvable. Thanks, Dan From rt at steait.net Fri Oct 10 20:58:22 2014 From: rt at steait.net (Rune Tipsmark) Date: Fri, 10 Oct 2014 20:58:22 +0000 Subject: [OmniOS-discuss] ZFS pool allocation remains after removing all files In-Reply-To: <19800DA1-528D-4442-8326-63BB97B2B700@richardelling.com> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2BFC@AIRA-SRV.aira.local> <056B925E-9202-46DA-8AB2-22D741122C96@richardelling.com> <8BE5066F-6143-4907-9C7F-C2134AB7986C@omniti.com> <7facc48a42cb4a51b9fd096c3c39e9ca@EX1301.steait.net> <0B94C1FB-8659-4024-9815-383809F660F0@omniti.com> <1471cfc65e4d453a98d873dfc6608a6e@EX1301.steait.net> <3599C645-8A66-49E0-94A4-A110CB5092D5@omniti.com> <3AD8817E-34CB-4CE3-A916-8A29382F85D3@omniti.com> <52148cc65f0e41c2a538e5e66241970e@EX1301.steait.net> <127df2661d744d70b22377bd854de648@EX1301.steait.net> <19800DA1-528D-4442-8326-63BB97B2B700@richardelling.com> Message-ID: Same acceleration on iSCSI naa.600144f0908abf5d0000539106e40001 VAAI Plugin Name: ATS Status: unsupported Clone Status: unsupported Zero Status: supported Delete Status: unsupported Rune -----Original Message----- From: Richard Elling [mailto:richard.elling at richardelling.com] Sent: Friday, October 10, 2014 10:01 AM To: Rune Tipsmark Cc: Dan McDonald; omnios-discuss Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after removing all files On Oct 9, 2014, at 4:58 PM, Rune Tipsmark wrote: > Just updated to latest version r151012 > > Still same... I checked for vdev settings, is there another place I can check? It won't be a ZFS feature. On the initiator, use something like sg3_utils thusly: [root at congo ~]# sg_opcodes /dev/rdsk/c0t5000C50030117C3Bd0 SEAGATE ST800FM0043 0005 Peripheral device type: disk Opcode Service CDB Name (hex) action(h) size ----------------------------------------------- 00 6 Test Unit Ready 01 6 Rezero Unit 03 6 Request Sense 04 6 Format Unit 07 6 Reassign Blocks 08 6 Read(6) 0a 6 Write(6) 0b 6 Seek(6) 12 6 Inquiry 15 6 Mode select(6) 16 6 Reserve(6) 17 6 Release(6) 1a 6 Mode sense(6) 1b 6 Start stop unit 1c 6 Receive diagnostic results 1d 6 Send diagnostic 25 10 Read capacity(10) 28 10 Read(10) 2a 10 Write(10) 2b 10 Seek(10) 2e 10 Write and verify(10) 2f 10 Verify(10) 35 10 Synchronize cache(10) 37 10 Read defect data(10) 3b 0 10 Write buffer, combined header and data [or multiple modes] 3b 2 10 Write buffer, data 3b 4 10 Write buffer, download microcode and activate 3b 5 10 Write buffer, download microcode, save, and activate 3b 6 10 Write buffer, download microcode with offsets and activate 3b 7 10 Write buffer, download microcode with offsets, save, and activate 3b a 10 Write buffer, write data to echo buffer 3b d 10 Write buffer, download microcode with offsets, select activation events, save and defer activate 3b e 10 Write buffer, download microcode with offsets, save and defer activate 3b f 10 Write buffer, activate deferred microcode 3b 1a 10 Write buffer, enable expander comms protocol and echo buffer 3b 1c 10 Write buffer, download application client error history 3c 0 10 Read buffer, combined header and data [or multiple modes] 3c 2 10 Read buffer, data 3c 3 10 Read buffer, descriptor 3c a 10 Read buffer, read data from echo buffer 3c b 10 Read buffer, echo buffer descriptor 3c 1c 10 Read buffer, error history 3e 10 Read long(10) 3f 10 Write long(10) 41 10 Write same(10) 42 10 Unmap 48 2 10 Sanitize, block erase 48 1f 10 Sanitize, exit failure mode 4c 10 Log select 4d 10 Log sense 55 10 Mode select(10) 56 10 Reserve(10) 57 10 Release(10) 5a 10 Mode sense(10) 5e 0 10 Persistent reserve in, read keys 5e 1 10 Persistent reserve in, read reservation 5e 2 10 Persistent reserve in, report capabilities 5e 3 10 Persistent reserve in, read full status 5f 0 10 Persistent reserve out, register 5f 1 10 Persistent reserve out, reserve 5f 2 10 Persistent reserve out, release 5f 3 10 Persistent reserve out, clear 5f 4 10 Persistent reserve out, preempt 5f 5 10 Persistent reserve out, preempt and abort 5f 6 10 Persistent reserve out, register and ignore existing key 5f 7 10 Persistent reserve out, register and move 7f 9 32 Read(32) 7f a 32 Verify(32) 7f b 32 Write(32) 7f c 32 Write an verify(32) 7f d 32 Write same(32) 88 16 Read(16) 8a 16 Write(16) 8e 16 Write and verify(16) 8f 16 Verify(16) 91 16 Synchronize cache(16) 93 16 Write same(16) 9e 10 16 Read capacity(16) 9e 11 16 Read long(16) 9f 11 16 Write long(16) a0 12 Report luns a3 5 12 Report identifying information a3 c 12 Report supported operation codes a3 d 12 Report supported task management functions a4 6 12 Set identifying information a4 f 12 Set timestamp b7 12 Read defect data(12) e0 10 Vendor specific [0xe0] e1 10 Vendor specific [0xe1] e2 10 Vendor specific [0xe2] e6 10 Vendor specific [0xe6] f7 10 Vendor specific [0xf7] I'd try it for iSCSI, but since I no longer use iSCSI (yea AoE!) I can't test :-P -- richard > > root at zfs10:/root# echo "::zfs_params" | mdb -k | grep vdev > zfs_vdev_max_active = 0x3e8 zfs_vdev_sync_read_min_active = 0xa > zfs_vdev_sync_read_max_active = 0xa zfs_vdev_sync_write_min_active = > 0xa zfs_vdev_sync_write_max_active = 0xa > zfs_vdev_async_read_min_active = 0x1 zfs_vdev_async_read_max_active = > 0x3 zfs_vdev_async_write_min_active = 0x1 > zfs_vdev_async_write_max_active = 0xa zfs_vdev_scrub_min_active = 0x1 > zfs_vdev_scrub_max_active = 0x2 > zfs_vdev_async_write_active_min_dirty_percent = 0x1e > zfs_vdev_async_write_active_max_dirty_percent = 0x3c > mdb: variable reference_tracking_enable not found: unknown symbol name > mdb: variable reference_history not found: unknown symbol name > zfs_vdev_cache_max = 0x4000 zfs_vdev_cache_size = 0x0 > zfs_vdev_cache_bshift = 0x10 vdev_mirror_shift = 0x15 > zfs_vdev_aggregation_limit = 0x20000 > > Rune > > -----Original Message----- > From: OmniOS-discuss [mailto:omnios-discuss-bounces at lists.omniti.com] > On Behalf Of Rune Tipsmark > Sent: Thursday, October 09, 2014 3:33 PM > To: Dan McDonald > Cc: omnios-discuss > Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after > removing all files > > Is there a command I can run to check? > > Rune > > -----Original Message----- > From: Dan McDonald [mailto:danmcd at omniti.com] > Sent: Thursday, October 09, 2014 11:51 AM > To: Rune Tipsmark > Cc: omnios-discuss > Subject: Re: [OmniOS-discuss] ZFS pool allocation remains after > removing all files > > > On Oct 9, 2014, at 2:38 PM, Rune Tipsmark wrote: > >> So if I just upgrade to latest it should be supported? > > It should be available in r151010! That's why I'm surprised. > > Dan > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From sjorge+ml at blackdot.be Fri Oct 10 22:01:33 2014 From: sjorge+ml at blackdot.be (Jorge Schrauwen) Date: Sat, 11 Oct 2014 00:01:33 +0200 Subject: [OmniOS-discuss] =?utf-8?q?What_do_you_*actually*_want_when_you_w?= =?utf-8?q?ant_to_build_illumos-gate=3F?= In-Reply-To: <5EE18079-52B1-420A-B2B3-6C8DFAD88DF0@omniti.com> References: <5EE18079-52B1-420A-B2B3-6C8DFAD88DF0@omniti.com> Message-ID: <86f2fe543d2926f1f556d6cebadcdd64@blackdot.be> For me one and two... I can test stuff on illumos-omnos... if they seem good to upstream I'd love to be able to do one and two to make sure they don't break things. Regards Jorge On 2014-10-10 22:24, Dan McDonald wrote: > Pardon the long subject line. > > Some people have asked, "Why can't I build illumos-gate on OmniOS?" > It's a legitimate question, and I'm in the middle of an attempt to > say, "Well, now you can." > > As I'm doing this, however, I want to make sure expectations for you > (and for me) are properly set. So my subject-line question stands. > Are you interested because: > > - I want to make sure my changes compile cleanly. > > - I want to make sure my changes lint cleanly. > > - I want to generate replacement binaries (e.g. > /kernel/drv/amd64/e1000g). > > - I want to generate packages to ONU on my OmniOS box. > > - More? > > I can answer now with my take on the above. > > - I want to make sure my changes compile. > I believe I have this working now. It will require some .env file > hacking on an OmniOS build box, but I will include a sample when I > upstream. > > - I want to make sure my changes lint cleanly. > Not yet, but there's a set of changes ALREADY in OmniOS (like the > already in use java ones for this project) that may need to show up in > illumos-gate as well. > > - I want to generate replacement binaries. > You should be able to do this. And if you build small (per > http://kebesays.blogspot.com/2011/03/for-illumos-newbies-on-developing-small.html), > you probably already could more a good chunk of binaries. > > - I want to generate packages to ONU on my OmniOS box. > This will be REALLY hard. If you really want to do this, I'd just use > illumos-omnios. > > - More? > I'm all ears for now. > > > Before I go any further along, I want to make sure I'm solving > people's problems, or that they are, indeed, solvable. > > Thanks, > Dan > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From henson at acm.org Sat Oct 11 00:15:28 2014 From: henson at acm.org (Paul B. Henson) Date: Fri, 10 Oct 2014 17:15:28 -0700 Subject: [OmniOS-discuss] What do you *actually* want when you want to build illumos-gate? Message-ID: <23c401cfe4e8$78cb2ac0$6a618040$@acm.org> > From: Dan McDonald > Sent: Friday, October 10, 2014 1:25 PM > > - I want to make sure my changes compile cleanly. > > - I want to make sure my changes lint cleanly. > > - I want to generate replacement binaries (e.g. > /kernel/drv/amd64/e1000g). I'd say my main goal would be to do upstream illumos development without having to have an OI box around :). I've upstreamed a couple of things based on illumos-omnios, but as you need a bloody omnios box to compile bloody illumos-omnios, that's a bit problematic. I don't really want an unstable development box :). Ideally I'd like to be able to work on upstream illumos-gate code on a stable omnios box. So that would definitely include the first two line items, as well as whatever else is required for an RTI submission. At least for me, for replacement binaries, I'd probably just use stable illumos-omnios and backport whatever new bits I think I need. Depending on the divergence between stable omnios and upstream illumos-gate, and what piece you're playing with (particularly kernel bits?), pulling something out of upstream and trying to shove it in stable sounds potentially unstable ;). Thanks. From fabio at fabiorabelo.wiki.br Mon Oct 13 13:18:52 2014 From: fabio at fabiorabelo.wiki.br (=?UTF-8?Q?F=C3=A1bio_Rabelo?=) Date: Mon, 13 Oct 2014 10:18:52 -0300 Subject: [OmniOS-discuss] napp-it forum ? Message-ID: Anyone knows a forum about nnap-it ? I cannot found anything in the site .... F?bio Rabelo -------------- next part -------------- An HTML attachment was scrubbed... URL: From johan.kragsterman at capvert.se Mon Oct 13 13:27:02 2014 From: johan.kragsterman at capvert.se (Johan Kragsterman) Date: Mon, 13 Oct 2014 15:27:02 +0200 Subject: [OmniOS-discuss] Ang: napp-it forum ? In-Reply-To: References: Message-ID: An HTML attachment was scrubbed... URL: From groups at tierarzt-mueller.de Mon Oct 13 17:49:03 2014 From: groups at tierarzt-mueller.de (Alexander Lesle) Date: Mon, 13 Oct 2014 19:49:03 +0200 Subject: [OmniOS-discuss] napp-it forum ? In-Reply-To: References: Message-ID: <13510188261.20141013194903@tierarzt-mueller.de> Hello F?bio Rabelo and List, there is no forum but you get help here in German: http://www.hardwareluxx.de/community/f101/zfs-stammtisch-570052.html in English: http://hardforum.com/showthread.php?t=1573272 https://forums.servethehome.com/index.php?forums/solaris-nexenta-openindiana-and-napp-it.26/ In both forum @gea is there. On Oktober, 13 2014, 15:18 wrote in [1]: > Anyone knows a forum about nnap-it ? > I cannot found anything in the site .... -- Best Regards Alexander Oktober, 13 2014 ........ [1] mid:CAEekY65Dp-m-JxqePqeMmkz3V00LYMdT3wapN975=+5mszEpvA at mail.gmail.com ........ From danmcd at omniti.com Mon Oct 13 20:48:19 2014 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 13 Oct 2014 16:48:19 -0400 Subject: [OmniOS-discuss] Any "bloody" users out there use libm? Message-ID: Coming very soon upstream in illumos-gate is the fully open-sourced Math Library (libm). It may make it into this week's bloody update. If it does, I wanted to get a feel for who all would be affected by it. Thanks, Dan From matthew.lagoe at subrigo.net Wed Oct 15 00:15:54 2014 From: matthew.lagoe at subrigo.net (Matthew Lagoe) Date: Tue, 14 Oct 2014 17:15:54 -0700 Subject: [OmniOS-discuss] backup fiber channel SAN Message-ID: <016b01cfe80d$323fbad0$96bf3070$@subrigo.net> Has anyone ever done async backups of a fiber channel san to omnios and has any recommendations on how to accomplish this? Any ideas are greatly appreciated. Thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: From johan.kragsterman at capvert.se Wed Oct 15 04:10:17 2014 From: johan.kragsterman at capvert.se (Johan Kragsterman) Date: Wed, 15 Oct 2014 06:10:17 +0200 Subject: [OmniOS-discuss] Ang: backup fiber channel SAN In-Reply-To: <016b01cfe80d$323fbad0$96bf3070$@subrigo.net> References: <016b01cfe80d$323fbad0$96bf3070$@subrigo.net> Message-ID: -----"OmniOS-discuss" skrev: ----- Till: Fr?n: "Matthew Lagoe" S?nt av: "OmniOS-discuss" Datum: 2014-10-15 02:18 ?rende: [OmniOS-discuss] backup fiber channel SAN Has anyone ever done async backups of a fiber channel san to omnios and has any recommendations on how to accomplish this? ? Any ideas are greatly appreciated. ? Yeah, I do that all the time. I use omnios both as a SAN Storage controller, and as a SAN backup server. I use comstar on the backup server to provide LUN's to send backups to, both to the SAN Storage controller, and to the hosts. From the Storage controller I send mainly snapshots, and from hosts I can send both snapshots and use file backup services of any kind. Johan Thanks _______________________________________________ OmniOS-discuss mailing list OmniOS-discuss at lists.omniti.com http://lists.omniti.com/mailman/listinfo/omnios-discuss From filip.marvan at aira.cz Wed Oct 15 09:17:56 2014 From: filip.marvan at aira.cz (Filip Marvan) Date: Wed, 15 Oct 2014 11:17:56 +0200 Subject: [OmniOS-discuss] Crashing qemu process Message-ID: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2E93@AIRA-SRV.aira.local> Hi, I'm doing some tests with KVM on OmniOS. I have two virtual servers, one with Windows XP, which works without any problems, but the second one with Windows 2008 R2 server is crashing after some time. Qemu proces just crashed (after about one hour after VPS starts). There is nothing strange in /var/adm/messages, and i think that -d aption for debuging informations is not working (/tmp/qemu.log is empty even for -d exec for example). Do you have any idea, how can I find what causes, that qemu proces is crashing? Do anybody havy any experiences with something like this? Thank you, Filip Marvan My configuration looks like this: # configuration VNIC=vnic3 HDD=/dev/zvol/rdsk/data/server_system CD=/root/kvm/iso/virtio-win-0.1-81.iso VNC=8 MEM=8192 mac="2:8:20:dd:19:ed" /usr/bin/qemu-system-x86_64 \ -d exec \ -name "VPS" \ -boot c \ -enable-kvm \ -vnc 0.0.0.0:$VNC \ -smp cores=4,threads=1,sockets=1 \ -m $MEM \ -no-hpet \ -localtime \ -drive file=$HDD,if=virtio,cache=none,aio=native,index=0 \ -drive file=$CD,media=cdrom,if=ide,index=2 \ -net nic,vlan=0,name=net0,model=virtio,macaddr=$mac \ -net vnic,vlan=0,name=net0,ifname=$VNIC,macaddr=$mac \ -vga std \ -cpu core2duo \ -usbdevice tablet \ -balloon virtio \ -chardev socket,id=monitor,path=/tmp/VPS.monitor,server,nowait,nodelay \ -monitor chardev:monitor \ -daemonize -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Wed Oct 15 14:36:53 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 15 Oct 2014 10:36:53 -0400 Subject: [OmniOS-discuss] Crashing qemu process In-Reply-To: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2E93@AIRA-SRV.aira.local> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2E93@AIRA-SRV.aira.local> Message-ID: On Oct 15, 2014, at 5:17 AM, Filip Marvan wrote: > Hi, > > I?m doing some tests with KVM on OmniOS. I have two virtual servers, one with Windows XP, which works without any problems, but the second one with Windows 2008 R2 server is crashing after some time. Qemu proces just crashed (after about one hour after VPS starts). There is nothing strange in /var/adm/messages, and i think that ?d aption for debuging informations is not working (/tmp/qemu.log is empty even for ?d exec for example). > > Do you have any idea, how can I find what causes, that qemu proces is crashing? Do anybody havy any experiences with something like this? Any qemu core files lying around? Core dumps are the first thing to check, to see WHERE it crashed. Dan From danmcd at omniti.com Wed Oct 15 15:57:27 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 15 Oct 2014 11:57:27 -0400 Subject: [OmniOS-discuss] Please "pkg update" to get OpenSSL 1.0.1j Message-ID: Per this OpenSSL security advisory: https://www.openssl.org/news/secadv_20141015.txt Please "pkg update" your OmniOS installation to get OpenSSL 1.0.1j. We have updates for supported and development releases: r151006 (LTS) r151010 (last stable) r151012 (this stable) bloody/r151013 (development) If you are still on r151008, PLEASE UPGRADE ALREADY! :) Thank you, Dan McD. -- OmniOS Engineering From simon at nostalgeek.com Wed Oct 15 17:57:24 2014 From: simon at nostalgeek.com (Simon Boulet) Date: Wed, 15 Oct 2014 13:57:24 -0400 Subject: [OmniOS-discuss] Please "pkg update" to get OpenSSL 1.0.1j In-Reply-To: References: Message-ID: Hi Dan, On Wed, Oct 15, 2014 at 11:57 AM, Dan McDonald wrote: > Per this OpenSSL security advisory: > > https://www.openssl.org/news/secadv_20141015.txt > > Please "pkg update" your OmniOS installation to get OpenSSL 1.0.1j. We > have updates for supported and development releases: > > r151006 (LTS) > r151010 (last stable) > r151012 (this stable) > bloody/r151013 (development) > > How to make sure we are running the patched version? OmniOS version doesn't seems to contain the OpenSSL revision letter at the end. # pkg list openssl NAME (PUBLISHER) VERSION IFO library/security/openssl 1.0.1.10-0.151012 i-- Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Wed Oct 15 18:00:51 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 15 Oct 2014 14:00:51 -0400 Subject: [OmniOS-discuss] Please "pkg update" to get OpenSSL 1.0.1j In-Reply-To: References: Message-ID: On Oct 15, 2014, at 1:57 PM, Simon Boulet wrote: > > How to make sure we are running the patched version? OmniOS version doesn't seems to contain the OpenSSL revision letter at the end. > > # pkg list openssl > NAME (PUBLISHER) VERSION IFO > library/security/openssl 1.0.1.10-0.151012 i-- It does, but it's a number instead. 1.0.1j == 1.0.1.10 (j is the 10th letter). IPS doesn't let you use letters like that. Also, you can always utter: openssl version and see what it prints. You are updated, from what you showed me in your pkg list output, though. Happy Poodle Leashing! :) Dan From zmalone at omniti.com Wed Oct 15 18:03:07 2014 From: zmalone at omniti.com (Zach Malone) Date: Wed, 15 Oct 2014 14:03:07 -0400 Subject: [OmniOS-discuss] Please "pkg update" to get OpenSSL 1.0.1j In-Reply-To: References: Message-ID: Hi Simon, 1.0.1.10-0.151012 contains the revision letter at the fourth decimal place. IPS doesn't support letters in versions, so it's been mapped to a number. j is the tenth letter of the alphabet, 1.0.1.10. pkginfo openssl will return this in the human readable version field. --Zach On Wed, Oct 15, 2014 at 1:57 PM, Simon Boulet wrote: > Hi Dan, > > > On Wed, Oct 15, 2014 at 11:57 AM, Dan McDonald wrote: >> >> Per this OpenSSL security advisory: >> >> https://www.openssl.org/news/secadv_20141015.txt >> >> Please "pkg update" your OmniOS installation to get OpenSSL 1.0.1j. We >> have updates for supported and development releases: >> >> r151006 (LTS) >> r151010 (last stable) >> r151012 (this stable) >> bloody/r151013 (development) >> > > How to make sure we are running the patched version? OmniOS version doesn't > seems to contain the OpenSSL revision letter at the end. > > # pkg list openssl > NAME (PUBLISHER) VERSION > IFO > library/security/openssl 1.0.1.10-0.151012 > i-- > > > Simon > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > From eric.sproul at circonus.com Wed Oct 15 18:07:43 2014 From: eric.sproul at circonus.com (Eric Sproul) Date: Wed, 15 Oct 2014 14:07:43 -0400 Subject: [OmniOS-discuss] Please "pkg update" to get OpenSSL 1.0.1j In-Reply-To: References: Message-ID: On Wed, Oct 15, 2014 at 2:00 PM, Dan McDonald wrote: > It does, but it's a number instead. 1.0.1j == 1.0.1.10 (j is the 10th letter). IPS doesn't let you use letters like that. We preserve the upstream version string by setting the pkg.human-version attribute in the package manifest: $ pkg contents -m openssl | grep pkg.human-version set name=pkg.human-version value=1.0.1j pkg(1) displays this in parentheses following the component version when the 'info' subcommand is used: $ pkg info openssl | grep Version Version: 1.0.1.10 (1.0.1j) From simon at nostalgeek.com Wed Oct 15 18:19:46 2014 From: simon at nostalgeek.com (Simon Boulet) Date: Wed, 15 Oct 2014 14:19:46 -0400 Subject: [OmniOS-discuss] Please "pkg update" to get OpenSSL 1.0.1j In-Reply-To: References: Message-ID: Dan, Zach, Eric Excellent, thanks. Simon -------------- next part -------------- An HTML attachment was scrubbed... URL: From filip.marvan at aira.cz Wed Oct 15 18:43:49 2014 From: filip.marvan at aira.cz (Filip Marvan) Date: Wed, 15 Oct 2014 20:43:49 +0200 Subject: [OmniOS-discuss] Crashing qemu process In-Reply-To: References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2E93@AIRA-SRV.aira.local> Message-ID: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFFF376@AIRA-SRV.aira.local> Hi Dan, yes, you are right, I'm sorry. Here are more detailed informations from core dump. root at omnikvm:/# mdb core Loading modules: [ libc.so.1 ld.so.1 ] > ::status debugging core file of qemu-system-x86 (64-bit) from omnikvm file: /usr/bin/qemu-system-x86_64 initial argv: /usr/bin/qemu-system-x86_64 -d exec -name VPS -boot c -enable-kvm -vnc 0.0 threading model: native threads status: process terminated by SIGSEGV (Segmentation Fault), addr=0 > $C fffffd7fffdf8ee0 qemu_del_timer+0x10() fffffd7fffdf8f20 qemu_mod_timer+0x24() fffffd7fffdf8f60 virtio_net_handle_tx_timer+0xef() fffffd7fffdf8fa0 virtio_net_rein_tick+0x15d() fffffd7fffdf8ff0 qemu_run_timers+0xb0() fffffd7fffdf9000 qemu_run_all_timers+0x51() fffffd7fffdff090 main_loop_wait+0x480() fffffd7fffdff0c0 kvm_main_loop+0x119() fffffd7fffdff0e0 main_loop+0x17() fffffd7fffdff420 main+0x2da4() fffffd7fffdff430 _start+0x6c() And pstack core is here: http://pastebin.com/P1yfJWwh I'm using freshly installed OmniOS r151012 Thank you for any help. Filip -----Original Message----- From: Dan McDonald [mailto:danmcd at omniti.com] Sent: Wednesday, October 15, 2014 4:37 PM To: Filip Marvan Cc: omnios-discuss at lists.omniti.com Subject: Re: [OmniOS-discuss] Crashing qemu process On Oct 15, 2014, at 5:17 AM, Filip Marvan wrote: > Hi, > > I'm doing some tests with KVM on OmniOS. I have two virtual servers, one with Windows XP, which works without any problems, but the second one with Windows 2008 R2 server is crashing after some time. Qemu proces just crashed (after about one hour after VPS starts). There is nothing strange in /var/adm/messages, and i think that -d aption for debuging informations is not working (/tmp/qemu.log is empty even for -d exec for example). > > Do you have any idea, how can I find what causes, that qemu proces is crashing? Do anybody havy any experiences with something like this? Any qemu core files lying around? Core dumps are the first thing to check, to see WHERE it crashed. Dan From nsmith at careyweb.com Thu Oct 16 12:37:51 2014 From: nsmith at careyweb.com (Nate Smith) Date: Thu, 16 Oct 2014 08:37:51 -0400 Subject: [OmniOS-discuss] QLT Logging Message-ID: <66f986a0-5121-44e3-9782-5cb8184781a0@careyweb.com> I use Omnios to present LUN targets via Comstar/Fiber channel. Last night I had those targets drop and them come back up. This appears to be happening during a backup on the host servers. My question is twofold, A) how do I turn up logging to figure out what is causing this, and B) any idea what could cause this? I know it's not a switch issue. Both cards run through separate switches in a redundant mesh. I did some searching and found this thread, which only referenced the nebulous "IO storm" http://openindiana.org/pipermail/openindiana-discuss/2012-May/008211.html Thanks in advance. -Nate Smith Oct 15 22:08:47 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt3,0 LINK UP, portid 10000, topology Fabric Pt-to-Pt,speed 8G Oct 15 22:08:49 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt0,0 LINK UP, portid 20100, topology Fabric Pt-to-Pt,speed 8G Oct 15 22:08:50 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt1,0 LINK UP, portid 10100, topology Fabric Pt-to-Pt,speed 8G Oct 15 22:11:01 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt0,0 LINK UP, portid 20100, topology Fabric Pt-to-Pt,speed 8G Oct 15 22:11:01 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt3,0 LINK UP, portid 10000, topology Fabric Pt-to-Pt,speed 8G Oct 15 22:11:02 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt2,0 LINK UP, portid 20000, topology Fabric Pt-to-Pt,speed 8G Oct 15 22:12:30 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt0,0 LINK UP, portid 20100, topology Fabric Pt-to-Pt,speed 8G Oct 15 23:02:15 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt2,0 LINK UP, portid 20000, topology Fabric Pt-to-Pt,speed 8G Oct 15 23:02:16 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt3,0 LINK UP, portid 10000, topology Fabric Pt-to-Pt,speed 8G Oct 15 23:04:09 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt0,0 LINK UP, portid 20100, topology Fabric Pt-to-Pt,speed 8G Oct 15 23:04:10 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt1,0 LINK UP, portid 10100, topology Fabric Pt-to-Pt,speed 8G Oct 15 23:05:52 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt3,0 LINK UP, portid 10000, topology Fabric Pt-to-Pt,speed 8G Oct 15 23:05:53 newstorm fct: [ID 132490 kern.notice] NOTICE: qlt2,0 LINK UP, portid 20000, topology Fabric Pt-to-Pt,speed 8G -------------- next part -------------- An HTML attachment was scrubbed... URL: From filip.marvan at aira.cz Fri Oct 17 07:39:21 2014 From: filip.marvan at aira.cz (Filip Marvan) Date: Fri, 17 Oct 2014 09:39:21 +0200 Subject: [OmniOS-discuss] Crashing qemu process In-Reply-To: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFFF376@AIRA-SRV.aira.local> References: <3BE0DEED8863E5429BAE4CAEDF6245650368AFFA2E93@AIRA-SRV.aira.local> <3BE0DEED8863E5429BAE4CAEDF6245650368AFFFF376@AIRA-SRV.aira.local> Message-ID: <3BE0DEED8863E5429BAE4CAEDF62456503804DA92668@AIRA-SRV.aira.local> Just for any other who will face similar problems in future, I found, that problem was probably caused by virtio disk interface type. If I change: -drive file=$HDD,if=virtio,cache=none,aio=native,index=0 \ for -drive file=$HDD,if=ide,index=1,cache=none,aio=native,index=0 \ everything works great. Filip -----Original Message----- From: Filip Marvan Sent: Wednesday, October 15, 2014 8:44 PM To: 'Dan McDonald' Cc: omnios-discuss at lists.omniti.com Subject: RE: [OmniOS-discuss] Crashing qemu process Hi Dan, yes, you are right, I'm sorry. Here are more detailed informations from core dump. root at omnikvm:/# mdb core Loading modules: [ libc.so.1 ld.so.1 ] > ::status debugging core file of qemu-system-x86 (64-bit) from omnikvm file: /usr/bin/qemu-system-x86_64 initial argv: /usr/bin/qemu-system-x86_64 -d exec -name VPS -boot c -enable-kvm -vnc 0.0 threading model: native threads status: process terminated by SIGSEGV (Segmentation Fault), addr=0 > $C fffffd7fffdf8ee0 qemu_del_timer+0x10() fffffd7fffdf8f20 qemu_mod_timer+0x24() fffffd7fffdf8f60 virtio_net_handle_tx_timer+0xef() fffffd7fffdf8fa0 virtio_net_rein_tick+0x15d() fffffd7fffdf8ff0 qemu_run_timers+0xb0() fffffd7fffdf9000 qemu_run_all_timers+0x51() fffffd7fffdff090 main_loop_wait+0x480() fffffd7fffdff0c0 kvm_main_loop+0x119() fffffd7fffdff0e0 main_loop+0x17() fffffd7fffdff420 main+0x2da4() fffffd7fffdff430 _start+0x6c() And pstack core is here: http://pastebin.com/P1yfJWwh I'm using freshly installed OmniOS r151012 Thank you for any help. Filip -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6247 bytes Desc: not available URL: From vorob at yamalfin.ru Fri Oct 17 11:55:07 2014 From: vorob at yamalfin.ru (Yuri Vorobyev) Date: Fri, 17 Oct 2014 17:55:07 +0600 Subject: [OmniOS-discuss] ms.omniti.com resolving Message-ID: Hello. Unofficial repositories ms.omniti.com and perl.omniti.com do not resolve. They was shutdown intentionally? % host pkg.omniti.com pkg.omniti.com has address 199.15.226.221 % host ms.omniti.com Host ms.omniti.com not found: 3(NXDOMAIN) % host perl.omniti.com Host perl.omniti.com not found: 3(NXDOMAIN) From zmalone at omniti.com Fri Oct 17 14:16:48 2014 From: zmalone at omniti.com (Zach Malone) Date: Fri, 17 Oct 2014 10:16:48 -0400 Subject: [OmniOS-discuss] ms.omniti.com resolving In-Reply-To: References: Message-ID: Hi Yuri, While those look like host names, they are actually publishers. The URLs for them are http://pkg.omniti.com/omniti-ms and http://pkg.omniti.com/omniti-perl respectively. You can add them with pkg set-publisher -g http://pkg.omniti.com/omniti-ms ms.omniti.com but ms.omniti.com doesn't exist as a host. --Zach On Fri, Oct 17, 2014 at 7:55 AM, Yuri Vorobyev wrote: > Hello. > > > Unofficial repositories ms.omniti.com and perl.omniti.com do not resolve. > They was shutdown intentionally? > > % host pkg.omniti.com > pkg.omniti.com has address 199.15.226.221 > % host ms.omniti.com > Host ms.omniti.com not found: 3(NXDOMAIN) > % host perl.omniti.com > Host perl.omniti.com not found: 3(NXDOMAIN) > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From eric.sproul at circonus.com Fri Oct 17 14:20:56 2014 From: eric.sproul at circonus.com (Eric Sproul) Date: Fri, 17 Oct 2014 10:20:56 -0400 Subject: [OmniOS-discuss] ms.omniti.com resolving In-Reply-To: References: Message-ID: Also, the list of known community-provided additional repositories, as well as the official ones, is at http://omnios.omniti.com/wiki.php/Packaging From rafibeyli at gmail.com Mon Oct 20 07:01:25 2014 From: rafibeyli at gmail.com (Hafiz Rafibeyli) Date: Mon, 20 Oct 2014 10:01:25 +0300 (EEST) Subject: [OmniOS-discuss] BCM57810 bnxe driver In-Reply-To: <1860295082.284179.1413788215820.JavaMail.zimbra@cantekstil.com.tr> Message-ID: <1687578929.284277.1413788485634.JavaMail.zimbra@cantekstil.com.tr> Hello guys, have a trouble with Broadcom NetXtreme II BCM57810 10 Gigabit Ethernet on latest Omnios r151012,nic not recognized according http://wiki.openindiana.org/oi/Ethernet+Networking BCM57810 is supported on Openindiana since oi_151a7, anyone using BCM57810 on Omnios or on Openindiana? or howto install BCM57810 driver on Omnios? Regards, Hafiz. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. From richard at netbsd.org Mon Oct 20 07:42:57 2014 From: richard at netbsd.org (Richard PALO) Date: Mon, 20 Oct 2014 09:42:57 +0200 Subject: [OmniOS-discuss] strange ptime message with latest b151013 Message-ID: Funny thing after updating to latest b151013 (with libm integrated), never seen this message before when launching a nightly: > richard at omnis:/home/richard/src/illumos-gate$ ptime ./nightly.sh illumos.sh > newtask: failed to grab for process 8963: process is traced is this something to be concerned with? From danmcd at omniti.com Mon Oct 20 12:43:43 2014 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 20 Oct 2014 08:43:43 -0400 Subject: [OmniOS-discuss] BCM57810 bnxe driver In-Reply-To: <1687578929.284277.1413788485634.JavaMail.zimbra@cantekstil.com.tr> References: <1687578929.284277.1413788485634.JavaMail.zimbra@cantekstil.com.tr> Message-ID: <56BC2540-98B8-42EF-B45F-5A9461345B66@omniti.com> > On Oct 20, 2014, at 3:01 AM, Hafiz Rafibeyli wrote: > > Hello guys, > > have a trouble with Broadcom NetXtreme II BCM57810 10 Gigabit Ethernet on latest Omnios r151012,nic not recognized > > according http://wiki.openindiana.org/oi/Ethernet+Networking BCM57810 is supported on Openindiana since oi_151a7, > > anyone using BCM57810 on Omnios or on Openindiana? or howto install BCM57810 driver on Omnios? That card is currently driven by the old closed-source "bnxe" driver. It's likely your model is just too new for the old driver. There IS more modern open-source code for it, which was under review, but suffering a bit due to the lack of reviewers, and testers. Can you please utter "prtconf -v | grep pciex14e4" and share with me what you find? If it matches one of the newer /etc/driver_aliases entries for this WIP, I might be able to help. Are you willing to volunteer and test the newer open-source bits of it? This week is shaping up to be busy for me, but I MAY be able to squeeze in building it and having you try it out. Dan From danmcd at omniti.com Mon Oct 20 12:45:27 2014 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 20 Oct 2014 08:45:27 -0400 Subject: [OmniOS-discuss] strange ptime message with latest b151013 In-Reply-To: References: Message-ID: <15310372-2FC0-441D-8A0D-4D9CCDA97DEF@omniti.com> > On Oct 20, 2014, at 3:42 AM, Richard PALO wrote: > > Funny thing after updating to latest b151013 (with libm integrated), > never seen this message before when launching a nightly: > >> richard at omnis:/home/richard/src/illumos-gate$ ptime ./nightly.sh illumos.sh >> newtask: failed to grab for process 8963: process is traced > > is this something to be concerned with? Can you reproduce this with a simpler script than "nightly.sh"? And libm integration hasn't been shipped yet. Do you mean just the latest b151013? If so, I plan on updating that this week. Dan From richard at netbsd.org Mon Oct 20 12:57:24 2014 From: richard at netbsd.org (Richard PALO) Date: Mon, 20 Oct 2014 14:57:24 +0200 Subject: [OmniOS-discuss] strange ptime message with latest b151013 In-Reply-To: <15310372-2FC0-441D-8A0D-4D9CCDA97DEF@omniti.com> References: <15310372-2FC0-441D-8A0D-4D9CCDA97DEF@omniti.com> Message-ID: Le 20/10/14 14:45, Dan McDonald a ?crit : > >> On Oct 20, 2014, at 3:42 AM, Richard PALO wrote: >> >> Funny thing after updating to latest b151013 (with libm integrated), >> never seen this message before when launching a nightly: >> >>> richard at omnis:/home/richard/src/illumos-gate$ ptime ./nightly.sh illumos.sh >>> newtask: failed to grab for process 8963: process is traced >> >> is this something to be concerned with? > > Can you reproduce this with a simpler script than "nightly.sh"? > > And libm integration hasn't been shipped yet. Do you mean just the latest b151013? If so, I plan on updating that this week. > > Dan > richard at omnis:/home/richard$ uname -a SunOS omnis 5.11 omnios-f800a71 i86pc i386 i86pc From a.angelov at gmail.com Mon Oct 20 13:06:35 2014 From: a.angelov at gmail.com (Angel Angelov) Date: Mon, 20 Oct 2014 16:06:35 +0300 Subject: [OmniOS-discuss] Zpool fail to import after a system crash and reinstall Message-ID: Hello, I have experienced yesterday the issue tracked under illumos gate: https://www.illumos.org/issues/1778 After a system crash(zfs and zpool were unresponsive and sleeping when looking with truss) I needed to reboot the system which triggered it to go into system maintenance mode where I couldn't fix it and make it boot properly. Anyway that's not the issue I am reporting. Since after no success in few hours I have decided to re-install a LTS(it's what was installed so...) from scratch and import the zpool so I'd be able to send the data over to another box. Re-install finished and when tried importing the zpool I had: Assertion failed: rn->rn_nozpool == B_FALSE, file ../common/libzfs_import.c, line 1080, function zpool_open_func Abort (core dumped) The above was tested with current LTS and current Stable releases with no success. I have tried work around with re-creating the drive devices but it didn't helped. # mv /dev/rdsk /dev/old-rdsk # mv /dev/dsk /dev/old-dsk # mkdir /dev/dsk /dev/rdsk # devfsadm -c disk So after a few more hours of fighting with it I have actually managed to workaround it with installing Linux and ZFS where importing the zpool worked like a charm. I was then able to send the data over to another box(illumos) without any issues. So another successful night after a crash but still :) Thanks, A.A. -------------- next part -------------- An HTML attachment was scrubbed... URL: From sim.ple at live.nl Mon Oct 20 13:36:00 2014 From: sim.ple at live.nl (Randy S) Date: Mon, 20 Oct 2014 15:36:00 +0200 Subject: [OmniOS-discuss] Dell R730xd Message-ID: Hi all, Can anyone please tell me if I should expect any problems installing omnios R10 on a new Dell R730xd Thinking about chipsets, nic drivers etc (all standard on the dell) PowerEdge R730xd Specifications: (from site) Form factor: 2U rackProcessor: Intel Xeon processor E5-2600 v3 product familyProcessor sockets: 2Dimensions: H: 8.73cm (3.44in.) x W: 44.40cm (17.49in.) x D: 68.40cm (26.92in.)Cache: 2.5MB per core; core options: 4, 6, 8, 10, 12, 14, 16, 18Chipset: Intel C610 series chipsetMemory: Up to 768GB (24 DIMM slots): 4GB/8GB/16GB/32GB DDR4 up to 2133MT/sI/O slots: Up to 6 x PCIe 3.0 slots plus dedicated RAID card slotDrive bays: Internal hard drive bay and hot-plug backplane:Up to 16 x 3.5? SAS, SATA, nearline SAS, SSD + 2 x 2.5? drivesUp to 18 x 1.8? SAS, SATA, nearline SAS, SSD drives + 8 x 3.5? SAS, SATA, nearline SAS, SSD drives, + 2 x 2.5? HDDUp to 26 x 2.5? SAS, SATA, nearline SAS, SSD, PCIe SSD drives Rgds, Randy -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Mon Oct 20 14:53:48 2014 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 20 Oct 2014 10:53:48 -0400 Subject: [OmniOS-discuss] Zpool fail to import after a system crash and reinstall In-Reply-To: References: Message-ID: <17B3FA9A-A46C-4B4D-B39E-EE0A43B68904@omniti.com> Pinging the ZFS developers about this. If ZoL works, then this may be a fixed-there-but-not-yet-upstreamed bugfix. Dan From danmcd at omniti.com Mon Oct 20 15:15:19 2014 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 20 Oct 2014 11:15:19 -0400 Subject: [OmniOS-discuss] Dell R730xd In-Reply-To: References: Message-ID: > On Oct 20, 2014, at 9:36 AM, Randy S wrote: > > Hi all, > > Can anyone please tell me if I should expect any problems installing omnios R10 on a new Dell R730xd Possibly, but let's take a look... > Thinking about chipsets, nic drivers etc (all standard on the dell) > PowerEdge R730xd Specifications: (from site) > > ? Form factor: 2U rack > ? Processor: Intel Xeon processor E5-2600 v3 product family > ? Processor sockets: 2 > ? Dimensions: H: 8.73cm (3.44in.) x W: 44.40cm (17.49in.) x D: 68.40cm (26.92in.) > ? Cache: 2.5MB per core; core options: 4, 6, 8, 10, 12, 14, 16, 18 > ? Chipset: Intel C610 series chipset > ? Memory: Up to 768GB (24 DIMM slots): 4GB/8GB/16GB/32GB DDR4 up to 2133MT/s > ? I/O slots: Up to 6 x PCIe 3.0 slots plus dedicated RAID card slot > ? Drive bays: > ? Internal hard drive bay and hot-plug backplane: > ? Up to 16 x 3.5? SAS, SATA, nearline SAS, SSD + 2 x 2.5? drives > ? Up to 18 x 1.8? SAS, SATA, nearline SAS, SSD drives + 8 x 3.5? SAS, SATA, nearline SAS, SSD drives, + 2 x 2.5? HDD > ? Up to 26 x 2.5? SAS, SATA, nearline SAS, SSD, PCIe SSD drives How are those SAS and SATA slots being driven?! The C610 chipset has what's know as a Storage Controller Unit (SCU), which is not supported by any illumos distribution currently. SAS won't work at all, and SATA will only work if it presents as AHCI SATA. If there's a Dell-supplied raid controller, I'll need to know which model it is, AND what chipset drives is. There may be hope here, but Dell likes HW-RAID, which we generally frown up on in the illumos & ZFS communities. The H330 looks like an LSI 3008 with "skinny" firmware on it, and the H730 looks like an LSI 3108 with "skinny" firmware on it. Generally speaking, both of those controllers are HW-RAID. The H330 (like its predecessor the H310) can do passthrough/raw-disk, but I believe it will be driven by the "mr_sas" driver, which has not had a lot of performance work done upon it. Oooh, I do see, however, that the H730 now has raw-disk mode, unlike its predecessor the H710. Still, same warning about the mr_sas driver, however. Generally speaking, I'd try SuperMicro HW over Dell, but if your environment is insistent upon a Dell, I'd recommend: * Intel NICs if possible (though Broadcom 10GigE support is in the midst of being improved on illumos, and Mellanox has had some work as well). * Query the *illumos* community about the H330 or the H730 controllers. Maybe Nexenta (a Dell partner) has done additional work on these, so they may have something to say there. Hope this helps, Dan From hakansom at ohsu.edu Mon Oct 20 15:58:20 2014 From: hakansom at ohsu.edu (Marion Hakanson) Date: Mon, 20 Oct 2014 15:58:20 +0000 Subject: [OmniOS-discuss] Dell R730xd In-Reply-To: References: , Message-ID: <8D3E3DD7-EA48-4693-93DE-374C3E67BEFB@ohsu.edu> I don't know if it's true for the R730xd, but Dell will equip your R720xd with an LSI-branded HBA for the internal slots. It's not on their web configurator, you have to ask your salesperson directly. I've also bought LSI external HBA's from Dell. Regards, Marion Sent from somebody's iPad > On Oct 20, 2014, at 8:16 AM, "Dan McDonald" wrote: > > >> On Oct 20, 2014, at 9:36 AM, Randy S wrote: >> >> Hi all, >> >> Can anyone please tell me if I should expect any problems installing omnios R10 on a new Dell R730xd > > Possibly, but let's take a look... > >> Thinking about chipsets, nic drivers etc (all standard on the dell) >> PowerEdge R730xd Specifications: (from site) >> >> ? Form factor: 2U rack >> ? Processor: Intel Xeon processor E5-2600 v3 product family >> ? Processor sockets: 2 >> ? Dimensions: H: 8.73cm (3.44in.) x W: 44.40cm (17.49in.) x D: 68.40cm (26.92in.) >> ? Cache: 2.5MB per core; core options: 4, 6, 8, 10, 12, 14, 16, 18 >> ? Chipset: Intel C610 series chipset >> ? Memory: Up to 768GB (24 DIMM slots): 4GB/8GB/16GB/32GB DDR4 up to 2133MT/s >> ? I/O slots: Up to 6 x PCIe 3.0 slots plus dedicated RAID card slot >> ? Drive bays: >> ? Internal hard drive bay and hot-plug backplane: >> ? Up to 16 x 3.5? SAS, SATA, nearline SAS, SSD + 2 x 2.5? drives >> ? Up to 18 x 1.8? SAS, SATA, nearline SAS, SSD drives + 8 x 3.5? SAS, SATA, nearline SAS, SSD drives, + 2 x 2.5? HDD >> ? Up to 26 x 2.5? SAS, SATA, nearline SAS, SSD, PCIe SSD drives > > How are those SAS and SATA slots being driven?! > > The C610 chipset has what's know as a Storage Controller Unit (SCU), which is not supported by any illumos distribution currently. SAS won't work at all, and SATA will only work if it presents as AHCI SATA. > > If there's a Dell-supplied raid controller, I'll need to know which model it is, AND what chipset drives is. There may be hope here, but Dell likes HW-RAID, which we generally frown up on in the illumos & ZFS communities. The H330 looks like an LSI 3008 with "skinny" firmware on it, and the H730 looks like an LSI 3108 with "skinny" firmware on it. Generally speaking, both of those controllers are HW-RAID. The H330 (like its predecessor the H310) can do passthrough/raw-disk, but I believe it will be driven by the "mr_sas" driver, which has not had a lot of performance work done upon it. Oooh, I do see, however, that the H730 now has raw-disk mode, unlike its predecessor the H710. Still, same warning about the mr_sas driver, however. > > Generally speaking, I'd try SuperMicro HW over Dell, but if your environment is insistent upon a Dell, I'd recommend: > > * Intel NICs if possible (though Broadcom 10GigE support is in the midst of being improved on illumos, and Mellanox has had some work as well). > * Query the *illumos* community about the H330 or the H730 controllers. Maybe Nexenta (a Dell partner) has done additional work on these, so they may have something to say there. > > Hope this helps, > Dan > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From chip at innovates.com Mon Oct 20 20:26:31 2014 From: chip at innovates.com (Schweiss, Chip) Date: Mon, 20 Oct 2014 15:26:31 -0500 Subject: [OmniOS-discuss] Dell R730xd In-Reply-To: <8D3E3DD7-EA48-4693-93DE-374C3E67BEFB@ohsu.edu> References: <8D3E3DD7-EA48-4693-93DE-374C3E67BEFB@ohsu.edu> Message-ID: Recently we had a quote from Dell w/ Nexenta. They specify Intel NICs and LSI HBAs when building with Nexenta. -Chip On Mon, Oct 20, 2014 at 10:58 AM, Marion Hakanson wrote: > I don't know if it's true for the R730xd, but Dell will equip your R720xd > with an LSI-branded HBA for the internal slots. It's not on their web > configurator, you have to ask your salesperson directly. I've also bought > LSI external HBA's from Dell. > > Regards, > > Marion > > > Sent from somebody's iPad > > > On Oct 20, 2014, at 8:16 AM, "Dan McDonald" wrote: > > > > > >> On Oct 20, 2014, at 9:36 AM, Randy S wrote: > >> > >> Hi all, > >> > >> Can anyone please tell me if I should expect any problems installing > omnios R10 on a new Dell R730xd > > > > Possibly, but let's take a look... > > > >> Thinking about chipsets, nic drivers etc (all standard on the dell) > >> PowerEdge R730xd Specifications: (from site) > >> > >> ? Form factor: 2U rack > >> ? Processor: Intel Xeon processor E5-2600 v3 product family > >> ? Processor sockets: 2 > >> ? Dimensions: H: 8.73cm (3.44in.) x W: 44.40cm (17.49in.) x D: > 68.40cm (26.92in.) > >> ? Cache: 2.5MB per core; core options: 4, 6, 8, 10, 12, 14, 16, 18 > >> ? Chipset: Intel C610 series chipset > >> ? Memory: Up to 768GB (24 DIMM slots): 4GB/8GB/16GB/32GB DDR4 up to > 2133MT/s > >> ? I/O slots: Up to 6 x PCIe 3.0 slots plus dedicated RAID card slot > >> ? Drive bays: > >> ? Internal hard drive bay and hot-plug backplane: > >> ? Up to 16 x 3.5? SAS, SATA, nearline SAS, SSD + 2 x 2.5? drives > >> ? Up to 18 x 1.8? SAS, SATA, nearline SAS, SSD drives + 8 x 3.5? > SAS, SATA, nearline SAS, SSD drives, + 2 x 2.5? HDD > >> ? Up to 26 x 2.5? SAS, SATA, nearline SAS, SSD, PCIe SSD drives > > > > How are those SAS and SATA slots being driven?! > > > > The C610 chipset has what's know as a Storage Controller Unit (SCU), > which is not supported by any illumos distribution currently. SAS won't > work at all, and SATA will only work if it presents as AHCI SATA. > > > > If there's a Dell-supplied raid controller, I'll need to know which > model it is, AND what chipset drives is. There may be hope here, but Dell > likes HW-RAID, which we generally frown up on in the illumos & ZFS > communities. The H330 looks like an LSI 3008 with "skinny" firmware on it, > and the H730 looks like an LSI 3108 with "skinny" firmware on it. > Generally speaking, both of those controllers are HW-RAID. The H330 (like > its predecessor the H310) can do passthrough/raw-disk, but I believe it > will be driven by the "mr_sas" driver, which has not had a lot of > performance work done upon it. Oooh, I do see, however, that the H730 now > has raw-disk mode, unlike its predecessor the H710. Still, same warning > about the mr_sas driver, however. > > > > Generally speaking, I'd try SuperMicro HW over Dell, but if your > environment is insistent upon a Dell, I'd recommend: > > > > * Intel NICs if possible (though Broadcom 10GigE support is in the midst > of being improved on illumos, and Mellanox has had some work as well). > > * Query the *illumos* community about the H330 or the H730 controllers. > Maybe Nexenta (a Dell partner) has done additional work on these, so they > may have something to say there. > > > > Hope this helps, > > Dan > > > > _______________________________________________ > > OmniOS-discuss mailing list > > OmniOS-discuss at lists.omniti.com > > http://lists.omniti.com/mailman/listinfo/omnios-discuss > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Mon Oct 20 20:27:47 2014 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 20 Oct 2014 16:27:47 -0400 Subject: [OmniOS-discuss] Dell R730xd In-Reply-To: References: <8D3E3DD7-EA48-4693-93DE-374C3E67BEFB@ohsu.edu> Message-ID: <3F65C05E-30CA-4916-8304-578B0F927F0D@omniti.com> > On Oct 20, 2014, at 4:26 PM, Schweiss, Chip wrote: > > Recently we had a quote from Dell w/ Nexenta. They specify Intel NICs and LSI HBAs when building with Nexenta. Any HW setup that works with Nexenta should also work with OmniOS. Dan From sim.ple at live.nl Mon Oct 20 21:44:46 2014 From: sim.ple at live.nl (Randy S) Date: Mon, 20 Oct 2014 23:44:46 +0200 Subject: [OmniOS-discuss] Dell R730xd In-Reply-To: <3F65C05E-30CA-4916-8304-578B0F927F0D@omniti.com> References: , , <8D3E3DD7-EA48-4693-93DE-374C3E67BEFB@ohsu.edu>, , <3F65C05E-30CA-4916-8304-578B0F927F0D@omniti.com> Message-ID: Thanks for the input guys! I'll certainly be sure to have it assembled with intel nics and LSI hba's. If there is any interrest I can post results of preliminary tests here. It will be some weeks though before I receive the stuff (if all goes well). Regards, Randy > From: danmcd at omniti.com > Date: Mon, 20 Oct 2014 16:27:47 -0400 > To: chip at innovates.com > CC: omnios-discuss at lists.omniti.com > Subject: Re: [OmniOS-discuss] Dell R730xd > > > > On Oct 20, 2014, at 4:26 PM, Schweiss, Chip wrote: > > > > Recently we had a quote from Dell w/ Nexenta. They specify Intel NICs and LSI HBAs when building with Nexenta. > > Any HW setup that works with Nexenta should also work with OmniOS. > > Dan > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From contact at jacobvosmaer.nl Mon Oct 20 22:53:11 2014 From: contact at jacobvosmaer.nl (Jacob Vosmaer) Date: Tue, 21 Oct 2014 00:53:11 +0200 Subject: [OmniOS-discuss] r151012 nlockmgr fails to start In-Reply-To: References: <6824C2D7-9C1E-4E81-9C20-D4B08A394EC4@omniti.com> <5432E672.2080805@ColoState.EDU> <5949A3B6-E21F-4FC5-9F31-0AAA91EFCBA2@omniti.com> Message-ID: I appear to be having the same problem on a HP MicroServer N54L. Messages: /sbin/dhcpagent[95]: [ID 778557 daemon.warning] configure_v4_lease: no IP broadcast specified for bge0, making best guess rootnex: [ID 349649 kern.info] iscsi0 at root genunix: [ID 936769 kern.info] iscsi0 is /iscsi pseudo: [ID 129642 kern.info] pseudo-device: dtrace0 genunix: [ID 936769 kern.info] dtrace0 is /pseudo/dtrace at 0 klmmod: [ID 814159 kern.notice] NOTICE: Failed to connect to local statd (rpcerr=5) /usr/lib/nfs/lockd[473]: [ID 491006 daemon.error] Cannot establish NLM service over : I/O error. Exiting svc.startd[10]: [ID 652011 daemon.warning] svc:/network/nfs/nlockmgr:default: Method "/lib/svc/method/nlockmgr" failed with exit status 1. klmmod: [ID 814159 kern.notice] NOTICE: Failed to connect to local statd (rpcerr=5) /usr/lib/nfs/lockd[534]: [ID 491006 daemon.error] Cannot establish NLM service over : I/O error. Exiting svc.startd[10]: [ID 652011 daemon.warning] svc:/network/nfs/nlockmgr:default: Method "/lib/svc/method/nlockmgr" failed with exit status 1. pseudo: [ID 129642 kern.info] pseudo-device: pool0 genunix: [ID 936769 kern.info] pool0 is /pseudo/pool at 0 klmmod: [ID 814159 kern.notice] NOTICE: Failed to connect to local statd (rpcerr=5) /usr/lib/nfs/lockd[537]: [ID 491006 daemon.error] Cannot establish NLM service over : I/O error. Exiting svc.startd[10]: [ID 652011 daemon.warning] svc:/network/nfs/nlockmgr:default: Method "/lib/svc/method/nlockmgr" failed with exit status 1. svc.startd[10]: [ID 748625 daemon.error] network/nfs/nlockmgr:default failed: transitioned to maintenance (see 'svcs -xv' for details) Rebooting sometimes makes the nlockmgr problem go away, but sometimes it does not. If there is anything I can do diagnostics-wise let me know. I have no clue. 2014-10-10 21:09 GMT+02:00 Schweiss, Chip : > Apparently something common in my OmniOS setup is triggering this. I > have no idea what yet, and I'm feeling green at digging through this > issue. > > On one of my VMs for doing script development, I exported the data pool > planing to test importing it with a different cache location and the > problem immediately happened. Now I cannot get nlockmgr to start at all > on this VM. I tried disabling all nfs services and re-enabling. Still > failing with /usr/lib/nfs/lockd[862]: [ID 491006 daemon.error] Cannot > establish NLM service over : I/O error. Exiting > > root at ZFSsendTest1:/root# svcs -a|grep nfs > disabled 13:47:05 svc:/network/nfs/log:default > disabled 13:47:11 svc:/network/nfs/rquota:default > disabled 13:55:05 svc:/network/nfs/server:default > disabled 13:55:32 svc:/network/nfs/nlockmgr:default > disabled 13:55:32 svc:/network/nfs/mapid:default > disabled 13:55:32 svc:/network/nfs/status:default > disabled 13:55:32 svc:/network/nfs/client:default > disabled 13:55:57 svc:/network/nfs/cbd:default > root at ZFSsendTest1:/root# svcadm enable svc:/network/nfs/status:default > svc:/network/nfs/cbd:default svc:/network/nfs/mapid:default > svc:/network/nfs/server:default svc:/network/nfs/nlockmgr:default > root at ZFSsendTest1:/root# svcs -a|grep nfs > disabled 13:47:05 svc:/network/nfs/log:default > disabled 13:47:11 svc:/network/nfs/rquota:default > disabled 13:55:32 svc:/network/nfs/client:default > online 13:56:56 svc:/network/nfs/status:default > online 13:56:56 svc:/network/nfs/cbd:default > online 13:56:56 svc:/network/nfs/mapid:default > offline 13:56:56 svc:/network/nfs/server:default > offline* 13:56:56 svc:/network/nfs/nlockmgr:default > root at ZFSsendTest1:/root# svcs -a|grep nfs > disabled 13:47:05 svc:/network/nfs/log:default > disabled 13:47:11 svc:/network/nfs/rquota:default > disabled 13:55:32 svc:/network/nfs/client:default > online 13:56:56 svc:/network/nfs/status:default > online 13:56:56 svc:/network/nfs/cbd:default > online 13:56:56 svc:/network/nfs/mapid:default > offline 13:56:56 svc:/network/nfs/server:default > maintenance 13:58:11 svc:/network/nfs/nlockmgr:default > > This VM has never had RSF-1 on it, so that definitely isn't the trigger. > This VM has never exhibited this problem before today. It has been > rebooted many times. > > I wonder if the problem is triggered by exporting a pool with NFS exports > that have active client connections. That is always the case on my > production systems. This VM has one NFS client that was connected when I > exported the pool. > > Now nlockmgr dies and goes to maintenance mode regardless if I import the > data pool or not. > > Any advice on where to dig for better diagnosis of this would be > helpful. If any developers would like to get access to this VM I'd be > happy to arrange that too. > > -Chip > > > On Fri, Oct 10, 2014 at 9:26 AM, Richard Elling < > richard.elling at richardelling.com> wrote: > >> >> On Oct 10, 2014, at 6:15 AM, "Schweiss, Chip" wrote: >> >> >> On Thu, Oct 9, 2014 at 9:54 PM, Dan McDonald wrote: >> >>> >>> On Oct 9, 2014, at 10:23 PM, Schweiss, Chip wrote: >>> >>> > Just tried my 2nd system. r151010 nlockmgr starts after clearing >>> maintenance mode. r151012 it will not start at all. nfs/status was >>> enabled and online. >>> > >>> > The commonality I see on the two systems I have tried is they are both >>> part of an HA cluster. So they don't import the pool at boot, but RSF-1 >>> imports it with cache mapped to a different location. >>> >>> That could be something HA Inc. needs to further test. We don't >>> directly support RSF-1, after all. >>> >>> >> I there really isn't anything different than an auto imported pool. I'm >> suspecting using an alternate cache location my be triggering something >> else to go wrong in the nlockmgr. >> >> >> no, these are totally separate subsystems. RSF-1 imports the pool. NFS >> sharing is started by the zpool command, in userland, via sharing after the >> dataset is mounted. You can do the same procedure manually... no magic >> pixie dust needed. >> >> >> Here's the command RSF-1 uses to import the pool: >> zpool import -c /opt/HAC/RSF-1/etc/volume-cache/nrgpool.cache -o >> cachefile=/opt/HAC/RSF-1/etc/v >> olume-cache/nrgpool.cache-live -o failmode=panic nrgpool >> >> After the pool import it puts the ip addresses back and is done. That >> happens in less than 1 second. >> >> In the mean time NFS services auto start and nlockmgr starts spinning. >> >> >> Perhaps share doesn't properly start all of the services? Does it work ok >> if you manually "svcadm enable" all of the NFS services? >> >> >> -- richard >> >> >> >> >>> > nlockmgr is becoming a real show stopper. >>> >>> svcadm disable nlockmgr nfs/status >>> svcadm enable nfs/status >>> svcadm enable nlockmgr >>> >>> You may wish to discuss this on illumos as well, I'm not sure who all >>> else is seeing this save me one time, and you seemingly a lot of times. >>> >> >> I did that this time, no joy. Today I'm working on a virtual setup with >> HA to see if I can get this reproduced on r151012. >> >> I thought this nlockmgr propblem was related to lots of nfs exports >> until, I ran into this on my SSD pool. It used to be able to fail over in >> about 3-5 seconds. It takes nlockmgr now sits in a spinning state for a >> few minutes and fails every time. A clear of the maintenance mode, brings >> it back nearly instantly. This is on r151010. On r151012 it fails every >> time. >> >> Hopefully I can reproduce and I'll start a new thread copying Illumos too. >> >> -Chip >> >> >> _______________________________________________ >> OmniOS-discuss mailing list >> OmniOS-discuss at lists.omniti.com >> http://lists.omniti.com/mailman/listinfo/omnios-discuss >> >> > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Tue Oct 21 05:17:15 2014 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 21 Oct 2014 01:17:15 -0400 Subject: [OmniOS-discuss] OmniOS "bloody" now updated. Message-ID: <3736D518-8A4F-41E7-9B5A-4B8BF92A0699@omniti.com> Pardon the delay. I was waiting for one of the new features to integrate upstream before I updated bloody. So this one is not-quite three weeks later. I plan on resuming biweekly updates of bloody on a Thursday +/- 1 day of an update week. That means the next bloody update after this SHOULD be out in early November. I did a full sweeping package update, so you may need to "pkg update pkg" prior to this update. New with this release (r151013-20141021): - omnios-build master branch, revision 2412b2d. - open-vm-tools from Delphix - illumos-omnios master branch, revision 09bdbed. - Open-source math library. (This caused the delay. NOTE: library/math/header-math is now gone. Any dependencies upon that should be replaced with library/math.) - pam_timestamp option for PAM (to allow sudo-like last-login caching, handle with care). - Some ZFS fixes and improvments. - Replaceable sendmail packages (uses IPS mediators). - sTec S112x PCIe SSD support. I've also cut new install media as well, AND there's a new kayak .zfs.bz2 in place. Thanks! Dan From eric.sproul at circonus.com Tue Oct 21 18:55:41 2014 From: eric.sproul at circonus.com (Eric Sproul) Date: Tue, 21 Oct 2014 14:55:41 -0400 Subject: [OmniOS-discuss] Serving multiple OmniOS releases from one Kayak server Message-ID: Hi, Just wanted to share something I came up with a couple months ago: http://omnios.omniti.com/wiki.php/KayakMultiRelease At Circonus we run LTS on our infrastructure, but need to install machines at different releases for build automation and testing. I set up our Kayak server to serve LTS, current stable and bloody, allowing me to choose at PXE-boot time which release to install. It works well, but does require manually moving the kayak-kernel files around and editing the GRUB config, though someone clever could probably figure out a way to make it easier. I'm not sufficiently motivated to do that right now. ;) Enjoy, Eric From sjorge+ml at blackdot.be Tue Oct 21 18:59:53 2014 From: sjorge+ml at blackdot.be (Jorge Schrauwen) Date: Tue, 21 Oct 2014 20:59:53 +0200 Subject: [OmniOS-discuss] Serving multiple OmniOS releases from one Kayak server In-Reply-To: References: Message-ID: <34c0ad6429ddac3d7f4aefbc947bee73@blackdot.be> For those interested, I worked around it by using iPXE, I just load different kernel + bootarchive for each release. Although it is a bit flaky sometimes. On 2014-10-21 20:55, Eric Sproul wrote: > Hi, > Just wanted to share something I came up with a couple months ago: > http://omnios.omniti.com/wiki.php/KayakMultiRelease > > At Circonus we run LTS on our infrastructure, but need to install > machines at different releases for build automation and testing. I > set up our Kayak server to serve LTS, current stable and bloody, > allowing me to choose at PXE-boot time which release to install. It > works well, but does require manually moving the kayak-kernel files > around and editing the GRUB config, though someone clever could > probably figure out a way to make it easier. I'm not sufficiently > motivated to do that right now. ;) > > Enjoy, > Eric > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From hasslerd at gmx.li Tue Oct 21 19:25:08 2014 From: hasslerd at gmx.li (Dominik Hassler) Date: Tue, 21 Oct 2014 21:25:08 +0200 Subject: [OmniOS-discuss] HP Z230 as a home server Message-ID: <5446B314.9020600@gmx.li> hi, it's about a year since i got in touch with illumos/solaris and zfs for the first time. i absolutely don't want to miss it anymore, especially zfs. omniti/dan and the whole illumos community are doing a great job. thank you! of course i'd like to have an omnios server at home, too. nothing fancy, basically just for data storage (nfs and smb sharing) and 1-2 linux kvms for multimedia services and auxiliary stuff. since i don't have a dedicated room where noise does not matter i have to consider quiet "workstation hardware" instead of noisy "server hardware". here is the hardware setup i am planning to go with: - HP Z230 (WM583EA#UUZ) which is basically -> Intel PCH C226 chipset -> E3-1225v3 quad core XEON -> Intel I217LM PCIe Gigabit-Controller - 4x 8 GB ECC RAM (HP A2Z50AA) - 1x 1 TB WD Re, S-ATA III (512n) as rpool - 2x 4 TB WD Re, S-ATA III (512n) as a data mirror what do you think about this hardware setup? any no-gos? thanks dominik -------------- next part -------------- A non-text attachment was scrubbed... Name: 0xF9ECC6A5.asc Type: application/pgp-keys Size: 2686 bytes Desc: not available URL: From danmcd at omniti.com Tue Oct 21 20:24:51 2014 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 21 Oct 2014 16:24:51 -0400 Subject: [OmniOS-discuss] HP Z230 as a home server In-Reply-To: <5446B314.9020600@gmx.li> References: <5446B314.9020600@gmx.li> Message-ID: > On Oct 21, 2014, at 3:25 PM, Dominik Hassler wrote: > of course i'd like to have an omnios server at home, too. nothing fancy, > basically just for data storage (nfs and smb sharing) and 1-2 linux kvms > for multimedia services and auxiliary stuff. since i don't have a > dedicated room where noise does not matter i have to consider quiet > "workstation hardware" instead of noisy "server hardware". > > here is the hardware setup i am planning to go with: > - HP Z230 (WM583EA#UUZ) which is basically http://www8.hp.com/us/en/campaigns/workstations/z230.html Interesting. It's a socket 1150 workstation. Good for illumos. > -> Intel PCH C226 chipset > -> E3-1225v3 quad core XEON Gotta get that ECC support. Also, while not HT, it does have good clock speed (> 3GHz). > -> Intel I217LM PCIe Gigabit-Controller Supported, and relatively new. Should show up as e1000g. I notice you can get a 2nd ethernet with an I210. This would show up as igb. > - 4x 8 GB ECC RAM (HP A2Z50AA) Gotta do ECC! > - 1x 1 TB WD Re, S-ATA III (512n) as rpool > - 2x 4 TB WD Re, S-ATA III (512n) as a data mirror I might swap in an SSD for rpool, but that's just me. (NOTE: If you do use an SSD, don't use it for swap, though with 32GB, you won't need swap at all). And for your data, it doesn't matter (I don't think) if it's 512 or 4k sectors, just so long as you know what it is up front, they MATCH, and set things appropriately when creating the pool. If others here have opinions, you ought to share 'em. I'm running a Supermicro not unlike what the HP offers: http://kebesays.blogspot.com/2014/06/home-data-center-20-dogfooding-again.html I got more NICs though. :) Thanks, and hope this helps, Dan From hasslerd at gmx.li Tue Oct 21 21:01:03 2014 From: hasslerd at gmx.li (Dominik Hassler) Date: Tue, 21 Oct 2014 23:01:03 +0200 Subject: [OmniOS-discuss] HP Z230 as a home server In-Reply-To: References: <5446B314.9020600@gmx.li> Message-ID: <5446C98F.9070609@gmx.li> dan, i will definitely go for the 2 nic option; did overlook that somehow... and take an intel S3500 SSD as rpool (non mirrored). thanks again dominik On 10/21/2014 10:24 PM, Dan McDonald wrote: > >> On Oct 21, 2014, at 3:25 PM, Dominik Hassler wrote: > > > >> of course i'd like to have an omnios server at home, too. nothing fancy, >> basically just for data storage (nfs and smb sharing) and 1-2 linux kvms >> for multimedia services and auxiliary stuff. since i don't have a >> dedicated room where noise does not matter i have to consider quiet >> "workstation hardware" instead of noisy "server hardware". >> >> here is the hardware setup i am planning to go with: >> - HP Z230 (WM583EA#UUZ) which is basically > > http://www8.hp.com/us/en/campaigns/workstations/z230.html > > Interesting. It's a socket 1150 workstation. Good for illumos. > >> -> Intel PCH C226 chipset >> -> E3-1225v3 quad core XEON > > Gotta get that ECC support. Also, while not HT, it does have good clock speed (> 3GHz). > >> -> Intel I217LM PCIe Gigabit-Controller > > Supported, and relatively new. Should show up as e1000g. I notice you can get a 2nd ethernet with an I210. This would show up as igb. > >> - 4x 8 GB ECC RAM (HP A2Z50AA) > > Gotta do ECC! > >> - 1x 1 TB WD Re, S-ATA III (512n) as rpool >> - 2x 4 TB WD Re, S-ATA III (512n) as a data mirror > > I might swap in an SSD for rpool, but that's just me. (NOTE: If you do use an SSD, don't use it for swap, though with 32GB, you won't need swap at all). And for your data, it doesn't matter (I don't think) if it's 512 or 4k sectors, just so long as you know what it is up front, they MATCH, and set things appropriately when creating the pool. > > If others here have opinions, you ought to share 'em. I'm running a Supermicro not unlike what the HP offers: > > http://kebesays.blogspot.com/2014/06/home-data-center-20-dogfooding-again.html > > I got more NICs though. :) > > Thanks, and hope this helps, > Dan > -------------- next part -------------- A non-text attachment was scrubbed... Name: 0xF9ECC6A5.asc Type: application/pgp-keys Size: 2686 bytes Desc: not available URL: From rme at clozure.com Tue Oct 21 21:09:52 2014 From: rme at clozure.com (R. Matthew Emerson) Date: Tue, 21 Oct 2014 17:09:52 -0400 Subject: [OmniOS-discuss] HP Z230 as a home server In-Reply-To: References: <5446B314.9020600@gmx.li> Message-ID: <52DCA0DF-14A9-44F8-8B62-E29749D3645D@clozure.com> On Oct 21, 2014, at 4:24 PM, Dan McDonald wrote: > >> On Oct 21, 2014, at 3:25 PM, Dominik Hassler wrote: > > > >> of course i'd like to have an omnios server at home, too. nothing fancy, >> basically just for data storage (nfs and smb sharing) and 1-2 linux kvms >> for multimedia services and auxiliary stuff. since i don't have a >> dedicated room where noise does not matter i have to consider quiet >> "workstation hardware" instead of noisy "server hardware". >> >> here is the hardware setup i am planning to go with: >> - HP Z230 (WM583EA#UUZ) which is basically > > http://www8.hp.com/us/en/campaigns/workstations/z230.html > > Interesting. It's a socket 1150 workstation. Good for illumos. > >> -> Intel PCH C226 chipset >> -> E3-1225v3 quad core XEON > > Gotta get that ECC support. Also, while not HT, it does have good clock speed (> 3GHz). > >> -> Intel I217LM PCIe Gigabit-Controller > > Supported, and relatively new. Should show up as e1000g. I notice you can get a 2nd ethernet with an I210. This would show up as igb. > >> - 4x 8 GB ECC RAM (HP A2Z50AA) > > Gotta do ECC! > >> - 1x 1 TB WD Re, S-ATA III (512n) as rpool >> - 2x 4 TB WD Re, S-ATA III (512n) as a data mirror > > I might swap in an SSD for rpool, but that's just me. (NOTE: If you do use an SSD, don't use it for swap, though with 32GB, you won't need swap at all). And for your data, it doesn't matter (I don't think) if it's 512 or 4k sectors, just so long as you know what it is up front, they MATCH, and set things appropriately when creating the pool. > > If others here have opinions, you ought to share 'em. I'm running a Supermicro not unlike what the HP offers: > > http://kebesays.blogspot.com/2014/06/home-data-center-20-dogfooding-again.html > > I got more NICs though. :) My small home server is a Supermicro X10SLM-F-O with a Xeon 1230v3 in a Supermicro CSE-731i-300B case. The board is nice because it has two NICs (one e1000g, one igb), and also a dedicated one for IPMI. I used memory from Crucial (CT2KIT102472BD160B) and a pair of Toshiba 2 TB disks (PH3200U-1I72) as my data pool (on sale at the local Micro Center). I'm using a 500GB Hitachi disk that I happened to have around for my rpool. This ended up being around $900, which I thought was way cheap. If I did it again, I might consider the larger 732 series case, which is supposed to be a little bit quieter. My desktop box is a pretty quiet 27" iMac, and I can hear the Supermicro box's fans from about 5 feet away. The noise not really objectionable, but it's not silent either. The only problem I had with this hardware was that fastboot would hang, but the suggestion in http://article.gmane.org/gmane.os.omnios.general/1043 solved that. I would have gone with a factory-built box, but Dell wasn't selling Haswell Xeon E3 systems at the time, and HP was trying hard to discourage customers from using non-HP disks. From brogyi at gmail.com Tue Oct 21 22:03:14 2014 From: brogyi at gmail.com (=?ISO-8859-1?Q?Brogy=E1nyi_J=F3zsef?=) Date: Wed, 22 Oct 2014 00:03:14 +0200 Subject: [OmniOS-discuss] HP Z230 as a home server In-Reply-To: <5446B314.9020600@gmx.li> References: <5446B314.9020600@gmx.li> Message-ID: <5446D822.4060207@gmail.com> Hi - I definitely choose E3-1230v3 CPU. - Not need any VGA and better CPU. - 2x WD7500BFCX mirror rpool - I'm sure if I were you I sliced the HDD. 100GB is enough for rpool. Some guys use only 32GB for rpool. The rest part of HDD is for your data. Brogyi 2014.10.21. 21:25 keltez?ssel, Dominik Hassler ?rta: > hi, > > it's about a year since i got in touch with illumos/solaris and zfs for > the first time. i absolutely don't want to miss it anymore, especially > zfs. omniti/dan and the whole illumos community are doing a great job. > thank you! > > > of course i'd like to have an omnios server at home, too. nothing fancy, > basically just for data storage (nfs and smb sharing) and 1-2 linux kvms > for multimedia services and auxiliary stuff. since i don't have a > dedicated room where noise does not matter i have to consider quiet > "workstation hardware" instead of noisy "server hardware". > > here is the hardware setup i am planning to go with: > - HP Z230 (WM583EA#UUZ) which is basically > -> Intel PCH C226 chipset > -> E3-1225v3 quad core XEON > -> Intel I217LM PCIe Gigabit-Controller > > - 4x 8 GB ECC RAM (HP A2Z50AA) > - 1x 1 TB WD Re, S-ATA III (512n) as rpool > - 2x 4 TB WD Re, S-ATA III (512n) as a data mirror > > > what do you think about this hardware setup? any no-gos? > > thanks > dominik > > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From hasslerd at gmx.li Wed Oct 22 17:07:09 2014 From: hasslerd at gmx.li (Dominik Hassler) Date: Wed, 22 Oct 2014 19:07:09 +0200 Subject: [OmniOS-discuss] HP Z230 as a home server In-Reply-To: <52DCA0DF-14A9-44F8-8B62-E29749D3645D@clozure.com> References: <5446B314.9020600@gmx.li> <52DCA0DF-14A9-44F8-8B62-E29749D3645D@clozure.com> Message-ID: <5447E43D.5070303@gmx.li> your inputs made me rethink my "strategy". especially IPMI capability of the X10SLM-F. and all that for a decent price... so instead of going for a dealer pre-configured HP solution which offers me very little flexibility, i'll make my own based on your suggestions whilst saving some money on top: - MBD-X10SLM-F-O mainboard - Xeon E3-1231v3 (this time w/ HT) - 4x 8 GB KTH-PL316EK4/32G - 2x SSD DC S3500 (rpool mirror) - 2x WD RE 4T (data mirror) - Lian Li PC-V358 case @dan: can you give some details about your SSD wear-leveling setup? is it basically just a slice of 20GB (why 20?) of unallocated disk space or is there more magic behind it? does anyone have (good or bad) experience with the kingston RAM mentioned above? thanks dominik On 10/21/2014 11:09 PM, R. Matthew Emerson wrote: > > On Oct 21, 2014, at 4:24 PM, Dan McDonald wrote: > >> >>> On Oct 21, 2014, at 3:25 PM, Dominik Hassler wrote: >> >> >> >>> of course i'd like to have an omnios server at home, too. nothing fancy, >>> basically just for data storage (nfs and smb sharing) and 1-2 linux kvms >>> for multimedia services and auxiliary stuff. since i don't have a >>> dedicated room where noise does not matter i have to consider quiet >>> "workstation hardware" instead of noisy "server hardware". >>> >>> here is the hardware setup i am planning to go with: >>> - HP Z230 (WM583EA#UUZ) which is basically >> >> http://www8.hp.com/us/en/campaigns/workstations/z230.html >> >> Interesting. It's a socket 1150 workstation. Good for illumos. >> >>> -> Intel PCH C226 chipset >>> -> E3-1225v3 quad core XEON >> >> Gotta get that ECC support. Also, while not HT, it does have good clock speed (> 3GHz). >> >>> -> Intel I217LM PCIe Gigabit-Controller >> >> Supported, and relatively new. Should show up as e1000g. I notice you can get a 2nd ethernet with an I210. This would show up as igb. >> >>> - 4x 8 GB ECC RAM (HP A2Z50AA) >> >> Gotta do ECC! >> >>> - 1x 1 TB WD Re, S-ATA III (512n) as rpool >>> - 2x 4 TB WD Re, S-ATA III (512n) as a data mirror >> >> I might swap in an SSD for rpool, but that's just me. (NOTE: If you do use an SSD, don't use it for swap, though with 32GB, you won't need swap at all). And for your data, it doesn't matter (I don't think) if it's 512 or 4k sectors, just so long as you know what it is up front, they MATCH, and set things appropriately when creating the pool. >> >> If others here have opinions, you ought to share 'em. I'm running a Supermicro not unlike what the HP offers: >> >> http://kebesays.blogspot.com/2014/06/home-data-center-20-dogfooding-again.html >> >> I got more NICs though. :) > > > My small home server is a Supermicro X10SLM-F-O with a Xeon 1230v3 in a Supermicro CSE-731i-300B case. The board is nice because it has two NICs (one e1000g, one igb), and also a dedicated one for IPMI. I used memory from Crucial (CT2KIT102472BD160B) and a pair of Toshiba 2 TB disks (PH3200U-1I72) as my data pool (on sale at the local Micro Center). I'm using a 500GB Hitachi disk that I happened to have around for my rpool. > > This ended up being around $900, which I thought was way cheap. > > If I did it again, I might consider the larger 732 series case, which is supposed to be a little bit quieter. My desktop box is a pretty quiet 27" iMac, and I can hear the Supermicro box's fans from about 5 feet away. The noise not really objectionable, but it's not silent either. > > The only problem I had with this hardware was that fastboot would hang, but the suggestion in http://article.gmane.org/gmane.os.omnios.general/1043 solved that. > > I would have gone with a factory-built box, but Dell wasn't selling Haswell Xeon E3 systems at the time, and HP was trying hard to discourage customers from using non-HP disks. > -------------- next part -------------- A non-text attachment was scrubbed... Name: 0xF9ECC6A5.asc Type: application/pgp-keys Size: 2686 bytes Desc: not available URL: From doug at will.to Wed Oct 22 19:14:58 2014 From: doug at will.to (Doug Hughes) Date: Wed, 22 Oct 2014 15:14:58 -0400 Subject: [OmniOS-discuss] strange console messages Message-ID: I'm seeing these on a 'mature' x4275 server. I tried googling this, but didn't turn up anything. Any idea what it means or how severe it is? ###!!! [Child][DispatchInterruptMessage] Error: Unknown message: not processed ###!!! [Child][DispatchInterruptMessage] Error: Unknown message: not processed ###!!! [Child][DispatchInterruptMessage] Error: Unknown message: not processed ###!!! [Child][DispatchInterruptMessage] Error: Unknown message: not processed (they stagger and appear in bunches) -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdg117 at elvis.arl.psu.edu Thu Oct 23 13:49:14 2014 From: jdg117 at elvis.arl.psu.edu (John D Groenveld) Date: Thu, 23 Oct 2014 09:49:14 -0400 Subject: [OmniOS-discuss] r151012 Message-ID: <201410231349.s9NDnExQ000576@elvis.arl.psu.edu> Trying to update from r151010 to r151012, and hit this... # pkg update -v DOWNLOAD PKGS FILES XFER (MB) library/python-2/lxml-26 12/398 292/17474 2.7/310.7$<3> Errors were encountered while attempting to retrieve package or file data for the requested operation. Details follow: Framework error: code: 52 reason: Empty reply from server URL: 'http://pkg.omniti.com/omnios/r151012/omnios/file/1/9b8d58d2ec902dc305f5e64a7aa4184dbdc3b6c1'. (happened 4 times) $ wget http://pkg.omniti.com/omnios/r151012/omnios/file/1/9b8d58d2ec902dc305f5e64a7aa4184dbdc3b6c1 Resolving pkg.omniti.com... 199.15.226.221 Connecting to pkg.omniti.com|199.15.226.221|:80... connected. HTTP request sent, awaiting response... 404 Not Found 2014-10-23 09:46:55 ERROR 404: Not Found. John groenveld at acm.org From danmcd at omniti.com Thu Oct 23 14:33:28 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 23 Oct 2014 10:33:28 -0400 Subject: [OmniOS-discuss] r151012 In-Reply-To: <201410231349.s9NDnExQ000576@elvis.arl.psu.edu> References: <201410231349.s9NDnExQ000576@elvis.arl.psu.edu> Message-ID: <88F70275-5BDD-4E83-A7F2-09C23BFBC9D4@omniti.com> > On Oct 23, 2014, at 9:49 AM, John D Groenveld wrote: > > Trying to update from r151010 to r151012, and hit this... > # pkg update -v > DOWNLOAD PKGS FILES XFER (MB) > library/python-2/lxml-26 12/398 292/17474 2.7/310.7$<3> > > > Errors were encountered while attempting to retrieve package or file data for > the requested operation. > Details follow: > > Framework error: code: 52 reason: Empty reply from server > URL: 'http://pkg.omniti.com/omnios/r151012/omnios/file/1/9b8d58d2ec902dc305f5e64a7aa4184dbdc3b6c1'. (happened 4 times) > > $ wget http://pkg.omniti.com/omnios/r151012/omnios/file/1/9b8d58d2ec902dc305f5e64a7aa4184dbdc3b6c1 > Resolving pkg.omniti.com... 199.15.226.221 > Connecting to pkg.omniti.com|199.15.226.221|:80... connected. > HTTP request sent, awaiting response... 404 Not Found > 2014-10-23 09:46:55 ERROR 404: Not Found. Can you try again? It may have been a blip in pkg.omniti.com. Meanwhile, I've pinged ops about it, in case there's something actually wrong. Thanks! Dan From cks at cs.toronto.edu Thu Oct 23 16:59:38 2014 From: cks at cs.toronto.edu (Chris Siebenmann) Date: Thu, 23 Oct 2014 12:59:38 -0400 Subject: [OmniOS-discuss] Oddity in how much reserved space there is in ZFS pools? Message-ID: <20141023165938.C30A95A0789@testapps.cs.toronto.edu> If you have a ZFS pool with mirror vdevs and you look at 'zfs list' versus 'zpool list', you can see that the available space is somewhat different between the two. This is a known issue and comes about because the ZFS code reserves some amount of space that can't be consumed in normal use and 'zfs list' accounts for this; see eg http://cuddletech.com/blog/pivot/entry.php?id=1013 In current OmniOS and Illumos source this is discussed (and handled) in the code and comments around spa_slop_shift, spa_get_slop_space(), and dsl_pool_adjustedsize(). In this, both the code and the comments say that the reserved space should be about 3.2% of the pool's space, or more specifically 1/32nd of it. However, this is not the behavior I observe in test pools; instead when I run a test pool completely out of space at the user level (eg with 'dd if=/dev/zero of=spaceeater'), with 'zfs list' reporting 0 bytes free, 'zpool list' reports only about 1.56% free. I'm concerned that there's something funny going on here such that the space limitation is not working properly. Are things working as expected and I'm just missing something? Thanks in advance. (This is on OmniOS r151010 on an ashift=12 pool with two vdevs, each of them a 2-way mirror. Specific numbers are that my test pool reports a total size (in 'zpool list') of 1392 GB and when 'zfs list' reports 0 bytes free 'zpool list' reports 21.7 GB free. Unless I'm doing the math wrong, 1/32nd would be 43 GB or so.) - cks From mir at miras.org Thu Oct 23 23:19:04 2014 From: mir at miras.org (Michael Rasmussen) Date: Fri, 24 Oct 2014 01:19:04 +0200 Subject: [OmniOS-discuss] infiniband Message-ID: <20141024011904.08b18afc@sleipner.datanom.net> Hi all, Is infiniband supported in r151012. It is these cards I had in mind: Voltaire 500 EX-D Dual Port Host Channel Adapter PCIe 20Gbps DDR InfiniBand Card -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: Process promptly. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From turbo124 at gmail.com Thu Oct 23 23:24:19 2014 From: turbo124 at gmail.com (David Bomba) Date: Fri, 24 Oct 2014 10:24:19 +1100 Subject: [OmniOS-discuss] infiniband In-Reply-To: <20141024011904.08b18afc@sleipner.datanom.net> References: <20141024011904.08b18afc@sleipner.datanom.net> Message-ID: yes, tavor and hermon drivers are present, and function well. On 24 October 2014 10:19, Michael Rasmussen wrote: > Hi all, > > Is infiniband supported in r151012. > > It is these cards I had in mind: > Voltaire 500 EX-D Dual Port Host Channel Adapter PCIe 20Gbps DDR > InfiniBand Card > > -- > Hilsen/Regards > Michael Rasmussen > > Get my public GnuPG keys: > michael rasmussen cc > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E > mir datanom net > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C > mir miras org > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 > -------------------------------------------------------------- > /usr/games/fortune -es says: > Process promptly. > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Thu Oct 23 23:25:47 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 23 Oct 2014 19:25:47 -0400 Subject: [OmniOS-discuss] infiniband In-Reply-To: <20141024011904.08b18afc@sleipner.datanom.net> References: <20141024011904.08b18afc@sleipner.datanom.net> Message-ID: > On Oct 23, 2014, at 7:19 PM, Michael Rasmussen wrote: > > Hi all, > > Is infiniband supported in r151012. Only older cards. > It is these cards I had in mind: > Voltaire 500 EX-D Dual Port Host Channel Adapter PCIe 20Gbps DDR > InfiniBand Card Do you know the PCI ID of that card? I can double-check, but I'm guessing the answer will be no. Dan From mir at miras.org Thu Oct 23 23:49:09 2014 From: mir at miras.org (Michael Rasmussen) Date: Fri, 24 Oct 2014 01:49:09 +0200 Subject: [OmniOS-discuss] infiniband In-Reply-To: References: <20141024011904.08b18afc@sleipner.datanom.net> Message-ID: <20141024014909.48a7bd3a@sleipner.datanom.net> On Thu, 23 Oct 2014 19:25:47 -0400 Dan McDonald wrote: > > Only older cards. > By older cards do you the mean only 10 gbps cards? > > It is these cards I had in mind: > > Voltaire 500 EX-D Dual Port Host Channel Adapter PCIe 20Gbps DDR > > InfiniBand Card > > Do you know the PCI ID of that card? I can double-check, but I'm guessing the answer will be no. > > Dan > prtconf -v from Solaris 11.1 https://forums.servethehome.com/index.php?threads/connectx-vpi-infiniband-card-and-solaris-11-1.3211/ -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: [Sir Stafford Cripps] has all the virtues I dislike and none of the vices I admire. -- Winston Churchill -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From danmcd at omniti.com Fri Oct 24 00:02:16 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 23 Oct 2014 20:02:16 -0400 Subject: [OmniOS-discuss] infiniband In-Reply-To: <20141024014909.48a7bd3a@sleipner.datanom.net> References: <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> Message-ID: <1AB52ABA-A846-4015-A797-5118283B82A4@omniti.com> > On Oct 23, 2014, at 7:49 PM, Michael Rasmussen wrote: > > On Thu, 23 Oct 2014 19:25:47 -0400 > Dan McDonald wrote: > >> >> Only older cards. >> > By older cards do you the mean only 10 gbps cards? I mean there's been no active development in IB on illumos. Doesn't mean an older driver won't work, however, because... >>> It is these cards I had in mind: >>> Voltaire 500 EX-D Dual Port Host Channel Adapter PCIe 20Gbps DDR >>> InfiniBand Card >> >> Do you know the PCI ID of that card? I can double-check, but I'm guessing the answer will be no. >> >> Dan >> > prtconf -v from Solaris 11.1 > https://forums.servethehome.com/index.php?threads/connectx-vpi-infiniband-card-and-solaris-11-1.3211/ ...buried in that output, I found a match: -bash-4.3$ grep 15b3,634a /etc/driver_aliases hermon "pciex15b3,634a" -bash-4.3$ So I'd give it a whirl. Dan From ikaufman at eng.ucsd.edu Fri Oct 24 00:07:25 2014 From: ikaufman at eng.ucsd.edu (Ian Kaufman) Date: Thu, 23 Oct 2014 17:07:25 -0700 Subject: [OmniOS-discuss] infiniband In-Reply-To: <20141024014909.48a7bd3a@sleipner.datanom.net> References: <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> Message-ID: I have had an issue with the Hermon driver and Mellanox ConnectX2 QDR cards. On one system, I saw this in the log: massive02-a hermon: [ID 492207 kern.info] hermon1: CQE RNR retry counter exceeded and the card went offline. Now, the interface does not even come up. Three other identical systems are fine, as are two other systems with the same card running Illumian 1.0a. I haven't found a way to reset the RNR counter. Ian On Thu, Oct 23, 2014 at 4:49 PM, Michael Rasmussen wrote: > On Thu, 23 Oct 2014 19:25:47 -0400 > Dan McDonald wrote: > >> >> Only older cards. >> > By older cards do you the mean only 10 gbps cards? > >> > It is these cards I had in mind: >> > Voltaire 500 EX-D Dual Port Host Channel Adapter PCIe 20Gbps DDR >> > InfiniBand Card >> >> Do you know the PCI ID of that card? I can double-check, but I'm guessing the answer will be no. >> >> Dan >> > prtconf -v from Solaris 11.1 > https://forums.servethehome.com/index.php?threads/connectx-vpi-infiniband-card-and-solaris-11-1.3211/ > > -- > Hilsen/Regards > Michael Rasmussen > > Get my public GnuPG keys: > michael rasmussen cc > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E > mir datanom net > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C > mir miras org > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 > -------------------------------------------------------------- > /usr/games/fortune -es says: > [Sir Stafford Cripps] has all the virtues I dislike and none of the > vices I admire. > -- Winston Churchill > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -- Ian Kaufman Research Systems Administrator UC San Diego, Jacobs School of Engineering ikaufman AT ucsd DOT edu From mir at miras.org Fri Oct 24 00:11:05 2014 From: mir at miras.org (Michael Rasmussen) Date: Fri, 24 Oct 2014 02:11:05 +0200 Subject: [OmniOS-discuss] infiniband In-Reply-To: <1AB52ABA-A846-4015-A797-5118283B82A4@omniti.com> References: <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <1AB52ABA-A846-4015-A797-5118283B82A4@omniti.com> Message-ID: <20141024021105.6b35f5cf@sleipner.datanom.net> On Thu, 23 Oct 2014 20:02:16 -0400 Dan McDonald wrote: > > ...buried in that output, I found a match: > > -bash-4.3$ grep 15b3,634a /etc/driver_aliases > hermon "pciex15b3,634a" > -bash-4.3$ > Nice. My winter project is saved;-) I intend to convert storage and hypervisor communication to infiniband for fast and cheaper speed pr. $. 20 gbps cards and switches come cheap these days. Dual port cards at about 50$ and a 24 port switch can be purchased for about 300$ Thanks Dan and David. -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: If you can read this, you're too close. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From turbo124 at gmail.com Fri Oct 24 00:51:30 2014 From: turbo124 at gmail.com (David Bomba) Date: Fri, 24 Oct 2014 11:51:30 +1100 Subject: [OmniOS-discuss] infiniband In-Reply-To: References: <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> Message-ID: I have the exact same issue Ian. I have multiple XenServers connected to OmniOS storage units. Under specific circumstances, I get iscsi disconnects which prove fatal for the XenServer host guests. These machines run a mixed bag of ConnectX and ConnectX-2 cards I have managed to get the system stable, but with severe compromises. Previously using 10Gb cards, I didnt have any problems what so ever. On 24 October 2014 11:07, Ian Kaufman wrote: > I have had an issue with the Hermon driver and Mellanox ConnectX2 QDR > cards. On one system, I saw this in the log: > > massive02-a hermon: [ID 492207 kern.info] hermon1: CQE RNR retry > counter exceeded > > and the card went offline. Now, the interface does not even come up. > Three other identical systems are fine, as are two other systems with > the same card running Illumian 1.0a. I haven't found a way to reset > the RNR counter. > > Ian > > On Thu, Oct 23, 2014 at 4:49 PM, Michael Rasmussen wrote: > > On Thu, 23 Oct 2014 19:25:47 -0400 > > Dan McDonald wrote: > > > >> > >> Only older cards. > >> > > By older cards do you the mean only 10 gbps cards? > > > >> > It is these cards I had in mind: > >> > Voltaire 500 EX-D Dual Port Host Channel Adapter PCIe 20Gbps DDR > >> > InfiniBand Card > >> > >> Do you know the PCI ID of that card? I can double-check, but I'm > guessing the answer will be no. > >> > >> Dan > >> > > prtconf -v from Solaris 11.1 > > > https://forums.servethehome.com/index.php?threads/connectx-vpi-infiniband-card-and-solaris-11-1.3211/ > > > > -- > > Hilsen/Regards > > Michael Rasmussen > > > > Get my public GnuPG keys: > > michael rasmussen cc > > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E > > mir datanom net > > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C > > mir miras org > > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 > > -------------------------------------------------------------- > > /usr/games/fortune -es says: > > [Sir Stafford Cripps] has all the virtues I dislike and none of the > > vices I admire. > > -- Winston Churchill > > > > _______________________________________________ > > OmniOS-discuss mailing list > > OmniOS-discuss at lists.omniti.com > > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > > > > > -- > Ian Kaufman > Research Systems Administrator > UC San Diego, Jacobs School of Engineering ikaufman AT ucsd DOT edu > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mir at miras.org Fri Oct 24 00:57:47 2014 From: mir at miras.org (Michael Rasmussen) Date: Fri, 24 Oct 2014 02:57:47 +0200 Subject: [OmniOS-discuss] infiniband In-Reply-To: References: <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> Message-ID: <20141024025747.1a7c959e@sleipner.datanom.net> On Fri, 24 Oct 2014 11:51:30 +1100 David Bomba wrote: > I have the exact same issue Ian. > > I have multiple XenServers connected to OmniOS storage units. Under > specific circumstances, I get iscsi disconnects which prove fatal for the > XenServer host guests. These machines run a mixed bag of ConnectX and > ConnectX-2 cards > > I have managed to get the system stable, but with severe compromises. > > Previously using 10Gb cards, I didnt have any problems what so ever. > ConnectX and ConnectX-2 is 40 gpbs cards? -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: Indifference will certainly be the downfall of mankind, but who cares? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From ikaufman at eng.ucsd.edu Fri Oct 24 01:01:22 2014 From: ikaufman at eng.ucsd.edu (Ian Kaufman) Date: Thu, 23 Oct 2014 18:01:22 -0700 Subject: [OmniOS-discuss] infiniband In-Reply-To: <20141024025747.1a7c959e@sleipner.datanom.net> References: <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <20141024025747.1a7c959e@sleipner.datanom.net> Message-ID: Mellanox ConnectX-2 cards came in SDR (10Gbps) , DDR (20Gbps), and QDR (40Gpbs) configs. Ian On Thu, Oct 23, 2014 at 5:57 PM, Michael Rasmussen wrote: > On Fri, 24 Oct 2014 11:51:30 +1100 > David Bomba wrote: > >> I have the exact same issue Ian. >> >> I have multiple XenServers connected to OmniOS storage units. Under >> specific circumstances, I get iscsi disconnects which prove fatal for the >> XenServer host guests. These machines run a mixed bag of ConnectX and >> ConnectX-2 cards >> >> I have managed to get the system stable, but with severe compromises. >> >> Previously using 10Gb cards, I didnt have any problems what so ever. >> > ConnectX and ConnectX-2 is 40 gpbs cards? > > -- > Hilsen/Regards > Michael Rasmussen > > Get my public GnuPG keys: > michael rasmussen cc > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E > mir datanom net > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C > mir miras org > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 > -------------------------------------------------------------- > /usr/games/fortune -es says: > Indifference will certainly be the downfall of mankind, but who cares? > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -- Ian Kaufman Research Systems Administrator UC San Diego, Jacobs School of Engineering ikaufman AT ucsd DOT edu From danmcd at omniti.com Fri Oct 24 01:56:30 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 23 Oct 2014 21:56:30 -0400 Subject: [OmniOS-discuss] Initial review --> Changes that allow illumos-gate to (mostly) build on OmniOS Message-ID: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> Hello folks from two lists! I've been tweaking around with illumos-gate on OmniOS again. And this time, I have it to where: * You can still build with NO NOISE on OpenIndiana - i.e. it's a NOP to OI builders. * You can *build* and *package* on OmniOS, but there are still 2-3 ELF runtime check hits, and some can't-help-it lint. The "can't-help-it-lint" is because I either generate lint errors on OI due to not having lint-bypass comments, or noise on OmniOS because I DO have lint-bypass comments. I chose to have the OmniOS people cope with the noise. I need 8 commits (all independent of each other) to go into illumos-gate to make what I just mentioned above happen. They are here: http://kebe.com/~danmcd/webrevs/for-omnios/ Some of these may have value ANYWAY for upstream, the rest are modifications that do not affect an OI build, but help an OmniOS one. The 8 commits are: * Some check_rtime modifications for the version of things in OmniOS * Updates to allow buildings with openjdk7, but STILL ALLOW jdk1.6 that's on OI. * Lose the IPP listener if so specified in your .env file, which depends on bits not in OmniOS. * Lose the dependency on docbooks, only existing in the Shell Style Guide. * Lose the SMF dependency on sysidtool. * Allow linkage with more recent versions of net-snmp. * Lint fixes that the OmniOS sunstudio12.1 lint are picky about. * More perl fixes to work with OmniOS's alternative approach to perl support. Additionally, there's a sample .env file that anyone who applies these fixes can use on their OmniOS build machine. I also have a mail_msg from an OmniOS build, to demonstrate what errors are still there in the build even AFTER these changes. I'd like both the illumos developer community and the OmniOS community to look, and share their reviews and opinions. If you have a distro that's not OI or OmniOS, see if these mods help things work on YOUR distro. If I can get more non-OI distros to build or mostly-build stock illumos-gate, it's better for everyone. Thanks! Dan From mir at miras.org Fri Oct 24 05:36:12 2014 From: mir at miras.org (Michael Rasmussen) Date: Fri, 24 Oct 2014 07:36:12 +0200 Subject: [OmniOS-discuss] infiniband In-Reply-To: References: <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <20141024025747.1a7c959e@sleipner.datanom.net> Message-ID: <20141024073612.474ba3ca@sleipner.datanom.net> On Fri, 24 Oct 2014 12:08:14 +1100 David Bomba wrote: > Correct, they are 40Gb/s cards. I have heard same stories of problems with 40Gb/s cards on Linux and FreeBSD. What it boils down to is that at this speed software subnet manager is causing problems so always use a switch with hardware subnet manager. -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: LordHavoc: I'm already insane. damn straight. or curvy, crooked, or what have you -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From johan.kragsterman at capvert.se Fri Oct 24 06:37:12 2014 From: johan.kragsterman at capvert.se (Johan Kragsterman) Date: Fri, 24 Oct 2014 08:37:12 +0200 Subject: [OmniOS-discuss] Ang: Re: infiniband In-Reply-To: <20141024073612.474ba3ca@sleipner.datanom.net> References: <20141024073612.474ba3ca@sleipner.datanom.net>, <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <20141024025747.1a7c959e@sleipner.datanom.net> Message-ID: -----"OmniOS-discuss" skrev: ----- Till: omnios-discuss Fr?n: Michael Rasmussen S?nt av: "OmniOS-discuss" Datum: 2014-10-24 07:37 ?rende: Re: [OmniOS-discuss] infiniband On Fri, 24 Oct 2014 12:08:14 +1100 David Bomba wrote: > Correct, they are 40Gb/s cards. I have heard same stories of problems with 40Gb/s cards on Linux and FreeBSD. What it boils down to is that at this speed software subnet manager is causing problems so always use a switch with hardware subnet manager. Hi, Michael! Interesting! Do you have any links to discussions or documentation of that? Rgrds Johan -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: LordHavoc: I'm already insane. damn straight. or curvy, crooked, or what have you _______________________________________________ OmniOS-discuss mailing list OmniOS-discuss at lists.omniti.com http://lists.omniti.com/mailman/listinfo/omnios-discuss [bilagan "attej6sv.dat" borttagen av Johan Kragsterman/Capvert] From narayan.desai at gmail.com Fri Oct 24 07:07:59 2014 From: narayan.desai at gmail.com (Narayan Desai) Date: Fri, 24 Oct 2014 00:07:59 -0700 Subject: [OmniOS-discuss] infiniband In-Reply-To: <20141024073612.474ba3ca@sleipner.datanom.net> References: <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <20141024025747.1a7c959e@sleipner.datanom.net> <20141024073612.474ba3ca@sleipner.datanom.net> Message-ID: On Thu, Oct 23, 2014 at 10:36 PM, Michael Rasmussen wrote: > On Fri, 24 Oct 2014 12:08:14 +1100 > David Bomba wrote: > > > Correct, they are 40Gb/s cards. > I have heard same stories of problems with 40Gb/s cards on Linux and > FreeBSD. What it boils down to is that at this speed software subnet > manager is causing problems so always use a switch with hardware subnet > manager. > This sounds apocryphal and wrong to me. I know of a several large infiniband networks (1k+ host ports, non-blocking) that run with subnet managers running on linux machines and generally work pretty well. A few remarks: - subnet managers are a part of the control plane, they don't participate directly in the data plane after initial connection setup. This means that the footprint of a QDR port is the same as an SDR port; data packets aren't processed through the subnet manager. - subnet managers aren't bug-free -- the versions integrated into switch firmware can't be upgraded a la carte, so you are dependent on the vendor to release a new version of the switch firmware to get bugfixes. - the processors included in switches are embedded and hence slow. this could easily make a difference to subnet performance, if you need to talk to the subnet manager frequently to get new path records. I'd strongly recommend against running switch provided subnet managers. -nld -------------- next part -------------- An HTML attachment was scrubbed... URL: From mir at miras.org Fri Oct 24 09:53:05 2014 From: mir at miras.org (mir at miras.org) Date: Fri, 24 Oct 2014 11:53:05 +0200 Subject: [OmniOS-discuss] Ang: Re: infiniband In-Reply-To: References: <20141024073612.474ba3ca@sleipner.datanom.net>, " <20141024011904.08b18afc@sleipner.datanom.net>" " <20141024014909.48a7bd3a@sleipner.datanom.net>" " " <20141024025747.1a7c959e@sleipner.datanom.net> Message-ID: <04f3d7688b5e75dbafa57e787c37ed74@miras.org> On 2014-10-24 08:37, Johan Kragsterman wrote: > > Hi, Michael! > > Interesting! Do you have any links to discussions or documentation of > that? > See this thread: http://forum.proxmox.com/threads/19789-Infiniband-40gbps-capped-at-12-5gbps ---- This mail was virus scanned and spam checked before delivery. This mail is also DKIM signed. See header dkim-signature. From johan.kragsterman at capvert.se Fri Oct 24 11:17:39 2014 From: johan.kragsterman at capvert.se (Johan Kragsterman) Date: Fri, 24 Oct 2014 13:17:39 +0200 Subject: [OmniOS-discuss] Ang: Re: Ang: Re: infiniband In-Reply-To: <04f3d7688b5e75dbafa57e787c37ed74@miras.org> References: <04f3d7688b5e75dbafa57e787c37ed74@miras.org>, <20141024073612.474ba3ca@sleipner.datanom.net>, " <20141024011904.08b18afc@sleipner.datanom.net>" " <20141024014909.48a7bd3a@sleipner.datanom.net>" " " <20141024025747.1a7c959e@sleipner.datanom.net> Message-ID: -----mir at miras.org skrev: ----- Till: Johan Kragsterman Fr?n: mir at miras.org Datum: 2014-10-24 11:53 Kopia: omnios-discuss ?rende: Re: Ang: Re: [OmniOS-discuss] infiniband On 2014-10-24 08:37, Johan Kragsterman wrote: > > Hi, Michael! > > Interesting! Do you have any links to discussions or documentation of > that? > See this thread: http://forum.proxmox.com/threads/19789-Infiniband-40gbps-capped-at-12-5gbps Thanks, Michael! Well, I can only find one person in that thread that BELIEVES that running SM on a switch is preferable to run OpenSM software. He hardly comes with any documentation of this, and I can't see any other that agrees... But of coarse, if you provide too little resources for OpenSM it will cause problems, or if you have any problems with the S/W or the configuration... Rgrds Johan ---- This mail was virus scanned and spam checked before delivery. This mail is also DKIM signed. See header dkim-signature. From mir at miras.org Fri Oct 24 14:55:50 2014 From: mir at miras.org (Michael Rasmussen) Date: Fri, 24 Oct 2014 16:55:50 +0200 Subject: [OmniOS-discuss] Ang: Re: Ang: Re: infiniband In-Reply-To: References: <04f3d7688b5e75dbafa57e787c37ed74@miras.org> <20141024073612.474ba3ca@sleipner.datanom.net> <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <20141024025747.1a7c959e@sleipner.datanom.net> Message-ID: <20141024165550.4e82ee5c@sleipner.datanom.net> On Fri, 24 Oct 2014 13:17:39 +0200 Johan Kragsterman wrote: > > Well, I can only find one person in that thread that BELIEVES that running SM on a switch is preferable to run OpenSM software. He hardly comes with any documentation of this, and I can't see any other that agrees... > > But of coarse, if you provide too little resources for OpenSM it will cause problems, or if you have any problems with the S/W or the configuration... > I think the issue here is that OpenSM is a single threaded application so when the CPU which is assigned to it maxes out performance will drop. So I guess it comes down to a trade-off between the performance of the host CPU and the performance of the switch hardware. -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: The default Magic Word, "Abracadabra", actually is a corruption of the Hebrew phrase "ha-Bracha dab'ra" which means "pronounce the blessing". -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From ikaufman at eng.ucsd.edu Fri Oct 24 14:58:16 2014 From: ikaufman at eng.ucsd.edu (Ian Kaufman) Date: Fri, 24 Oct 2014 07:58:16 -0700 Subject: [OmniOS-discuss] Ang: Re: Ang: Re: infiniband In-Reply-To: <20141024165550.4e82ee5c@sleipner.datanom.net> References: <04f3d7688b5e75dbafa57e787c37ed74@miras.org> <20141024073612.474ba3ca@sleipner.datanom.net> <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <20141024025747.1a7c959e@sleipner.datanom.net> <20141024165550.4e82ee5c@sleipner.datanom.net> Message-ID: I use a QLogic 12300 with SM built in. I haven't had any issues, saving my cluster frontend's cycles for other things. Ian On Fri, Oct 24, 2014 at 7:55 AM, Michael Rasmussen wrote: > On Fri, 24 Oct 2014 13:17:39 +0200 > Johan Kragsterman wrote: > >> >> Well, I can only find one person in that thread that BELIEVES that running SM on a switch is preferable to run OpenSM software. He hardly comes with any documentation of this, and I can't see any other that agrees... >> >> But of coarse, if you provide too little resources for OpenSM it will cause problems, or if you have any problems with the S/W or the configuration... >> > I think the issue here is that OpenSM is a single threaded application > so when the CPU which is assigned to it maxes out performance will > drop. So I guess it comes down to a trade-off between the > performance of the host CPU and the performance of the switch hardware. > > -- > Hilsen/Regards > Michael Rasmussen > > Get my public GnuPG keys: > michael rasmussen cc > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E > mir datanom net > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C > mir miras org > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 > -------------------------------------------------------------- > /usr/games/fortune -es says: > The default Magic Word, "Abracadabra", actually is a corruption of the > Hebrew phrase "ha-Bracha dab'ra" which means "pronounce the blessing". > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -- Ian Kaufman Research Systems Administrator UC San Diego, Jacobs School of Engineering ikaufman AT ucsd DOT edu From ikaufman at eng.ucsd.edu Fri Oct 24 15:35:54 2014 From: ikaufman at eng.ucsd.edu (Ian Kaufman) Date: Fri, 24 Oct 2014 08:35:54 -0700 Subject: [OmniOS-discuss] Ang: Re: Ang: Re: infiniband In-Reply-To: References: <04f3d7688b5e75dbafa57e787c37ed74@miras.org> <20141024073612.474ba3ca@sleipner.datanom.net> <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <20141024025747.1a7c959e@sleipner.datanom.net> <20141024165550.4e82ee5c@sleipner.datanom.net> Message-ID: I wonder if Nexenta has added more/better support for IB? I believe that Illumian 1.0 is finally out (after having run the Alpha for years), and I know that they did have some interest in supporting IB originally. If they have, then I would hope some of that support would make it back upstream. Mayhaps I have an email to send out ... Ian On Fri, Oct 24, 2014 at 7:58 AM, Ian Kaufman wrote: > I use a QLogic 12300 with SM built in. I haven't had any issues, > saving my cluster frontend's cycles for other things. > > Ian > > On Fri, Oct 24, 2014 at 7:55 AM, Michael Rasmussen wrote: >> On Fri, 24 Oct 2014 13:17:39 +0200 >> Johan Kragsterman wrote: >> >>> >>> Well, I can only find one person in that thread that BELIEVES that running SM on a switch is preferable to run OpenSM software. He hardly comes with any documentation of this, and I can't see any other that agrees... >>> >>> But of coarse, if you provide too little resources for OpenSM it will cause problems, or if you have any problems with the S/W or the configuration... >>> >> I think the issue here is that OpenSM is a single threaded application >> so when the CPU which is assigned to it maxes out performance will >> drop. So I guess it comes down to a trade-off between the >> performance of the host CPU and the performance of the switch hardware. >> >> -- >> Hilsen/Regards >> Michael Rasmussen >> >> Get my public GnuPG keys: >> michael rasmussen cc >> http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E >> mir datanom net >> http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C >> mir miras org >> http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 >> -------------------------------------------------------------- >> /usr/games/fortune -es says: >> The default Magic Word, "Abracadabra", actually is a corruption of the >> Hebrew phrase "ha-Bracha dab'ra" which means "pronounce the blessing". >> >> _______________________________________________ >> OmniOS-discuss mailing list >> OmniOS-discuss at lists.omniti.com >> http://lists.omniti.com/mailman/listinfo/omnios-discuss >> > > > > -- > Ian Kaufman > Research Systems Administrator > UC San Diego, Jacobs School of Engineering ikaufman AT ucsd DOT edu -- Ian Kaufman Research Systems Administrator UC San Diego, Jacobs School of Engineering ikaufman AT ucsd DOT edu From jeffpc at josefsipek.net Fri Oct 24 15:56:47 2014 From: jeffpc at josefsipek.net (Josef 'Jeff' Sipek) Date: Fri, 24 Oct 2014 11:56:47 -0400 Subject: [OmniOS-discuss] [developer] Initial review --> Changes that allow illumos-gate to (mostly) build on OmniOS In-Reply-To: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> References: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> Message-ID: <20141024155647.GA1317@meili> On Thu, Oct 23, 2014 at 09:56:30PM -0400, Dan McDonald via illumos-developer wrote: > Hello folks from two lists! > > I've been tweaking around with illumos-gate on OmniOS again. And this time, I have it to where: > > * You can still build with NO NOISE on OpenIndiana - i.e. it's a NOP to OI builders. > > * You can *build* and *package* on OmniOS, but there are still 2-3 ELF runtime check hits, and some can't-help-it lint. > > The "can't-help-it-lint" is because I either generate lint errors on OI due to not having lint-bypass comments, or noise on OmniOS because I DO have lint-bypass comments. I chose to have the OmniOS people cope with the noise. > > I need 8 commits (all independent of each other) to go into illumos-gate to make what I just mentioned above happen. They are here: > > http://kebe.com/~danmcd/webrevs/for-omnios/ > > Some of these may have value ANYWAY for upstream, the rest are modifications that do not affect an OI build, but help an OmniOS one. The 8 commits are: > > * Some check_rtime modifications for the version of things in OmniOS > * Updates to allow buildings with openjdk7, but STILL ALLOW jdk1.6 that's on OI. jdtrace/Makefile java_api/Makefile libdtrace_jni/java/Makefile you changed the -target from 1.5 to 1.6. Is this right? (FYI, I know nothing about Java.) damain/pools/Makefile does removing this change the output from 1.5 to 1.6? dnssd/Makefile libdtrace_jni/java/Makefile I'm curious about the details here... Javadoc uses different filenames for the generated gifs and this allows packaging to work since it'll see both old and new files? Am I understanding correctly? Would it make sense to document this hack in the package manifest as well for when someone decides that 1.6 is ancient? A couple of the files have #pragma ident. > * Lose the IPP listener if so specified in your .env file, which depends on bits not in OmniOS. Makefile Is this the only IPP related package? Did the change a few years ago that introduced ENABLE_IPP_PRINTING not change any packaging? > * Lose the dependency on docbooks, only existing in the Shell Style Guide. Cool! I think this also nukes one of the sources of wsdiff noise. > * Lose the SMF dependency on sysidtool. > * Allow linkage with more recent versions of net-snmp. > * Lint fixes that the OmniOS sunstudio12.1 lint are picky about. > * More perl fixes to work with OmniOS's alternative approach to perl support. Makefile.master Hrm? Why are both -64int? Is that right? Ah, I see... you just moved it from Makefile.perl. LGTM. > Additionally, there's a sample .env file that anyone who applies these > fixes can use on their OmniOS build machine. I also have a mail_msg from > an OmniOS build, to demonstrate what errors are still there in the build > even AFTER these changes. > > I'd like both the illumos developer community and the OmniOS community to > look, and share their reviews and opinions. If you have a distro that's > not OI or OmniOS, see if these mods help things work on YOUR distro. If I > can get more non-OI distros to build or mostly-build stock illumos-gate, > it's better for everyone. You said OI... did you try both /dev (151a9) as well as hipster? Just curious. Jeff. -- NT is to UNIX what a doughnut is to a particle accelerator. From danmcd at omniti.com Fri Oct 24 16:01:21 2014 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 24 Oct 2014 12:01:21 -0400 Subject: [OmniOS-discuss] [developer] Initial review --> Changes that allow illumos-gate to (mostly) build on OmniOS In-Reply-To: <20141024155647.GA1317@meili> References: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> <20141024155647.GA1317@meili> Message-ID: <8B714042-F636-4C10-8105-3838053B1FD9@omniti.com> > On Oct 24, 2014, at 11:56 AM, Josef 'Jeff' Sipek wrote: > > On Thu, Oct 23, 2014 at 09:56:30PM -0400, Dan McDonald via illumos-developer wrote: >> >> * Some check_rtime modifications for the version of things in OmniOS >> * Updates to allow buildings with openjdk7, but STILL ALLOW jdk1.6 that's on OI. > > jdtrace/Makefile > java_api/Makefile > libdtrace_jni/java/Makefile > you changed the -target from 1.5 to 1.6. Is this right? (FYI, I > know nothing about Java.) I did, because from what I can tell, OI uses 1.6 now. OI's 1.6 is the earliest Java we support, from what I understand. > damain/pools/Makefile > does removing this change the output from 1.5 to 1.6? Yes. > dnssd/Makefile > libdtrace_jni/java/Makefile > I'm curious about the details here... Javadoc uses different > filenames for the generated gifs and this allows packaging to work > since it'll see both old and new files? Am I understanding > correctly? Would it make sense to document this hack in the package > manifest as well for when someone decides that 1.6 is ancient? openjdk7's javadoc generates other files, yes. This hack does allow 1.6 and 1.7 to both work. I probably should document it somewhere. I can't remember if pkg(5) files can take comments in the middle or not. > A couple of the files have #pragma ident. Yeah, should fix those pre-integration. >> * Lose the IPP listener if so specified in your .env file, which depends on bits not in OmniOS. > > Makefile > Is this the only IPP related package? Did the change a few years > ago that introduced ENABLE_IPP_PRINTING not change any packaging? ENABLE_IPP_PRINTING is NEW with this wad. >> * Lose the dependency on docbooks, only existing in the Shell Style Guide. > > Cool! I think this also nukes one of the sources of wsdiff noise. Even better! >> * Lose the SMF dependency on sysidtool. >> * Allow linkage with more recent versions of net-snmp. >> * Lint fixes that the OmniOS sunstudio12.1 lint are picky about. >> * More perl fixes to work with OmniOS's alternative approach to perl support. > > Makefile.master > Hrm? Why are both -64int? Is that right? Ah, I see... you just > moved it from Makefile.perl. LGTM. Phew. And for building on OmniOS, you need to override that in your .env file. >> Additionally, there's a sample .env file that anyone who applies these >> fixes can use on their OmniOS build machine. I also have a mail_msg from >> an OmniOS build, to demonstrate what errors are still there in the build >> even AFTER these changes. >> >> I'd like both the illumos developer community and the OmniOS community to >> look, and share their reviews and opinions. If you have a distro that's >> not OI or OmniOS, see if these mods help things work on YOUR distro. If I >> can get more non-OI distros to build or mostly-build stock illumos-gate, >> it's better for everyone. > > You said OI... did you try both /dev (151a9) as well as hipster? Just > curious. I used /dev only (oi_151a9) for my tests. I'd LOVE it if someone took these and tried on hipster. Dan From jeffpc at josefsipek.net Fri Oct 24 16:11:31 2014 From: jeffpc at josefsipek.net (Josef 'Jeff' Sipek) Date: Fri, 24 Oct 2014 12:11:31 -0400 Subject: [OmniOS-discuss] [developer] Initial review --> Changes that allow illumos-gate to (mostly) build on OmniOS In-Reply-To: <8B714042-F636-4C10-8105-3838053B1FD9@omniti.com> References: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> <20141024155647.GA1317@meili> <8B714042-F636-4C10-8105-3838053B1FD9@omniti.com> Message-ID: <20141024161130.GB1317@meili> On Fri, Oct 24, 2014 at 12:01:21PM -0400, Dan McDonald via illumos-developer wrote: > > > On Oct 24, 2014, at 11:56 AM, Josef 'Jeff' Sipek wrote: > > > > On Thu, Oct 23, 2014 at 09:56:30PM -0400, Dan McDonald via illumos-developer wrote: > >> > >> * Some check_rtime modifications for the version of things in OmniOS > >> * Updates to allow buildings with openjdk7, but STILL ALLOW jdk1.6 that's on OI. > > > > jdtrace/Makefile > > java_api/Makefile > > libdtrace_jni/java/Makefile > > you changed the -target from 1.5 to 1.6. Is this right? (FYI, I > > know nothing about Java.) > > I did, because from what I can tell, OI uses 1.6 now. OI's 1.6 is the > earliest Java we support, from what I understand. Fair enough. ... > > dnssd/Makefile > > libdtrace_jni/java/Makefile > > I'm curious about the details here... Javadoc uses different > > filenames for the generated gifs and this allows packaging to work > > since it'll see both old and new files? Am I understanding > > correctly? Would it make sense to document this hack in the package > > manifest as well for when someone decides that 1.6 is ancient? > > openjdk7's javadoc generates other files, yes. This hack does allow 1.6 > and 1.7 to both work. I probably should document it somewhere. I can't > remember if pkg(5) files can take comments in the middle or not. IIRC, it does. ... > >> * Lose the IPP listener if so specified in your .env file, which depends on bits not in OmniOS. > > > > Makefile > > Is this the only IPP related package? Did the change a few years > > ago that introduced ENABLE_IPP_PRINTING not change any packaging? > > ENABLE_IPP_PRINTING is NEW with this wad. Sorry, I was thinking of ENABLE_SMB_PRINTING. ... > >> Additionally, there's a sample .env file that anyone who applies these > >> fixes can use on their OmniOS build machine. I also have a mail_msg from > >> an OmniOS build, to demonstrate what errors are still there in the build > >> even AFTER these changes. > >> > >> I'd like both the illumos developer community and the OmniOS community to > >> look, and share their reviews and opinions. If you have a distro that's > >> not OI or OmniOS, see if these mods help things work on YOUR distro. If I > >> can get more non-OI distros to build or mostly-build stock illumos-gate, > >> it's better for everyone. > > > > You said OI... did you try both /dev (151a9) as well as hipster? Just > > curious. > > I used /dev only (oi_151a9) for my tests. I'd LOVE it if someone took > these and tried on hipster. I'm going to try to run a nightly to see what happens. Jeff. From narayan.desai at gmail.com Fri Oct 24 17:18:48 2014 From: narayan.desai at gmail.com (Narayan Desai) Date: Fri, 24 Oct 2014 10:18:48 -0700 Subject: [OmniOS-discuss] Ang: Re: Ang: Re: infiniband In-Reply-To: <20141024165550.4e82ee5c@sleipner.datanom.net> References: <04f3d7688b5e75dbafa57e787c37ed74@miras.org> <20141024073612.474ba3ca@sleipner.datanom.net> <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <20141024025747.1a7c959e@sleipner.datanom.net> <20141024165550.4e82ee5c@sleipner.datanom.net> Message-ID: On Fri, Oct 24, 2014 at 7:55 AM, Michael Rasmussen wrote: > > I think the issue here is that OpenSM is a single threaded application > so when the CPU which is assigned to it maxes out performance will > drop. So I guess it comes down to a trade-off between the > performance of the host CPU and the performance of the switch hardware. > You're right in a general sort of sense; if the opensm process is resource starved, that will result in slower performance for subnet manager operations. That said, the subnet manager isn't involved in data transfers, except to program switch forwarding tables and answer path requests. Neither of these are particularly important for data transfers, unless the topology is constantly changing (like if links are flapping), or the network is otherwise sick. I wouldn't expect subnet manager performance to effect infiniband transfer performance, unless it was picking poor routes. That would be a function of opensm routing algorithm choice/implementation, not of opensm performance. I'd be much more concerned about fixing bugs in the subnet manager, which is trickier in the case of switch-based subnet managers. I've seen all sorts of weird subnet manager failures over the years, including the subnet manager completely going out to lunch, which can be fixed by upgrading to newer versions. -nld -------------- next part -------------- An HTML attachment was scrubbed... URL: From gmason at msu.edu Fri Oct 24 18:46:47 2014 From: gmason at msu.edu (Greg Mason) Date: Fri, 24 Oct 2014 14:46:47 -0400 Subject: [OmniOS-discuss] Ang: Re: Ang: Re: infiniband In-Reply-To: References: <04f3d7688b5e75dbafa57e787c37ed74@miras.org> <20141024073612.474ba3ca@sleipner.datanom.net> <20141024011904.08b18afc@sleipner.datanom.net> <20141024014909.48a7bd3a@sleipner.datanom.net> <20141024025747.1a7c959e@sleipner.datanom.net> <20141024165550.4e82ee5c@sleipner.datanom.net> Message-ID: <52A5C8FB-A953-4638-9C85-50CE66427997@msu.edu> Enabling IB support for our ZFS-based filers is important enough to us that we have made the decision to migrate away from OmniOS/Illumos over to FreeBSD. FreeBSD 9 has a partial OFED stack, FreeBSD 10 is further along, and development continues for IB support in FreeBSD. I?ve successfully tested a Mellanox ConnectX-3 card on our FDR fabric on a FreeBSD 10.0-RELEASE system. This was with the subnet manager running on a Mellanox switch (which we regularly upgrade). If you?d like more details, please contact me off-list, I?m happy to share :) -Greg Greg Mason Michigan State University Institute for Cyber-Enabled Research High Performance Computing Center web: www.hpcc.msu.edu email: gmason at msu.edu > On Oct 24, 2014, at 11:35 AM, Ian Kaufman wrote: > > I wonder if Nexenta has added more/better support for IB? I believe > that Illumian 1.0 is finally out (after having run the Alpha for > years), and I know that they did have some interest in supporting IB > originally. If they have, then I would hope some of that support would > make it back upstream. > > Mayhaps I have an email to send out ... > > Ian > > On Fri, Oct 24, 2014 at 7:58 AM, Ian Kaufman wrote: >> I use a QLogic 12300 with SM built in. I haven't had any issues, >> saving my cluster frontend's cycles for other things. >> >> Ian >> >> On Fri, Oct 24, 2014 at 7:55 AM, Michael Rasmussen wrote: >>> On Fri, 24 Oct 2014 13:17:39 +0200 >>> Johan Kragsterman wrote: >>> >>>> >>>> Well, I can only find one person in that thread that BELIEVES that running SM on a switch is preferable to run OpenSM software. He hardly comes with any documentation of this, and I can't see any other that agrees... >>>> >>>> But of coarse, if you provide too little resources for OpenSM it will cause problems, or if you have any problems with the S/W or the configuration... >>>> >>> I think the issue here is that OpenSM is a single threaded application >>> so when the CPU which is assigned to it maxes out performance will >>> drop. So I guess it comes down to a trade-off between the >>> performance of the host CPU and the performance of the switch hardware. >>> >>> -- >>> Hilsen/Regards >>> Michael Rasmussen >>> >>> Get my public GnuPG keys: >>> michael rasmussen cc >>> http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E >>> mir datanom net >>> http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C >>> mir miras org >>> http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 >>> -------------------------------------------------------------- >>> /usr/games/fortune -es says: >>> The default Magic Word, "Abracadabra", actually is a corruption of the >>> Hebrew phrase "ha-Bracha dab'ra" which means "pronounce the blessing". >>> >>> _______________________________________________ >>> OmniOS-discuss mailing list >>> OmniOS-discuss at lists.omniti.com >>> http://lists.omniti.com/mailman/listinfo/omnios-discuss >>> >> >> >> >> -- >> Ian Kaufman >> Research Systems Administrator >> UC San Diego, Jacobs School of Engineering ikaufman AT ucsd DOT edu > > > > -- > Ian Kaufman > Research Systems Administrator > UC San Diego, Jacobs School of Engineering ikaufman AT ucsd DOT edu > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From gate03 at landcroft.co.uk Fri Oct 24 20:53:13 2014 From: gate03 at landcroft.co.uk (Michael Mounteney) Date: Sat, 25 Oct 2014 06:53:13 +1000 Subject: [OmniOS-discuss] best guide to tftp ? Message-ID: <20141025065313.4b67dcbc@punda-mlia> Hello, I want to use OmniOS as a boot server for various Linux thin clients. What's the most appropriate guide out there? There's a few, referring to obsolete problems, other OSs etc. Thanks, Michael. From gate03 at landcroft.co.uk Fri Oct 24 20:56:24 2014 From: gate03 at landcroft.co.uk (Michael Mounteney) Date: Sat, 25 Oct 2014 06:56:24 +1000 Subject: [OmniOS-discuss] switching from Stable to LTS In-Reply-To: <20141003011131.5e81500b@sleipner.datanom.net> References: <39bcb14c201f7d1cacc45f29f6ceebfa.squirrel@landcroft.co.uk> <20141003011131.5e81500b@sleipner.datanom.net> Message-ID: <20141025065624.5e60e578@punda-mlia> On Fri, 3 Oct 2014 01:11:31 +0200 Michael Rasmussen wrote: > I think the generel consensus is that unless you are very very very > knowledgeable in the darkest corners of Illumos, having patience above > and beyond the call of duty, have loads of spare time, and only have a > server populated with data of absolut nonsense you should never even > think of downgrading your server(s). OK but let's say despite all the warnings I were going to try anyway. Would it be possible to set up a LTS non-global zone on a machine on which Stable is installed? Michael. From danmcd at omniti.com Fri Oct 24 21:11:11 2014 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 24 Oct 2014 17:11:11 -0400 Subject: [OmniOS-discuss] best guide to tftp ? In-Reply-To: <20141025065313.4b67dcbc@punda-mlia> References: <20141025065313.4b67dcbc@punda-mlia> Message-ID: <530EAE82-EB10-4DE9-9EB8-E237909DF944@omniti.com> > On Oct 24, 2014, at 4:53 PM, Michael Mounteney wrote: > > Hello, I want to use OmniOS as a boot server for various Linux thin > clients. What's the most appropriate guide out there? There's a few, > referring to obsolete problems, other OSs etc. This: http://omnios.omniti.com/wiki.php/KayakMultiRelease specifically talks about setting up for Kayak installation (PXE installation of OmniOS), but I suspect what's there can be modified to put entries into tftpboot/boot/grub/menu.lst that will help your Linux thin clients. Dan From danmcd at omniti.com Fri Oct 24 21:13:50 2014 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 24 Oct 2014 17:13:50 -0400 Subject: [OmniOS-discuss] switching from Stable to LTS In-Reply-To: <20141025065624.5e60e578@punda-mlia> References: <39bcb14c201f7d1cacc45f29f6ceebfa.squirrel@landcroft.co.uk> <20141003011131.5e81500b@sleipner.datanom.net> <20141025065624.5e60e578@punda-mlia> Message-ID: <1E550F75-3AEA-400F-B2DE-59A22B4319E3@omniti.com> > On Oct 24, 2014, at 4:56 PM, Michael Mounteney wrote: > > On Fri, 3 Oct 2014 01:11:31 +0200 > Michael Rasmussen wrote: > >> I think the generel consensus is that unless you are very very very >> knowledgeable in the darkest corners of Illumos, having patience above >> and beyond the call of duty, have loads of spare time, and only have a >> server populated with data of absolut nonsense you should never even >> think of downgrading your server(s). > > OK but let's say despite all the warnings I were going to try anyway. > Would it be possible to set up a LTS non-global zone on a machine on > which Stable is installed? No. The only way to even come close to approaching that would be to run a KVM zone, and put OmniOS LTS in that. There are also pitfalls in doing so (similar to the pitfalls of running TCP over TCP... in this case, running ZFS volumes that will become proper ZFS filesystems). You do realize that the next stable OmniOS is also going to be the next LTS release, right? If you can manage until springtime, there will be a new LTS release, and I suspect much of omniti-ms will be appropriately updated. Dan From sergei25 at gmail.com Fri Oct 24 21:22:46 2014 From: sergei25 at gmail.com (sergei) Date: Fri, 24 Oct 2014 14:22:46 -0700 Subject: [OmniOS-discuss] ZFS Logical Unit stuck in the offlining state with COMSTAR Message-ID: I'm having same issue with OmniOS v11 r151006. Found few threads on this including http://lists.omniti.com/pipermail/omnios-discuss/2013-March/000537.html and https://illumos.org/issues/3621 but still can't figure out whats going on. Basic symptoms are that some zvol-backed LUs can not be deleted. And stmfadm list-lu -v reports LU state as "Offlining". For what its worth we run same Dell R720xd as one of the earlier posters. I would really appreciate few tips on how to resolve this or if COMSTAR / stmf is more reliable in some other Illumos flavor ? -S -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Fri Oct 24 21:26:47 2014 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 24 Oct 2014 17:26:47 -0400 Subject: [OmniOS-discuss] ZFS Logical Unit stuck in the offlining state with COMSTAR In-Reply-To: References: Message-ID: <1675854F-C90E-4153-85B0-328348C78CC7@omniti.com> > On Oct 24, 2014, at 5:22 PM, sergei wrote: > > I'm having same issue with OmniOS v11 r151006. Found few threads on this including http://lists.omniti.com/pipermail/omnios-discuss/2013-March/000537.html and https://illumos.org/issues/3621 but still can't figure out whats going on. Basic symptoms are that some zvol-backed LUs can not be deleted. And stmfadm list-lu -v reports LU state as "Offlining". > > For what its worth we run same Dell R720xd as one of the earlier posters. > > I would really appreciate few tips on how to resolve this or if COMSTAR / stmf is more reliable in some other Illumos flavor ? You can try r151012, which has many bugfixes in it, some of which are COMSTAR/stmf related. Dan From gate03 at landcroft.co.uk Fri Oct 24 21:30:42 2014 From: gate03 at landcroft.co.uk (Michael Mounteney) Date: Sat, 25 Oct 2014 07:30:42 +1000 Subject: [OmniOS-discuss] switching from Stable to LTS In-Reply-To: <1E550F75-3AEA-400F-B2DE-59A22B4319E3@omniti.com> References: <39bcb14c201f7d1cacc45f29f6ceebfa.squirrel@landcroft.co.uk> <20141003011131.5e81500b@sleipner.datanom.net> <20141025065624.5e60e578@punda-mlia> <1E550F75-3AEA-400F-B2DE-59A22B4319E3@omniti.com> Message-ID: <20141025073042.4a35ccde@punda-mlia> On Fri, 24 Oct 2014 17:13:50 -0400 Dan McDonald wrote: > > OK but let's say despite all the warnings I were going to try > > anyway. Would it be possible to set up a LTS non-global zone on a > > machine on which Stable is installed? > > No. So, er, that's a "no" then ? ;-) > The only way [...] > > You do realize that the next stable OmniOS is also going to be the > next LTS release, right? If you can manage until springtime, there > will be a new LTS release, and I suspect much of omniti-ms will be > appropriately updated. Ah ... thinks Northern hemisphere ... so that would be say March 2015? I can keep busy with other things till then, or Steffen Kram's build. Thanks. Of course the real answer is to set up my own package-building system, but time. Michael. From zembower at criterion.com Fri Oct 24 22:08:51 2014 From: zembower at criterion.com (Chris Zembower) Date: Fri, 24 Oct 2014 18:08:51 -0400 Subject: [OmniOS-discuss] Ang: Re: Ang: Re: infiniband Message-ID: ConnectX-2 cards work very well with OmniOS, at least over SRP, and those HCA's can be found for next to nothing. Comstar rules. I feel there's some new momentum behind Infiniband, and hoping that a new driver for the latest hardware will end up in Illumos sometime in the near future (I'm willing to pay a fair bounty). That said, QDR is still extremely fast and can easily co-mingle with FDR fabrics. The subnet manager can be any linux or windows box sitting on the fabric, no need for a managed switch in my experience, although my fabrics are modestly sized and quite stable. To reiterate what others have said, the subnet manager has nothing to do with target/initiator data transfer. OpenSM just scans the fabric at regular intervals for topology changes and handles session setup and teardown. I think of OpenSM as a sort of DHCP server for Infiniband, and in that analogy, it would be strange to believe that even the most resource-deprived DHCP server would affect data transfer on a network. A managed switch is akin to a DHCP server embedded in a router. Convenient, but not necessary. Greg, I'll probably contact you off-list to learn about your FreeBSD implementation. My ConnectX-3 FDR fabric is currently ZoL-based because I couldn't find concrete documentation on the ULP's available in FreeBSD OFED, whereas Linux is quite straightforward, albeit arguably not the most robust ZFS implementation. I'm doing Linux zvol's exported over SRP via SCST on CentOS to a mix of Linux and Windows initiators and it's been rock-solid for about a year now. -------------- next part -------------- An HTML attachment was scrubbed... URL: From sergei25 at gmail.com Fri Oct 24 23:14:52 2014 From: sergei25 at gmail.com (sergei) Date: Fri, 24 Oct 2014 16:14:52 -0700 Subject: [OmniOS-discuss] ZFS Logical Unit stuck in the offlining state with COMSTAR In-Reply-To: <1675854F-C90E-4153-85B0-328348C78CC7@omniti.com> References: <1675854F-C90E-4153-85B0-328348C78CC7@omniti.com> Message-ID: Does r151012 include Illumos 3464 commit ? I went to the latest r151013 and that proved to be a very radical change. We've spent a lot to tune performance and make this system work well. And with 3464 all TXG logic has changed. Sergei On Fri, Oct 24, 2014 at 2:26 PM, Dan McDonald wrote: > > > On Oct 24, 2014, at 5:22 PM, sergei wrote: > > > > I'm having same issue with OmniOS v11 r151006. Found few threads on this > including > http://lists.omniti.com/pipermail/omnios-discuss/2013-March/000537.html > and https://illumos.org/issues/3621 but still can't figure out whats > going on. Basic symptoms are that some zvol-backed LUs can not be deleted. > And stmfadm list-lu -v reports LU state as "Offlining". > > > > For what its worth we run same Dell R720xd as one of the earlier posters. > > > > I would really appreciate few tips on how to resolve this or if COMSTAR > / stmf is more reliable in some other Illumos flavor ? > > You can try r151012, which has many bugfixes in it, some of which are > COMSTAR/stmf related. > > Dan > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From nsmith at careyweb.com Fri Oct 24 23:19:04 2014 From: nsmith at careyweb.com (Nate Smith) Date: Fri, 24 Oct 2014 19:19:04 -0400 Subject: [OmniOS-discuss] =?utf-8?q?ZFS_Logical_Unit_stuck_in_the_offlinin?= =?utf-8?q?g_state_with_COMSTAR?= Message-ID: <297520921-4948@mail.careyweb.com> An HTML attachment was scrubbed... URL: -------------- next part -------------- will say I've been running into very similar comstar/fibre lu issues with r720 Sent from my Android phone using TouchDown (www.nitrodesk.com) -----Original Message----- From: sergei [sergei25 at gmail.com] Received: Friday, 24 Oct 2014, 7:15PM To: Dan McDonald [danmcd at omniti.com] CC: omnios-discuss at lists.omniti.com [omnios-discuss at lists.omniti.com] Subject: Re: [OmniOS-discuss] ZFS Logical Unit stuck in the offlining state with COMSTAR Does r151012 include Illumos 3464 commit ? I went to the latest r151013 and that proved to be a very radical change. We've spent a lot to tune performance and make this system work well. And with 3464 all TXG logic has changed. Sergei On Fri, Oct 24, 2014 at 2:26 PM, Dan McDonald wrote: > > > On Oct 24, 2014, at 5:22 PM, sergei wrote: > > > > I'm having same issue with OmniOS v11 r151006. Found few threads on this > including > http://lists.omniti.com/pipermail/omnios-discuss/2013-March/000537.html > and https://illumos.org/issues/3621 but still can't figure out whats > going on. Basic symptoms are that some zvol-backed LUs can not be deleted. > And stmfadm list-lu -v reports LU state as "Offlining". > > > > For what its worth we run same Dell R720xd as one of the earlier posters. > > > > I would really appreciate few tips on how to resolve this or if COMSTAR > / stmf is more reliable in some other Illumos flavor ? > > You can try r151012, which has many bugfixes in it, some of which are > COMSTAR/stmf related. > > Dan > > From danmcd at omniti.com Sat Oct 25 01:14:18 2014 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 24 Oct 2014 21:14:18 -0400 Subject: [OmniOS-discuss] ZFS Logical Unit stuck in the offlining state with COMSTAR In-Reply-To: References: <1675854F-C90E-4153-85B0-328348C78CC7@omniti.com> Message-ID: > On Oct 24, 2014, at 7:14 PM, sergei wrote: > > Does r151012 include Illumos 3464 commit ? > > I went to the latest r151013 and that proved to be a very radical change. We've spent a lot to tune performance and make this system work well. And with 3464 all TXG logic has changed. Actually, 006 does as well. That's a ZFS bugfix, BTW, so while it may affect COMSTAR, it's not a fix in COMSTAR proper. Dan From jeffpc at josefsipek.net Sat Oct 25 01:31:40 2014 From: jeffpc at josefsipek.net (Josef 'Jeff' Sipek) Date: Fri, 24 Oct 2014 21:31:40 -0400 Subject: [OmniOS-discuss] [developer] Initial review --> Changes that allow illumos-gate to (mostly) build on OmniOS In-Reply-To: <20141024161130.GB1317@meili> References: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> <20141024155647.GA1317@meili> <8B714042-F636-4C10-8105-3838053B1FD9@omniti.com> <20141024161130.GB1317@meili> Message-ID: <20141025013140.GA1311@meili> On Fri, Oct 24, 2014 at 12:11:31PM -0400, Josef 'Jeff' Sipek wrote: > On Fri, Oct 24, 2014 at 12:01:21PM -0400, Dan McDonald via illumos-developer wrote: > > > On Oct 24, 2014, at 11:56 AM, Josef 'Jeff' Sipek wrote: ... > > > You said OI... did you try both /dev (151a9) as well as hipster? Just > > > curious. > > > > I used /dev only (oi_151a9) for my tests. I'd LOVE it if someone took > > these and tried on hipster. > > I'm going to try to run a nightly to see what happens. FWIW, nightly seems happy when run on hipster. I haven't tried using the produced packages. Jeff. -- Computer Science is no more about computers than astronomy is about telescopes. - Edsger Dijkstra From natxo.asenjo at gmail.com Sat Oct 25 07:51:13 2014 From: natxo.asenjo at gmail.com (Natxo Asenjo) Date: Sat, 25 Oct 2014 09:51:13 +0200 Subject: [OmniOS-discuss] best guide to tftp ? In-Reply-To: <20141025065313.4b67dcbc@punda-mlia> References: <20141025065313.4b67dcbc@punda-mlia> Message-ID: On Fri, Oct 24, 2014 at 10:53 PM, Michael Mounteney wrote: > Hello, I want to use OmniOS as a boot server for various Linux thin > clients. What's the most appropriate guide out there? There's a few, > referring to obsolete problems, other OSs etc. Just tested and it works. Searching for in.tftpd # pkg search in.tftpd INDEX ACTION VALUE PACKAGE basename file usr/sbin/in.tftpd pkg:/service/network/tftp at 0.5.11-0.151012 # pkg publisher PUBLISHER TYPE STATUS URI omnios origin online http://pkg.omniti.com/omnios/r151012/ So you should just install the tftp package in the omnios publisher and then activate the tftp service by editing inetd.conf and adding the line: tftp dgram udp6 wait root /usr/sbin/in.tftpd in.tftpd -s /tftpboot and then running the "inetconv" command. This will create an SMF service, network/tftp/udp6. Create a /tftpboot dir in your omnios host and there you go. Edit the pxe files like you are used to. -- Groeten, natxo From tim at multitalents.net Sat Oct 25 19:35:21 2014 From: tim at multitalents.net (Tim Rice) Date: Sat, 25 Oct 2014 12:35:21 -0700 (PDT) Subject: [OmniOS-discuss] HP Z230 as a home server In-Reply-To: <5447E43D.5070303@gmx.li> References: <5446B314.9020600@gmx.li> <52DCA0DF-14A9-44F8-8B62-E29749D3645D@clozure.com> <5447E43D.5070303@gmx.li> Message-ID: On Wed, 22 Oct 2014, Dominik Hassler wrote: [snip] > - MBD-X10SLM-F-O mainboard > - Xeon E3-1231v3 (this time w/ HT) > - 4x 8 GB KTH-PL316EK4/32G > - 2x SSD DC S3500 (rpool mirror) > - 2x WD RE 4T (data mirror) > - Lian Li PC-V358 case [snip] > does anyone have (good or bad) experience with the kingston RAM > mentioned above? Kingston RAM didn't work on a MBD-X9SCM-F-0 board here. Had to buy new ram. I recommend you stick with whatever RAM supermicro says is tested on your board. -- Tim Rice Multitalents tim at multitalents.net From henrikj at henkis.net Sun Oct 26 10:03:09 2014 From: henrikj at henkis.net (Henrik Johansson) Date: Sun, 26 Oct 2014 11:03:09 +0100 Subject: [OmniOS-discuss] ARC, L2ARC and metadata questions Message-ID: <28626250-04CC-4A64-AE09-FC0D9B6C0115@henkis.net> Hi, I have a random read-only workload against 50M which are just above 400KB each. I have built the dataset on a set of mirrored(10 mirrors) devices for best performance but are looking into using a larger ARC or L2ARC to improve performance. I do realise that the dataset is way to large to cache, but caching metadata improves performance and latency for reading. In a test on a subset and accesses 1M files with primarycache set to metadata for the dataset, it did improve performance but used 12-18GB of RAM in the ARC (will run it again when possible forgot the exact number). Anyway, it was far to much RAM to be realistic for us to put in each node for metadata caching. I have three questions I need help with: 1. How do I calculate required ARC metadata size for a dataset, x bytes per block and file size / block size to get the number of blocks? 2. If I where to use a L2ARC for metadata I need file size / block size * no files * 200 bytes of RAM ( 400 / 128 * 50M * 200 = ~ 37GB? ) to address the larger L2ARC? 3. I have disabled atime and prefetch since it had 100% miss, are there anything else we could do to to improve performance? Thanks Henrik From richard.elling at richardelling.com Sun Oct 26 21:50:42 2014 From: richard.elling at richardelling.com (Richard Elling) Date: Sun, 26 Oct 2014 14:50:42 -0700 Subject: [OmniOS-discuss] ARC, L2ARC and metadata questions In-Reply-To: <28626250-04CC-4A64-AE09-FC0D9B6C0115@henkis.net> References: <28626250-04CC-4A64-AE09-FC0D9B6C0115@henkis.net> Message-ID: On Oct 26, 2014, at 3:03 AM, Henrik Johansson wrote: > Hi, > > I have a random read-only workload against 50M which are just above 400KB each. I have built the dataset on a set of mirrored(10 mirrors) devices for best performance but are looking into using a larger ARC or L2ARC to improve performance. I do realise that the dataset is way to large to cache, but caching metadata improves performance and latency for reading. [I presume you mean quantity 50M of files, each just above 400KB in size] > > In a test on a subset and accesses 1M files with primarycache set to metadata for the dataset, it did improve performance but used 12-18GB of RAM in the ARC (will run it again when possible forgot the exact number). Anyway, it was far to much RAM to be realistic for us to put in each node for metadata caching. You'll need to look at arc_meta_used, not the (ARC) size. For convenience, the high water mark of arc_meta_used is stored in arc_meta_max. For example, on a smallish VM I have: # kstat -p ::arcstats:arc_meta_* zfs:0:arcstats:arc_meta_limit 1202388480 zfs:0:arcstats:arc_meta_max 48319112 zfs:0:arcstats:arc_meta_used 48273456 NB, for OmniOS/illumos, arc_meta_limit is not a hard limit -- it is a decision point above which the eviction algorithm changes priority. > > I have three questions I need help with: > > 1. How do I calculate required ARC metadata size for a dataset, x bytes per block and file size / block size to get the number of blocks? It is much easier to measure. > > 2. If I where to use a L2ARC for metadata I need file size / block size * no files * 200 bytes of RAM ( 400 / 128 * 50M * 200 = ~ 37GB? ) to address the larger L2ARC? This is a good starting point. Recall that this is a cache, so it depends on your working set size. Again, this is easier to measure the size of the headers used. The "200 bytes" can vary based on OS release and whether you are using a debug kernel (non-debug kernels are smaller) kstat -p ::arcstats:l2_hdr_size > > 3. I have disabled atime and prefetch since it had 100% miss, are there anything else we could do to to improve performance? Disabling data prefetch doesn't impact metadata prefetch. You can measure both. DNLC is probably too small, so you should look at the DNLC hit rate, too. Note that when you get above the arc_meta_limit, DLNC gets reduced, which is probably not what you want. So if your DNLC hit rate looks poor (< 90%) then consider increasing both the DNLC size and arc_meta_limit. -- richard > > Thanks > Henrik > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From alp at rsu.ru Sat Oct 25 18:22:30 2014 From: alp at rsu.ru (Alexander Pyhalov) Date: Sat, 25 Oct 2014 22:22:30 +0400 Subject: [OmniOS-discuss] [developer] Initial review --> Changes that allow illumos-gate to (mostly) build on OmniOS In-Reply-To: <8B714042-F636-4C10-8105-3838053B1FD9@omniti.com> References: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> <20141024155647.GA1317@meili> <8B714042-F636-4C10-8105-3838053B1FD9@omniti.com> Message-ID: Hello. Dan McDonald via illumos-developer ????? 24.10.2014 20:01: >> dnssd/Makefile >> libdtrace_jni/java/Makefile >> I'm curious about the details here... Javadoc uses different >> filenames for the generated gifs and this allows packaging to work >> since it'll see both old and new files? Am I understanding >> correctly? Would it make sense to document this hack in the package >> manifest as well for when someone decides that 1.6 is ancient? > > openjdk7's javadoc generates other files, yes. This hack does allow > 1.6 and 1.7 to both work. I probably should document it somewhere. I > can't remember if pkg(5) files can take comments in the middle or not. If you speak about comments in the middle of the manifest, of course, you can use them. --- System Administrator of Southern Federal University Computer Center From yuri.pankov at nexenta.com Sat Oct 25 20:23:18 2014 From: yuri.pankov at nexenta.com (Yuri Pankov) Date: Sun, 26 Oct 2014 00:23:18 +0400 Subject: [OmniOS-discuss] [developer] Initial review --> Changes that allow illumos-gate to (mostly) build on OmniOS In-Reply-To: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> References: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> Message-ID: <544C06B6.8000900@nexenta.com> On Thu, 23 Oct 2014 21:56:30 -0400, Illumos-developer wrote: > Hello folks from two lists! > > I've been tweaking around with illumos-gate on OmniOS again. And this time, I have it to where: > > * You can still build with NO NOISE on OpenIndiana - i.e. it's a NOP to OI builders. > > * You can *build* and *package* on OmniOS, but there are still 2-3 ELF runtime check hits, and some can't-help-it lint. > > The "can't-help-it-lint" is because I either generate lint errors on OI due to not having lint-bypass comments, or noise on OmniOS because I DO have lint-bypass comments. I chose to have the OmniOS people cope with the noise. > > I need 8 commits (all independent of each other) to go into illumos-gate to make what I just mentioned above happen. They are here: > > http://kebe.com/~danmcd/webrevs/for-omnios/ > > Some of these may have value ANYWAY for upstream, the rest are modifications that do not affect an OI build, but help an OmniOS one. The 8 commits are: > > * Some check_rtime modifications for the version of things in OmniOS > * Updates to allow buildings with openjdk7, but STILL ALLOW jdk1.6 that's on OI. > * Lose the IPP listener if so specified in your .env file, which depends on bits not in OmniOS. Can we "lose" it entirely, dropping the apache1 dependency and not adding extra cruft in the tree? > * Lose the dependency on docbooks, only existing in the Shell Style Guide. This looks like another candidate to simply disappear, the entire documentation subset, which is NOT maintained by illumos-gate developers. > * Lose the SMF dependency on sysidtool. > * Allow linkage with more recent versions of net-snmp. > * Lint fixes that the OmniOS sunstudio12.1 lint are picky about. > * More perl fixes to work with OmniOS's alternative approach to perl support. > > Additionally, there's a sample .env file that anyone who applies these fixes can use on their OmniOS build machine. I also have a mail_msg from an OmniOS build, to demonstrate what errors are still there in the build even AFTER these changes. > > I'd like both the illumos developer community and the OmniOS community to look, and share their reviews and opinions. If you have a distro that's not OI or OmniOS, see if these mods help things work on YOUR distro. If I can get more non-OI distros to build or mostly-build stock illumos-gate, it's better for everyone. From mayuresh at kathe.in Mon Oct 27 13:13:25 2014 From: mayuresh at kathe.in (Mayuresh Kathe) Date: Mon, 27 Oct 2014 18:43:25 +0530 Subject: [OmniOS-discuss] =?utf-8?q?Compiling_GNUCOBOL_=3A_pre-requisites_?= =?utf-8?q?present=3F?= Message-ID: <85f940caa925131c4e59fe2b99c0ae89@kathe.in> Hello, Would the following be present under OmniOS or be easily installable? GNU MP (libgmp) 4.1.2 or later GNU Libtool (libltdl) Berkeley DB (libdb) 1.85 or later Ncurses (ncurses) 5.2 or later Unix curses Thanks, ~Mayuresh From ryan at zinascii.com Mon Oct 27 15:44:18 2014 From: ryan at zinascii.com (Ryan Zezeski) Date: Mon, 27 Oct 2014 11:44:18 -0400 Subject: [OmniOS-discuss] Compiling GNUCOBOL : pre-requisites present? In-Reply-To: <85f940caa925131c4e59fe2b99c0ae89@kathe.in> References: <85f940caa925131c4e59fe2b99c0ae89@kathe.in> Message-ID: <6F3F930C-13BC-4459-B212-7A32EAA757F4@zinascii.com> > On Oct 27, 2014, at 9:13 AM, Mayuresh Kathe wrote: > > GNU MP (libgmp) 4.1.2 or later > GNU Libtool (libltdl) > Berkeley DB (libdb) 1.85 or later > Ncurses (ncurses) 5.2 or later > Unix curses > From what I can tell, there are packages for all of these on the base system except for BDB. For BDB it looks like you'll need to add the ms.omniti.com publisher. http://omnios.omniti.com/wiki.php/Packaging Keep in mind that the OmniOS folk encourage you to build your own application stack so that you can own it and tailor it to your specific needs. Because of this philosophy they keep the base system as thin as possible. http://omnios.omniti.com/wiki.php/KYSTY -Z From mayuresh at kathe.in Mon Oct 27 16:03:53 2014 From: mayuresh at kathe.in (Mayuresh Kathe) Date: Mon, 27 Oct 2014 21:33:53 +0530 Subject: [OmniOS-discuss] Compiling GNUCOBOL : pre-requisites present? In-Reply-To: <6F3F930C-13BC-4459-B212-7A32EAA757F4@zinascii.com> References: <85f940caa925131c4e59fe2b99c0ae89@kathe.in> <6F3F930C-13BC-4459-B212-7A32EAA757F4@zinascii.com> Message-ID: <20141027160351.GA10223@hp-aio> On Mon, Oct 27, 2014 at 11:44:18AM -0400, Ryan Zezeski wrote: > > > On Oct 27, 2014, at 9:13 AM, Mayuresh Kathe wrote: > > > > GNU MP (libgmp) 4.1.2 or later > > GNU Libtool (libltdl) > > Berkeley DB (libdb) 1.85 or later > > Ncurses (ncurses) 5.2 or later > > Unix curses > > > > From what I can tell, there are packages for all of these on the base system except for BDB. For BDB it looks like you'll need to add the ms.omniti.com publisher. > > http://omnios.omniti.com/wiki.php/Packaging > > Keep in mind that the OmniOS folk encourage you to build your own application stack so that you can own it and tailor it to your specific needs. Because of this philosophy they keep the base system as thin as possible. > > http://omnios.omniti.com/wiki.php/KYSTY Thanks Ryan, will look into the ms.omniti.com publisher adding process. Yes, I am aware of the OmniOS philosophy, that's why I love the work they are doing. :) BTW, if I do end up getting GNUCOBOL to work and work well, I would have saved a rural community in India a lot of money by preventing them from going for an IBM zSeries system. Most of the functionality already seems available with Illumos based systems. Thanks again for the tip. :) ~Mayuresh From mayuresh at kathe.in Mon Oct 27 16:40:16 2014 From: mayuresh at kathe.in (Mayuresh Kathe) Date: Mon, 27 Oct 2014 22:10:16 +0530 Subject: [OmniOS-discuss] Compiling GNUCOBOL : pre-requisites present? In-Reply-To: <6F3F930C-13BC-4459-B212-7A32EAA757F4@zinascii.com> References: <85f940caa925131c4e59fe2b99c0ae89@kathe.in> <6F3F930C-13BC-4459-B212-7A32EAA757F4@zinascii.com> Message-ID: <20141027164015.GA17098@hp-aio> On Mon, Oct 27, 2014 at 11:44:18AM -0400, Ryan Zezeski wrote: > > > On Oct 27, 2014, at 9:13 AM, Mayuresh Kathe wrote: > > > > GNU MP (libgmp) 4.1.2 or later > > GNU Libtool (libltdl) > > Berkeley DB (libdb) 1.85 or later > > Ncurses (ncurses) 5.2 or later > > Unix curses > > > > From what I can tell, there are packages for all of these on the base system except for BDB. For BDB it looks like you'll need to add the ms.omniti.com publisher. > > http://omnios.omniti.com/wiki.php/Packaging I did what you'd suggested, got BDB installed, but somehow the configure script is unable to find "gmp.h" at "/usr/include/gmp/". Is there anything I need to do to make the system know that gmp.h exists at "/usr/include/gmp/"? Thanks, ~Mayuresh From danmcd at omniti.com Mon Oct 27 19:27:02 2014 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 27 Oct 2014 15:27:02 -0400 Subject: [OmniOS-discuss] Small Security Update -> wget (CVE-2014-4877) Message-ID: <838219AB-23FE-497E-A244-DCCAB86E85F7@omniti.com> wget was updated to version 1.16 today, patching a small security vulnerability. All supported repos (006/LTS, 010, 012/stable, bloody) are now updated with this. It's not a reboot-worthy update, but it will create a backup BE. Please utter "pkg update" on your OmniOS boxes at your earliest convenience. Happy patching! Dan From mayuresh at kathe.in Mon Oct 27 19:40:26 2014 From: mayuresh at kathe.in (Mayuresh Kathe) Date: Tue, 28 Oct 2014 01:10:26 +0530 Subject: [OmniOS-discuss] Small Security Update -> wget (CVE-2014-4877) In-Reply-To: <838219AB-23FE-497E-A244-DCCAB86E85F7@omniti.com> References: <838219AB-23FE-497E-A244-DCCAB86E85F7@omniti.com> Message-ID: <20141027194025.GA632@hp-aio> Done. Thanks for the heads up Dan, really appreciate your work. ~Mayuresh On Mon, Oct 27, 2014 at 03:27:02PM -0400, Dan McDonald wrote: > wget was updated to version 1.16 today, patching a small security vulnerability. > > All supported repos (006/LTS, 010, 012/stable, bloody) are now updated with this. It's not a reboot-worthy update, but it will create a backup BE. > > Please utter "pkg update" on your OmniOS boxes at your earliest convenience. > > Happy patching! > Dan > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From danmcd at omniti.com Mon Oct 27 19:55:51 2014 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 27 Oct 2014 15:55:51 -0400 Subject: [OmniOS-discuss] [developer] Initial review --> Changes that allow illumos-gate to (mostly) build on OmniOS In-Reply-To: <544C06B6.8000900@nexenta.com> References: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> <544C06B6.8000900@nexenta.com> Message-ID: <1631C904-99F2-4D50-9437-D6F2B59FA31C@omniti.com> Pardon the latency on this one, Yuri. > On Oct 25, 2014, at 4:23 PM, Yuri Pankov wrote: >> * Lose the IPP listener if so specified in your .env file, which depends on bits not in OmniOS. > > Can we "lose" it entirely, dropping the apache1 dependency and not adding extra cruft in the tree? I wouldn't be opposed to the wholesale destruction of the IPP listener from illumos-gate. Would need a large crowd with pitchforks & torches, though. >> * Lose the dependency on docbooks, only existing in the Shell Style Guide. > > This looks like another candidate to simply disappear, the entire documentation subset, which is NOT maintained by illumos-gate developers. This one was a far easier fix, and nothing else DOES depend on docbooks. I'm reluctant to go further here. Dan From mayuresh at kathe.in Mon Oct 27 19:59:06 2014 From: mayuresh at kathe.in (Mayuresh Kathe) Date: Tue, 28 Oct 2014 01:29:06 +0530 Subject: [OmniOS-discuss] include path : is there such a thing? Message-ID: <20141027195905.GA5911@hp-aio> Hello, Just the way there's a PATH environment variable and an optional MANPATH one too, is there anything for "includes" also? I am really getting busted trying to get GnuCOBOL compiled under OmniOS. All of the extra stuff I have installed has gone and sat down in it's own structure at various places /usr/local/include/..., /opt/omni/include/..., etc. Can I please be advised on the best way forward such that the "configure" script of GnuCOBOL finds all the headers? Thanks, ~Mayuresh From henson at acm.org Mon Oct 27 20:03:50 2014 From: henson at acm.org (Paul B. Henson) Date: Mon, 27 Oct 2014 13:03:50 -0700 Subject: [OmniOS-discuss] include path : is there such a thing? In-Reply-To: <20141027195905.GA5911@hp-aio> References: <20141027195905.GA5911@hp-aio> Message-ID: <20141027200350.GI20693@bender.unx.csupomona.edu> On Tue, Oct 28, 2014 at 01:29:06AM +0530, Mayuresh Kathe wrote: > Just the way there's a PATH environment variable and an optional MANPATH > one too, is there anything for "includes" also? https://gcc.gnu.org/onlinedocs/gcc/Environment-Variables.html C_INCLUDE_PATH CPLUS_INCLUDE_PATH From mayuresh at kathe.in Mon Oct 27 20:22:41 2014 From: mayuresh at kathe.in (Mayuresh Kathe) Date: Tue, 28 Oct 2014 01:52:41 +0530 Subject: [OmniOS-discuss] include path : is there such a thing? In-Reply-To: <20141027200350.GI20693@bender.unx.csupomona.edu> References: <20141027195905.GA5911@hp-aio> <20141027200350.GI20693@bender.unx.csupomona.edu> Message-ID: <20141027202240.GA11462@hp-aio> On Mon, Oct 27, 2014 at 01:03:50PM -0700, Paul B. Henson wrote: > On Tue, Oct 28, 2014 at 01:29:06AM +0530, Mayuresh Kathe wrote: > > > Just the way there's a PATH environment variable and an optional MANPATH > > one too, is there anything for "includes" also? > > https://gcc.gnu.org/onlinedocs/gcc/Environment-Variables.html > > C_INCLUDE_PATH > CPLUS_INCLUDE_PATH Thanks, it worked, but not well enough. Now the "configure" script tells me about not finding libraries. Have set the LIBRARY_PATH variable as mentioned in the page at the link you'd sent. No use. Any more suggestions? Thanks, ~Mayuresh From mayuresh at kathe.in Mon Oct 27 20:29:23 2014 From: mayuresh at kathe.in (Mayuresh Kathe) Date: Tue, 28 Oct 2014 01:59:23 +0530 Subject: [OmniOS-discuss] include path : is there such a thing? In-Reply-To: <20141027202240.GA11462@hp-aio> References: <20141027195905.GA5911@hp-aio> <20141027200350.GI20693@bender.unx.csupomona.edu> <20141027202240.GA11462@hp-aio> Message-ID: <20141027202922.GA19908@hp-aio> On Tue, Oct 28, 2014 at 01:52:41AM +0530, Mayuresh Kathe wrote: > On Mon, Oct 27, 2014 at 01:03:50PM -0700, Paul B. Henson wrote: > > On Tue, Oct 28, 2014 at 01:29:06AM +0530, Mayuresh Kathe wrote: > > > > > Just the way there's a PATH environment variable and an optional MANPATH > > > one too, is there anything for "includes" also? > > > > https://gcc.gnu.org/onlinedocs/gcc/Environment-Variables.html > > > > C_INCLUDE_PATH > > CPLUS_INCLUDE_PATH > > Thanks, it worked, but not well enough. > Now the "configure" script tells me about not finding libraries. > Have set the LIBRARY_PATH variable as mentioned in the page at the link > you'd sent. > No use. > Any more suggestions? Okay, got it working by manually copying the libdb.x files from /opt/omni/lib to /lib directory. But somehow, it doesn't feel like a proper/elegant solution. Best, ~Mayuresh From mir at miras.org Mon Oct 27 21:35:52 2014 From: mir at miras.org (Michael Rasmussen) Date: Mon, 27 Oct 2014 22:35:52 +0100 Subject: [OmniOS-discuss] include path : is there such a thing? In-Reply-To: <20141027202922.GA19908@hp-aio> References: <20141027195905.GA5911@hp-aio> <20141027200350.GI20693@bender.unx.csupomona.edu> <20141027202240.GA11462@hp-aio> <20141027202922.GA19908@hp-aio> Message-ID: <20141027223552.2ad690a9@sleipner.datanom.net> On Tue, 28 Oct 2014 01:59:23 +0530 Mayuresh Kathe wrote: > > Okay, got it working by manually copying the libdb.x files from > /opt/omni/lib to /lib directory. > > But somehow, it doesn't feel like a proper/elegant solution. > Instead of copying why not make links? On same file system: ln /opt/omni/lib/whatever.so /lib On different file systems: ln -s /opt/omni/lib/whatever.so /lib -- Hilsen/Regards Michael Rasmussen Get my public GnuPG keys: michael rasmussen cc http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E mir datanom net http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C mir miras org http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 -------------------------------------------------------------- /usr/games/fortune -es says: "Who is General Failure and why is he reading my hard disk ?" Microsoft spel chekar vor sail, worgs grate !! (By leitner at inf.fu-berlin.de, Felix von Leitner) -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: OpenPGP digital signature URL: From jesus at omniti.com Mon Oct 27 21:52:18 2014 From: jesus at omniti.com (Theo Schlossnagle) Date: Mon, 27 Oct 2014 22:52:18 +0100 Subject: [OmniOS-discuss] include path : is there such a thing? In-Reply-To: <20141027223552.2ad690a9@sleipner.datanom.net> References: <20141027195905.GA5911@hp-aio> <20141027200350.GI20693@bender.unx.csupomona.edu> <20141027202240.GA11462@hp-aio> <20141027202922.GA19908@hp-aio> <20141027223552.2ad690a9@sleipner.datanom.net> Message-ID: Just add LDFLAGS="-L/opt/omni/lib -R/opt/omni/lib" to the arguments of configure. On Mon, Oct 27, 2014 at 10:35 PM, Michael Rasmussen wrote: > On Tue, 28 Oct 2014 01:59:23 +0530 > Mayuresh Kathe wrote: > > > > > Okay, got it working by manually copying the libdb.x files from > > /opt/omni/lib to /lib directory. > > > > But somehow, it doesn't feel like a proper/elegant solution. > > > Instead of copying why not make links? > > On same file system: > ln /opt/omni/lib/whatever.so /lib > > On different file systems: > ln -s /opt/omni/lib/whatever.so /lib > > -- > Hilsen/Regards > Michael Rasmussen > > Get my public GnuPG keys: > michael rasmussen cc > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xD3C9A00E > mir datanom net > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE501F51C > mir miras org > http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xE3E80917 > -------------------------------------------------------------- > /usr/games/fortune -es says: > "Who is General Failure and why is he reading my hard disk ?" > Microsoft spel chekar vor sail, worgs grate !! > (By leitner at inf.fu-berlin.de, Felix von Leitner) > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > -- Theo Schlossnagle http://omniti.com/is/theo-schlossnagle -------------- next part -------------- An HTML attachment was scrubbed... URL: From bfriesen at simple.dallas.tx.us Tue Oct 28 01:48:53 2014 From: bfriesen at simple.dallas.tx.us (Bob Friesenhahn) Date: Mon, 27 Oct 2014 20:48:53 -0500 (CDT) Subject: [OmniOS-discuss] include path : is there such a thing? In-Reply-To: <20141027202922.GA19908@hp-aio> References: <20141027195905.GA5911@hp-aio> <20141027200350.GI20693@bender.unx.csupomona.edu> <20141027202240.GA11462@hp-aio> <20141027202922.GA19908@hp-aio> Message-ID: On Tue, 28 Oct 2014, Mayuresh Kathe wrote: > Okay, got it working by manually copying the libdb.x files from > /opt/omni/lib to /lib directory. > > But somehow, it doesn't feel like a proper/elegant solution. Ugh! For my own purposes I use a config.site file (e.g. /usr/local/share/config.site) which supplies autoconf (`configure') defaults for a given installation prefix tree. I make sure that it injects suitable LDFLAGS (e.g. LDFLAGS="-L/usr/local/lib -R/usr/local/lib") into the configure script. The required options depend on the compiler's defaults. The above works well (like magic!) for the typical package with a configure script but not likely for packages depending on something like cmake. This approach allows providing defaults on a per installation prefix basis so different installation prefixes can use different compilation options and libraries. I am happy to share ... Bob -- Bob Friesenhahn bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ From ryan at zinascii.com Tue Oct 28 01:50:44 2014 From: ryan at zinascii.com (Ryan Zezeski) Date: Mon, 27 Oct 2014 21:50:44 -0400 Subject: [OmniOS-discuss] [developer] Initial review --> Changes that allow illumos-gate to (mostly) build on OmniOS In-Reply-To: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> References: <18BEBA50-C025-4C48-ABBD-97D3DA8D6C24@omniti.com> Message-ID: > On Oct 23, 2014, at 9:56 PM, Dan McDonald via illumos-developer wrote: > > * You can still build with NO NOISE on OpenIndiana - i.e. it's a NOP to OI builders. > > > I'd like both the illumos developer community and the OmniOS community to look, and share their reviews and opinions. If you have a distro that's not OI or OmniOS, see if these mods help things work on YOUR distro. If I can get more non-OI distros to build or mostly-build stock illumos-gate, it's better for everyone. > I'm very excited to see changes that make the build easier across distributions, thank you Dan. I did not review any of the webrevs but I did test the build on my OI 151a8 node. It came out clean with no modifications to my environment. Attached is my mail_msg. -Z -------------- next part -------------- A non-text attachment was scrubbed... Name: mail_msg.gz Type: application/x-gzip Size: 861 bytes Desc: not available URL: -------------- next part -------------- From bfriesen at simple.dallas.tx.us Tue Oct 28 01:52:11 2014 From: bfriesen at simple.dallas.tx.us (Bob Friesenhahn) Date: Mon, 27 Oct 2014 20:52:11 -0500 (CDT) Subject: [OmniOS-discuss] include path : is there such a thing? In-Reply-To: <20141027195905.GA5911@hp-aio> References: <20141027195905.GA5911@hp-aio> Message-ID: On Tue, 28 Oct 2014, Mayuresh Kathe wrote: > Hello, > > Just the way there's a PATH environment variable and an optional MANPATH > one too, is there anything for "includes" also? > > I am really getting busted trying to get GnuCOBOL compiled under OmniOS. > All of the extra stuff I have installed has gone and sat down in it's > own structure at various places /usr/local/include/..., > /opt/omni/include/..., etc. > > Can I please be advised on the best way forward such that the > "configure" script of GnuCOBOL finds all the headers? export CPPFLAGS='-I /usr/local/include -I/opt/omni/include' Then run 'configure'. You likely also want to set LIBS and LDFLAGS. See if there is a file called INSTALL in the package which describes the various options. Bob -- Bob Friesenhahn bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ From jorge at blackdot.be Mon Oct 27 19:42:26 2014 From: jorge at blackdot.be (Jorge Schrauwen) Date: Mon, 27 Oct 2014 20:42:26 +0100 Subject: [OmniOS-discuss] Small Security Update -> wget (CVE-2014-4877) In-Reply-To: <20141027194025.GA632@hp-aio> References: <838219AB-23FE-497E-A244-DCCAB86E85F7@omniti.com> <20141027194025.GA632@hp-aio> Message-ID: <1c0248a3010f5d43c77fc27d121a3d26@blackdot.be> Same, just a quick wand^Wwave^w oneliner and eveything is up to date. --- ~ sjorge On 2014-10-27 20:40, Mayuresh Kathe wrote: > Done. Thanks for the heads up Dan, really appreciate your work. > > ~Mayuresh > > On Mon, Oct 27, 2014 at 03:27:02PM -0400, Dan McDonald wrote: >> wget was updated to version 1.16 today, patching a small security >> vulnerability. >> >> All supported repos (006/LTS, 010, 012/stable, bloody) are now updated >> with this. It's not a reboot-worthy update, but it will create a >> backup BE. >> >> Please utter "pkg update" on your OmniOS boxes at your earliest >> convenience. >> >> Happy patching! >> Dan >> >> _______________________________________________ >> OmniOS-discuss mailing list >> OmniOS-discuss at lists.omniti.com >> http://lists.omniti.com/mailman/listinfo/omnios-discuss > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From chip at innovates.com Tue Oct 28 19:18:41 2014 From: chip at innovates.com (Schweiss, Chip) Date: Tue, 28 Oct 2014 14:18:41 -0500 Subject: [OmniOS-discuss] OmniOS "bloody" now updated. In-Reply-To: <3736D518-8A4F-41E7-9B5A-4B8BF92A0699@omniti.com> References: <3736D518-8A4F-41E7-9B5A-4B8BF92A0699@omniti.com> Message-ID: I went to grab the latest Bloody today, the download links seems to be broken on this page: http://omnios.omniti.com/wiki.php/Installation Is there a better link? -Chip On Tue, Oct 21, 2014 at 12:17 AM, Dan McDonald wrote: > Pardon the delay. I was waiting for one of the new features to integrate > upstream before I updated bloody. So this one is not-quite three weeks > later. I plan on resuming biweekly updates of bloody on a Thursday +/- 1 > day of an update week. That means the next bloody update after this SHOULD > be out in early November. > > I did a full sweeping package update, so you may need to "pkg update pkg" > prior to this update. > > New with this release (r151013-20141021): > > - omnios-build master branch, revision 2412b2d. > > - open-vm-tools from Delphix > > - illumos-omnios master branch, revision 09bdbed. > > - Open-source math library. (This caused the delay. NOTE: > library/math/header-math is now gone. Any dependencies upon that should be > replaced with library/math.) > > - pam_timestamp option for PAM (to allow sudo-like last-login caching, > handle with care). > > - Some ZFS fixes and improvments. > > - Replaceable sendmail packages (uses IPS mediators). > > - sTec S112x PCIe SSD support. > > > I've also cut new install media as well, AND there's a new kayak .zfs.bz2 > in place. > > Thanks! > Dan > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Tue Oct 28 19:22:45 2014 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 28 Oct 2014 15:22:45 -0400 Subject: [OmniOS-discuss] OmniOS "bloody" now updated. In-Reply-To: References: <3736D518-8A4F-41E7-9B5A-4B8BF92A0699@omniti.com> Message-ID: <3B62C6F2-CF8A-4094-9959-ADD4C5F3EDC1@omniti.com> > On Oct 28, 2014, at 3:18 PM, Schweiss, Chip wrote: > > I went to grab the latest Bloody today, the download links seems to be broken on this page: http://omnios.omniti.com/wiki.php/Installation > > Is there a better link? Hang on. Investigating now. Dan From danmcd at omniti.com Tue Oct 28 19:24:45 2014 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 28 Oct 2014 15:24:45 -0400 Subject: [OmniOS-discuss] OmniOS "bloody" now updated. In-Reply-To: <3B62C6F2-CF8A-4094-9959-ADD4C5F3EDC1@omniti.com> References: <3736D518-8A4F-41E7-9B5A-4B8BF92A0699@omniti.com> <3B62C6F2-CF8A-4094-9959-ADD4C5F3EDC1@omniti.com> Message-ID: <51098692-FE5A-444C-84B3-512E50682265@omniti.com> > On Oct 28, 2014, at 3:22 PM, Dan McDonald wrote: > > >> On Oct 28, 2014, at 3:18 PM, Schweiss, Chip wrote: >> >> I went to grab the latest Bloody today, the download links seems to be broken on this page: http://omnios.omniti.com/wiki.php/Installation >> >> Is there a better link? > > Hang on. Investigating now. Try now, please? I'd improperly named them after uploading. Odd it took until now for someone to notice. :) Sorry about that! Dan From chip at innovates.com Tue Oct 28 19:25:54 2014 From: chip at innovates.com (Schweiss, Chip) Date: Tue, 28 Oct 2014 14:25:54 -0500 Subject: [OmniOS-discuss] OmniOS "bloody" now updated. In-Reply-To: <51098692-FE5A-444C-84B3-512E50682265@omniti.com> References: <3736D518-8A4F-41E7-9B5A-4B8BF92A0699@omniti.com> <3B62C6F2-CF8A-4094-9959-ADD4C5F3EDC1@omniti.com> <51098692-FE5A-444C-84B3-512E50682265@omniti.com> Message-ID: That did it. Thanks! -Chip On Tue, Oct 28, 2014 at 2:24 PM, Dan McDonald wrote: > > > On Oct 28, 2014, at 3:22 PM, Dan McDonald wrote: > > > > > >> On Oct 28, 2014, at 3:18 PM, Schweiss, Chip wrote: > >> > >> I went to grab the latest Bloody today, the download links seems to be > broken on this page: http://omnios.omniti.com/wiki.php/Installation > >> > >> Is there a better link? > > > > Hang on. Investigating now. > > Try now, please? I'd improperly named them after uploading. Odd it took > until now for someone to notice. :) > > Sorry about that! > Dan > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Tue Oct 28 19:27:01 2014 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 28 Oct 2014 15:27:01 -0400 Subject: [OmniOS-discuss] OmniOS "bloody" now updated. In-Reply-To: References: <3736D518-8A4F-41E7-9B5A-4B8BF92A0699@omniti.com> <3B62C6F2-CF8A-4094-9959-ADD4C5F3EDC1@omniti.com> <51098692-FE5A-444C-84B3-512E50682265@omniti.com> Message-ID: <785ECE05-F6E1-44A3-896A-4719C454D3A2@omniti.com> > On Oct 28, 2014, at 3:25 PM, Schweiss, Chip wrote: > > That did it. Thanks! You're welcome. Again, sorry about that. Dan From danmcd at omniti.com Tue Oct 28 20:03:23 2014 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 28 Oct 2014 16:03:23 -0400 Subject: [OmniOS-discuss] Re-ping --> Any Sun DHCP server users out there? Message-ID: <448C97CD-5DA0-4D2B-9526-34C5D19B22A0@omniti.com> There's a good chance the Sun DHCP server is disappearing upstream in illumos-gate very soon. Are there any current OmniOS users who are still using the Sun DHCP server? IF SO, please speak up now. You have a dual-DHCP server release in r151012, and there's a chance Sun DHCP will disappear in the next stable, which is also the next LTS, r151014. Thanks, Dan From chip at innovates.com Tue Oct 28 21:00:48 2014 From: chip at innovates.com (Schweiss, Chip) Date: Tue, 28 Oct 2014 16:00:48 -0500 Subject: [OmniOS-discuss] Installing on VMware Workstation Message-ID: I've been regularly using OmniOS on ESXi since r151006 without issue. I'm stuck trying to install on VMware Workstation 10 on Windows 7 64bit. At the "Welcome to OmniOS" screen it dies after pressing F2 and returns to installation menu every time. I've tried telling VMware it's Solaris 11/64, Solaris 10/64, Generic 64. Tried different SCSI adapters, SATA, etc. Nothing changes. I've tried r151008t, r151012, bloody. All the same. Solaris 11.2 installs, OpenIndiana 151a8 installs. Any one know the trick to get OmniOS on VMware Workstation? -Chip -------------- next part -------------- An HTML attachment was scrubbed... URL: From admdwrf at gmail.com Tue Oct 28 21:12:03 2014 From: admdwrf at gmail.com (=?UTF-8?B?RnLDqWTDqXJpYyBBbGl4?=) Date: Tue, 28 Oct 2014 22:12:03 +0100 Subject: [OmniOS-discuss] Installing on VMware Workstation In-Reply-To: References: Message-ID: Hi, Try to add virtual floppy drive. It should work. Fred 2014-10-28 22:00 GMT+01:00 Schweiss, Chip : > I've been regularly using OmniOS on ESXi since r151006 without issue. > I'm stuck trying to install on VMware Workstation 10 on Windows 7 64bit. > > At the "Welcome to OmniOS" screen it dies after pressing F2 and returns to > installation menu every time. > > I've tried telling VMware it's Solaris 11/64, Solaris 10/64, Generic 64. > Tried different SCSI adapters, SATA, etc. Nothing changes. I've tried > r151008t, r151012, bloody. All the same. > > Solaris 11.2 installs, OpenIndiana 151a8 installs. > > Any one know the trick to get OmniOS on VMware Workstation? > > -Chip > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From admdwrf at gmail.com Tue Oct 28 21:14:09 2014 From: admdwrf at gmail.com (=?UTF-8?B?RnLDqWTDqXJpYyBBbGl4?=) Date: Tue, 28 Oct 2014 22:14:09 +0100 Subject: [OmniOS-discuss] Installing on VMware Workstation In-Reply-To: References: Message-ID: Sorry, i forget the doc link http://omnios.omniti.com/wiki.php/VMwareNotes 2014-10-28 22:12 GMT+01:00 Fr?d?ric Alix : > Hi, > > Try to add virtual floppy drive. > It should work. > > Fred > > 2014-10-28 22:00 GMT+01:00 Schweiss, Chip : > >> I've been regularly using OmniOS on ESXi since r151006 without issue. >> I'm stuck trying to install on VMware Workstation 10 on Windows 7 64bit. >> >> At the "Welcome to OmniOS" screen it dies after pressing F2 and returns >> to installation menu every time. >> >> I've tried telling VMware it's Solaris 11/64, Solaris 10/64, Generic >> 64. Tried different SCSI adapters, SATA, etc. Nothing changes. I've >> tried r151008t, r151012, bloody. All the same. >> >> Solaris 11.2 installs, OpenIndiana 151a8 installs. >> >> Any one know the trick to get OmniOS on VMware Workstation? >> >> -Chip >> >> _______________________________________________ >> OmniOS-discuss mailing list >> OmniOS-discuss at lists.omniti.com >> http://lists.omniti.com/mailman/listinfo/omnios-discuss >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Tue Oct 28 21:15:01 2014 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 28 Oct 2014 17:15:01 -0400 Subject: [OmniOS-discuss] Installing on VMware Workstation In-Reply-To: References: Message-ID: > On Oct 28, 2014, at 5:00 PM, Schweiss, Chip wrote: > > I've been regularly using OmniOS on ESXi since r151006 without issue. I'm stuck trying to install on VMware Workstation 10 on Windows 7 64bit. > > At the "Welcome to OmniOS" screen it dies after pressing F2 and returns to installation menu every time. You get the installation menu, though? How much mucking around can you do with the shell you get form the installation menu? See if anything there (format(1M), zpool create) locks up or fails oddly. Dan From danmcd at omniti.com Tue Oct 28 21:15:38 2014 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 28 Oct 2014 17:15:38 -0400 Subject: [OmniOS-discuss] Installing on VMware Workstation In-Reply-To: References: Message-ID: > On Oct 28, 2014, at 5:12 PM, Fr?d?ric Alix wrote: > > Hi, > > Try to add virtual floppy drive. > It should work. > > Fred DOH! Yeah, that's right. Here's a wiki entry to help: http://omnios.omniti.com/wiki.php/VMwareNotes Dan From henrikj at henkis.net Tue Oct 28 21:18:15 2014 From: henrikj at henkis.net (Henrik Johansson) Date: Tue, 28 Oct 2014 22:18:15 +0100 Subject: [OmniOS-discuss] ARC, L2ARC and metadata questions In-Reply-To: References: <28626250-04CC-4A64-AE09-FC0D9B6C0115@henkis.net> Message-ID: > On 26 Oct 2014, at 22:50, Richard Elling wrote: > > > On Oct 26, 2014, at 3:03 AM, Henrik Johansson wrote: > >> Hi, >> >> I have a random read-only workload against 50M which are just above 400KB each. I have built the dataset on a set of mirrored(10 mirrors) devices for best performance but are looking into using a larger ARC or L2ARC to improve performance. I do realise that the dataset is way to large to cache, but caching metadata improves performance and latency for reading. > > [I presume you mean quantity 50M of files, each just above 400KB in size] Correct. > >> >> In a test on a subset and accesses 1M files with primarycache set to metadata for the dataset, it did improve performance but used 12-18GB of RAM in the ARC (will run it again when possible forgot the exact number). Anyway, it was far to much RAM to be realistic for us to put in each node for metadata caching. > > You'll need to look at arc_meta_used, not the (ARC) size. For convenience, the high water > mark of arc_meta_used is stored in arc_meta_max. For example, on a smallish VM I have: > # kstat -p ::arcstats:arc_meta_* > zfs:0:arcstats:arc_meta_limit 1202388480 > zfs:0:arcstats:arc_meta_max 48319112 > zfs:0:arcstats:arc_meta_used 48273456 > > NB, for OmniOS/illumos, arc_meta_limit is not a hard limit -- it is a decision point above which > the eviction algorithm changes priority. I have tuned the arc_meta_limit, the problem is that the amount of ARC needed is very large. > >> >> I have three questions I need help with: >> >> 1. How do I calculate required ARC metadata size for a dataset, x bytes per block and file size / block size to get the number of blocks? > > It is much easier to measure. > >> >> 2. If I where to use a L2ARC for metadata I need file size / block size * no files * 200 bytes of RAM ( 400 / 128 * 50M * 200 = ~ 37GB? ) to address the larger L2ARC? > > This is a good starting point. Recall that this is a cache, so it depends on your working set size. > Again, this is easier to measure the size of the headers used. The "200 bytes" can vary based on > OS release and whether you are using a debug kernel (non-debug kernels are smaller) > kstat -p ::arcstats:l2_hdr_size I did a test with 10% of the files, it only used 1GB of the ARC for the L2ARC headers. I had not used the L2ARC with compression before, it was a pleasant surprise that we got 70x for the metadata so we will not be needing a large SSD device, too bad persistent L2ARC have not been integrated, and maybe an option to compress the ARC. I think we will solve our problem with an metadata L2ARC but it will require that we read all files once after each reboot before (~24h warmup). kstat -n arcstats| egrep "asize|l2_size|hdr_size" hdr_size 98454272 l2_asize 3066635264 l2_hdr_size 1006407376 l2_size 82864884736 Regards Henrik From chip at innovates.com Tue Oct 28 21:18:31 2014 From: chip at innovates.com (Schweiss, Chip) Date: Tue, 28 Oct 2014 16:18:31 -0500 Subject: [OmniOS-discuss] Installing on VMware Workstation In-Reply-To: References: Message-ID: That was the trick. Thank you! -Chip On Tue, Oct 28, 2014 at 4:14 PM, Fr?d?ric Alix wrote: > Sorry, i forget the doc link > > http://omnios.omniti.com/wiki.php/VMwareNotes > > 2014-10-28 22:12 GMT+01:00 Fr?d?ric Alix : > >> Hi, >> >> Try to add virtual floppy drive. >> It should work. >> >> Fred >> >> 2014-10-28 22:00 GMT+01:00 Schweiss, Chip : >> >>> I've been regularly using OmniOS on ESXi since r151006 without issue. >>> I'm stuck trying to install on VMware Workstation 10 on Windows 7 64bit. >>> >>> At the "Welcome to OmniOS" screen it dies after pressing F2 and returns >>> to installation menu every time. >>> >>> I've tried telling VMware it's Solaris 11/64, Solaris 10/64, Generic >>> 64. Tried different SCSI adapters, SATA, etc. Nothing changes. I've >>> tried r151008t, r151012, bloody. All the same. >>> >>> Solaris 11.2 installs, OpenIndiana 151a8 installs. >>> >>> Any one know the trick to get OmniOS on VMware Workstation? >>> >>> -Chip >>> >>> _______________________________________________ >>> OmniOS-discuss mailing list >>> OmniOS-discuss at lists.omniti.com >>> http://lists.omniti.com/mailman/listinfo/omnios-discuss >>> >>> >> > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From henson at acm.org Wed Oct 29 01:48:24 2014 From: henson at acm.org (Paul B. Henson) Date: Tue, 28 Oct 2014 18:48:24 -0700 Subject: [OmniOS-discuss] Installing on VMware Workstation In-Reply-To: References: Message-ID: <007001cff31a$71e97be0$55bc73a0$@acm.org> > From: Dan McDonald > Sent: Tuesday, October 28, 2014 2:16 PM > > > Try to add virtual floppy drive. > > DOH! Yeah, that's right. Here's a wiki entry to help: > > http://omnios.omniti.com/wiki.php/VMwareNotes As I recall, the virtual floppy drive just has to be present; it doesn't need to have an actual floppy image associated with it. Also, another option would be to pop into the bios of the VM and configure it to not have a floppy drive installed. The problem arises when the bios says there is a floppy drive, but there is no virtual floppy drive? From nicolas.digregorio at gmail.com Wed Oct 29 13:26:06 2014 From: nicolas.digregorio at gmail.com (Nicolas Di Gregorio) Date: Wed, 29 Oct 2014 14:26:06 +0100 Subject: [OmniOS-discuss] webdav Message-ID: Hello, What are you advising as webdav client? I would like to mount a webdav share at boot time is possible? Thanks in advance Kind Regards digrouz -------------- next part -------------- An HTML attachment was scrubbed... URL: From tobi at oetiker.ch Wed Oct 29 13:31:37 2014 From: tobi at oetiker.ch (Tobias Oetiker) Date: Wed, 29 Oct 2014 14:31:37 +0100 (CET) Subject: [OmniOS-discuss] kvm performance regresseion r10 -> r12 Message-ID: I am still running omnios r10 ...being reluctant to upgrade, because we find that kvm virtio network performance seems to have taken a severe hit from r10 to r12 I am runing this simple test: hw-box$ ssh -c aes256-ctr kvm-guest dd if=/dev/zero bs=10M count=10 > /dev/null On r10 I get 31 MB/s On r12 I get 21 MB/s On both systems we are running kvm with the magic -device-virtio-net-pci,mac=$mac,tx=timer,x-txtimer=200000,x-txburst=128,vlan=0 -net vnic,vlan=0,name=net0,ifname=$VNIC,macaddr=$mac Questions a) do you see that too ? b) any ideas what could be causing it ? kvm-guest is configured with 8 CPUs and 16 GB ram cheers tobi -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten, Switzerland www.oetiker.ch tobi at oetiker.ch +41 62 775 9902 From danmcd at omniti.com Wed Oct 29 14:39:27 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 29 Oct 2014 10:39:27 -0400 Subject: [OmniOS-discuss] kvm performance regresseion r10 -> r12 In-Reply-To: References: Message-ID: > On Oct 29, 2014, at 9:31 AM, Tobias Oetiker wrote: > > I am still running omnios r10 ...being reluctant to upgrade, > because we find that kvm virtio network performance seems to have > taken a severe hit from r10 to r12 > > I am runing this simple test: > > hw-box$ ssh -c aes256-ctr kvm-guest dd if=/dev/zero bs=10M count=10 > /dev/null > > On r10 I get 31 MB/s > > On r12 I get 21 MB/s > > > On both systems we are running kvm with the magic > > -device-virtio-net-pci,mac=$mac,tx=timer,x-txtimer=200000,x-txburst=128,vlan=0 > -net vnic,vlan=0,name=net0,ifname=$VNIC,macaddr=$mac > > > Questions > > a) do you see that too ? > > b) any ideas what could be causing it ? I'll be honest --> there are no changes in the virtio code between r151010 and r151012: r151012(~/ws/illumos-omnios)[0]% git branch * r151012 r151012(~/ws/illumos-omnios)[0]% git whatchanged origin/r151010.. | grep vir 5004 load average should be virtualized for zones :100644 100644 558abd8... 2e82787... M usr/src/man/man5/environ.5 r151012(~/ws/illumos-omnios)[0]% Also, if you're measuring network performance only, use netperf or at least iperf (both are packages you'll have to download and compile for now), as they are better equipped to tell you (esp. netperf) if you're seeing problems in latency, bandwidth, or both. Dan From tobi at oetiker.ch Wed Oct 29 15:36:26 2014 From: tobi at oetiker.ch (Tobias Oetiker) Date: Wed, 29 Oct 2014 16:36:26 +0100 (CET) Subject: [OmniOS-discuss] kvm performance regresseion r10 -> r12 In-Reply-To: References: Message-ID: Hi Dan, Today Dan McDonald wrote: > > > On Oct 29, 2014, at 9:31 AM, Tobias Oetiker wrote: > > > > I am still running omnios r10 ...being reluctant to upgrade, > > because we find that kvm virtio network performance seems to have > > taken a severe hit from r10 to r12 > > > > I am runing this simple test: > > > > hw-box$ ssh -c aes256-ctr kvm-guest dd if=/dev/zero bs=10M count=10 > /dev/null > > > > On r10 I get 31 MB/s > > > > On r12 I get 21 MB/s > > > > > > On both systems we are running kvm with the magic > > > > -device-virtio-net-pci,mac=$mac,tx=timer,x-txtimer=200000,x-txburst=128,vlan=0 > > -net vnic,vlan=0,name=net0,ifname=$VNIC,macaddr=$mac > > > > > > Questions > > > > a) do you see that too ? > > > > b) any ideas what could be causing it ? > > I'll be honest --> there are no changes in the virtio code between r151010 and r151012: > > r151012(~/ws/illumos-omnios)[0]% git branch > * r151012 > r151012(~/ws/illumos-omnios)[0]% git whatchanged origin/r151010.. | grep vir > 5004 load average should be virtualized for zones > :100644 100644 558abd8... 2e82787... M usr/src/man/man5/environ.5 > r151012(~/ws/illumos-omnios)[0]% > > Also, if you're measuring network performance only, use netperf > or at least iperf (both are packages you'll have to download and > compile for now), as they are better equipped to tell you (esp. > netperf) if you're seeing problems in latency, bandwidth, or > both. the reason I am using this test ist that our thinclients are doing roughly that when they are streaming content from your thinlinc server (running in kvm on omni). but yes, netperf would be more reproducible ... cheers tobi -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten, Switzerland www.oetiker.ch tobi at oetiker.ch +41 62 775 9902 From danmcd at omniti.com Wed Oct 29 20:11:48 2014 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 29 Oct 2014 16:11:48 -0400 Subject: [OmniOS-discuss] Building illumos-gate on OmniOS progress References: <0FA0B8EF-1680-4BAD-85AA-583E831EBDB8@kebe.com> Message-ID: I have a series of commits to illumos-gate that will allow illumos-gate to be built on OmniOS, but with these build-failures specifically on OmniOS: 1.) Lint is still mildly dirty. I believe I've narrowed it down to a series of openssl errors only, but there's still going to be lint until OI switches to a more modern openssl or illumos-gate does the Keith W. thing and have its own copy. 2.) I've got two ELF-runtime check complaints left, but I don't want to do to much more to the exception_list/check_rtime file. 3.) Your .env file for building on OmniOS will need a few extra variables set to ='#', which will cause the IPP listener not to build, and SMB printing to not be available. Are those exceptions enough to make you go "Not good enough"? Or are they tolerable? If you, the OmniOS community, are okay with this, I can start attempting to upstream these changes. If you want me to have perfectly silent builds on OmniOS, I don't think it's going to happen realistically in the foreseeable future. Please let me know, and thanks, Dan McD. -- OmniOS Engineering From sjorge+ml at blackdot.be Wed Oct 29 20:47:58 2014 From: sjorge+ml at blackdot.be (Jorge Schrauwen) Date: Wed, 29 Oct 2014 21:47:58 +0100 Subject: [OmniOS-discuss] Building illumos-gate on OmniOS progress In-Reply-To: References: <0FA0B8EF-1680-4BAD-85AA-583E831EBDB8@kebe.com> Message-ID: <3a6318ca1ed6915e279c76db87cf98a8@blackdot.be> +1 for upstream, I don't care about silent as long as I can check if things compile or not. On 2014-10-29 21:11, Dan McDonald wrote: > I have a series of commits to illumos-gate that will allow > illumos-gate to be built on OmniOS, but with these build-failures > specifically on OmniOS: > > 1.) Lint is still mildly dirty. I believe I've narrowed it down to a > series of openssl errors only, but there's still going to be lint > until OI switches to a more modern openssl or illumos-gate does the > Keith W. thing and have its own copy. > > 2.) I've got two ELF-runtime check complaints left, but I don't want > to do to much more to the exception_list/check_rtime file. > > 3.) Your .env file for building on OmniOS will need a few extra > variables set to ='#', which will cause the IPP listener not to build, > and SMB printing to not be available. > > Are those exceptions enough to make you go "Not good enough"? Or are > they tolerable? > > If you, the OmniOS community, are okay with this, I can start > attempting to upstream these changes. If you want me to have > perfectly silent builds on OmniOS, I don't think it's going to happen > realistically in the foreseeable future. > > Please let me know, and thanks, > Dan McD. -- OmniOS Engineering > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From anm at pobox.com Wed Oct 29 21:42:52 2014 From: anm at pobox.com (Andrew McRae) Date: Wed, 29 Oct 2014 21:42:52 +0000 Subject: [OmniOS-discuss] Building illumos-gate on OmniOS progress In-Reply-To: References: <0FA0B8EF-1680-4BAD-85AA-583E831EBDB8@kebe.com> Message-ID: This is good for OmniOS and very good indeed for illumos, and the closer to a perfectly clean build the more easily people are motivated to help make it perfectly clean. What's not to like? Push it up. (I don't believe my opinion's worth anything much here, but want to cheer from the sidelines!) On 29 Oct 2014 20:16, "Dan McDonald" wrote: > I have a series of commits to illumos-gate that will allow illumos-gate to > be built on OmniOS, but with these build-failures specifically on OmniOS: > > 1.) Lint is still mildly dirty. I believe I've narrowed it down to a > series of openssl errors only, but there's still going to be lint until OI > switches to a more modern openssl or illumos-gate does the Keith W. thing > and have its own copy. > > 2.) I've got two ELF-runtime check complaints left, but I don't want to do > to much more to the exception_list/check_rtime file. > > 3.) Your .env file for building on OmniOS will need a few extra variables > set to ='#', which will cause the IPP listener not to build, and SMB > printing to not be available. > > Are those exceptions enough to make you go "Not good enough"? Or are they > tolerable? > > If you, the OmniOS community, are okay with this, I can start attempting > to upstream these changes. If you want me to have perfectly silent builds > on OmniOS, I don't think it's going to happen realistically in the > foreseeable future. > > Please let me know, and thanks, > Dan McD. -- OmniOS Engineering > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -------------- next part -------------- An HTML attachment was scrubbed... URL: From henson at acm.org Thu Oct 30 01:30:46 2014 From: henson at acm.org (Paul B. Henson) Date: Wed, 29 Oct 2014 18:30:46 -0700 Subject: [OmniOS-discuss] Building illumos-gate on OmniOS progress In-Reply-To: References: <0FA0B8EF-1680-4BAD-85AA-583E831EBDB8@kebe.com> Message-ID: <051e01cff3e1$23339d10$699ad730$@acm.org> > From: Dan McDonald > Sent: Wednesday, October 29, 2014 1:12 PM > > Are those exceptions enough to make you go "Not good enough"? Or are > they tolerable? As long as reviewers and advocates are willing to tolerate the noise when trying to submit changes upstream that sounds good enough to me :)... Thanks. From svavar at pipar-tbwa.is Thu Oct 30 18:39:19 2014 From: svavar at pipar-tbwa.is (=?iso-8859-1?Q?Svavar_=D6rn_Eysteinsson?=) Date: Thu, 30 Oct 2014 18:39:19 +0000 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform Message-ID: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> Hello List. I recently updated my OmniOS hp Microserver from r151010 to r151012. Well, it didn't work. As I rebooted my server and then it shows me the following error message : krtld: failed to open '/platform###' krtld: bind_primary(): no relocation information found for module /platform## krtld: error during initial load/link phase krtld could neither locate nor resolve symbols for: /platform### in the boot archive. Please verify that this file matches what is found in the boot archive You may need to boot using the solaris failsafe to fix this. Unable to Boot. Press any key to reboot. ### = means strange ASCII character. No matter what boot environment I choose I allways get the same error. I have also mounted some of the BE in my USB live version of OmniOS R151012 and updated the boot-archive with these commands : zpool import -f rpool beadm mount some_be /a rm /a/platform/i86pc/boot_archive rm /a/platform/i86pc/amd64/boot_archive touch /a/etc/system bootadm update-archive -v -R /a zpool export -f rpool init 6 And try to boot the BE in the next reboot. Still no go, I allways get this error message. Does anyone know what the heck is going on ? Any help and or information much appreciated. Thanks allot. Best regards, Svavar O Reykjavik Iceland -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Thu Oct 30 18:56:40 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 30 Oct 2014 14:56:40 -0400 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform In-Reply-To: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> References: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> Message-ID: > On Oct 30, 2014, at 2:39 PM, Svavar ?rn Eysteinsson wrote: > > Hello List. > > I recently updated my OmniOS hp Microserver from r151010 to r151012. > Well, it didn't work. As I rebooted my server and then it shows me the following error message : > > krtld: failed to open '/platform###' How exactly did you update it? I ask because I'd like to reproduce your problem on one of my VMs. Dan From svavar at pipar-tbwa.is Thu Oct 30 19:04:43 2014 From: svavar at pipar-tbwa.is (=?iso-8859-1?Q?Svavar_=D6rn_Eysteinsson?=) Date: Thu, 30 Oct 2014 19:04:43 +0000 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform In-Reply-To: References: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> Message-ID: I just did a pkg update Just plain and simple, and then reboot. :( > On 30 Oct 2014, at 18:56, Dan McDonald wrote: > > >> On Oct 30, 2014, at 2:39 PM, Svavar ?rn Eysteinsson wrote: >> >> Hello List. >> >> I recently updated my OmniOS hp Microserver from r151010 to r151012. >> Well, it didn't work. As I rebooted my server and then it shows me the following error message : >> >> krtld: failed to open '/platform###' > > How exactly did you update it? I ask because I'd like to reproduce your problem on one of my VMs. > > Dan > > From danmcd at omniti.com Thu Oct 30 19:14:15 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 30 Oct 2014 15:14:15 -0400 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform In-Reply-To: References: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> Message-ID: <2D4C08F4-B55B-44A3-B4FD-ACEE2C8EF9A2@omniti.com> > On Oct 30, 2014, at 3:04 PM, Svavar ?rn Eysteinsson wrote: > > I just did a pkg update > > Just plain and simple, and then reboot. > > :( But updating from 010 to 012 requires a packaging change. You set-publisher and all that good stuff?!? This is why I want to know EXACTLY what you did. I have several r151010 BEs ready to test now. I also just typed "pkg update" into one of them just now. All I got was some updates to openssl, bash, wget, etc. Dan From svavar at pipar-tbwa.is Thu Oct 30 19:15:53 2014 From: svavar at pipar-tbwa.is (=?Windows-1252?Q?Svavar_=D6rn_Eysteinsson?=) Date: Thu, 30 Oct 2014 19:15:53 +0000 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform In-Reply-To: References: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> Message-ID: <4396DD64-A54E-47BF-95CE-D51BB5263B22@pipar-tbwa.is> I?m sorry. I was mistaken. I didn?t unpublish the omnios repo and publish the R151012 so I was just updating my current r151010 setup with a regular ?pkg update? command. > On 30 Oct 2014, at 19:04, Svavar ?rn Eysteinsson wrote: > > I just did a pkg update > > Just plain and simple, and then reboot. > > :( > > > >> On 30 Oct 2014, at 18:56, Dan McDonald wrote: >> >> >>> On Oct 30, 2014, at 2:39 PM, Svavar ?rn Eysteinsson wrote: >>> >>> Hello List. >>> >>> I recently updated my OmniOS hp Microserver from r151010 to r151012. >>> Well, it didn't work. As I rebooted my server and then it shows me the following error message : >>> >>> krtld: failed to open '/platform###' >> >> How exactly did you update it? I ask because I'd like to reproduce your problem on one of my VMs. >> >> Dan >> >> > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From danmcd at omniti.com Thu Oct 30 19:17:06 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 30 Oct 2014 15:17:06 -0400 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform In-Reply-To: <4396DD64-A54E-47BF-95CE-D51BB5263B22@pipar-tbwa.is> References: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> <4396DD64-A54E-47BF-95CE-D51BB5263B22@pipar-tbwa.is> Message-ID: <8D319B84-F204-460F-B9FD-3DA22E10F07B@omniti.com> > On Oct 30, 2014, at 3:15 PM, Svavar ?rn Eysteinsson wrote: > > I?m sorry. I was mistaken. > > I didn?t unpublish the omnios repo and publish the R151012 so I was just updating my current r151010 setup > with a regular ?pkg update? command. Still, having a failure like that isn't helpful. My update I just did didn't create me a new BE and force a reboot. I'm going to reboot it now, though, to see if anything odd happens anyway. Expect a followup mail with the results. Dan From svavar at pipar-tbwa.is Thu Oct 30 19:21:17 2014 From: svavar at pipar-tbwa.is (=?Windows-1252?Q?Svavar_=D6rn_Eysteinsson?=) Date: Thu, 30 Oct 2014 19:21:17 +0000 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform In-Reply-To: <2D4C08F4-B55B-44A3-B4FD-ACEE2C8EF9A2@omniti.com> References: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> <2D4C08F4-B55B-44A3-B4FD-ACEE2C8EF9A2@omniti.com> Message-ID: <9001F483-30D9-47F9-BFC2-6AACC19ACFCD@pipar-tbwa.is> Thats actually what I noticed. There where some bash, wget ssl updates and some core updates if I recall. I did however create a be. And then I only issued a reboot statement as I was using a SSH connection. When I got home, kaboom. errors. Quite strange? > On 30 Oct 2014, at 19:14, Dan McDonald wrote: > > >> On Oct 30, 2014, at 3:04 PM, Svavar ?rn Eysteinsson wrote: >> >> I just did a pkg update >> >> Just plain and simple, and then reboot. >> >> :( > > But updating from 010 to 012 requires a packaging change. You set-publisher and all that good stuff?!? > > This is why I want to know EXACTLY what you did. I have several r151010 BEs ready to test now. > > I also just typed "pkg update" into one of them just now. All I got was some updates to openssl, bash, wget, etc. > > Dan > From danmcd at omniti.com Thu Oct 30 19:24:06 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 30 Oct 2014 15:24:06 -0400 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform In-Reply-To: <9001F483-30D9-47F9-BFC2-6AACC19ACFCD@pipar-tbwa.is> References: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> <2D4C08F4-B55B-44A3-B4FD-ACEE2C8EF9A2@omniti.com> <9001F483-30D9-47F9-BFC2-6AACC19ACFCD@pipar-tbwa.is> Message-ID: > On Oct 30, 2014, at 3:21 PM, Svavar ?rn Eysteinsson wrote: > > Thats actually what I noticed. > There where some bash, wget ssl updates and some core updates if I recall. > I did however create a be. > > And then I only issued a reboot statement as I was using a SSH connection. > When I got home, kaboom. errors. I just rebooted and No problems. My 010 was previously updated on August 20th, which may have included some reboot-worthy changes. Post August 20th, nothing is reboot worthy,and only a backup BE got created. You may want to (from a good BE) zpool scrub, just in case. Dan From svavar at pipar-tbwa.is Thu Oct 30 19:31:26 2014 From: svavar at pipar-tbwa.is (=?Windows-1252?Q?Svavar_=D6rn_Eysteinsson?=) Date: Thu, 30 Oct 2014 19:31:26 +0000 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform In-Reply-To: References: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> <2D4C08F4-B55B-44A3-B4FD-ACEE2C8EF9A2@omniti.com> <9001F483-30D9-47F9-BFC2-6AACC19ACFCD@pipar-tbwa.is> Message-ID: Oks. Thank you. Is it enough for me to boot from live usb and import the rpool and scrub-it ? (just to be sure?. as you stated before) As there aren?t any BE that I can boot from? Thanks Dan. > On 30 Oct 2014, at 19:24, Dan McDonald wrote: > > >> On Oct 30, 2014, at 3:21 PM, Svavar ?rn Eysteinsson wrote: >> >> Thats actually what I noticed. >> There where some bash, wget ssl updates and some core updates if I recall. >> I did however create a be. >> >> And then I only issued a reboot statement as I was using a SSH connection. >> When I got home, kaboom. errors. > > I just rebooted and No problems. My 010 was previously updated on August 20th, which may have included some reboot-worthy changes. Post August 20th, nothing is reboot worthy,and only a backup BE got created. > > You may want to (from a good BE) zpool scrub, just in case. > > Dan > From danmcd at omniti.com Thu Oct 30 19:33:04 2014 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 30 Oct 2014 15:33:04 -0400 Subject: [OmniOS-discuss] Cannot boot after update - failed to open /platform In-Reply-To: References: <37335E68-E589-40F1-AA36-5D0BDA58CF54@pipar-tbwa.is> <2D4C08F4-B55B-44A3-B4FD-ACEE2C8EF9A2@omniti.com> <9001F483-30D9-47F9-BFC2-6AACC19ACFCD@pipar-tbwa.is> Message-ID: <61B10495-C4A7-49FC-A17D-90EE46D73971@omniti.com> > On Oct 30, 2014, at 3:31 PM, Svavar ?rn Eysteinsson wrote: > > Oks. > Thank you. > > Is it enough for me to boot from live usb and import the rpool and scrub-it ? > (just to be sure?. as you stated before) > As there aren?t any BE that I can boot from? That's one thing you can do, yes. I'd recommend also using an 010 USB to scrub and fix your pool. Just in case there are bizarre zpool feature-flag interactions. It's the safer thing to do. Dan From rt at steait.net Fri Oct 31 06:33:32 2014 From: rt at steait.net (Rune Tipsmark) Date: Fri, 31 Oct 2014 06:33:32 +0000 Subject: [OmniOS-discuss] zfs pool 100% busy, disks less than 10% Message-ID: <4153d29f2b3a42349a5b701edfb06e92@EX1301.steait.net> Hi all, Hope someone can help me get this pool running as it should, I am seeing something like 200-300 MB/sec max which is much much less than I want to see... 11 mirrored vdevs... 2 spares and 2 SLOG devices, 192gb ram in host... Why is this pool showing near 100% busy when the underlying disks are doing nothing at all.... pool: pool04 state: ONLINE scan: scrub repaired 0 in 0h0m with 0 errors on Fri Oct 31 06:37:30 2014 config: NAME STATE READ WRITE CKSUM pool04 ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c4t5000C50055FC9533d0 ONLINE 0 0 0 c4t5000C50055FE6A63d0 ONLINE 0 0 0 mirror-1 ONLINE 0 0 0 c4t5000C5005708296Fd0 ONLINE 0 0 0 c4t5000C5005708351Bd0 ONLINE 0 0 0 mirror-2 ONLINE 0 0 0 c4t5000C500570858EFd0 ONLINE 0 0 0 c4t5000C50057085A6Bd0 ONLINE 0 0 0 mirror-3 ONLINE 0 0 0 c4t5000C50057086307d0 ONLINE 0 0 0 c4t5000C50057086B67d0 ONLINE 0 0 0 mirror-4 ONLINE 0 0 0 c4t5000C500570870D3d0 ONLINE 0 0 0 c4t5000C50057089753d0 ONLINE 0 0 0 mirror-5 ONLINE 0 0 0 c4t5000C500625B7EA7d0 ONLINE 0 0 0 c4t5000C500625B8137d0 ONLINE 0 0 0 mirror-6 ONLINE 0 0 0 c4t5000C500625B8427d0 ONLINE 0 0 0 c4t5000C500625B86E3d0 ONLINE 0 0 0 mirror-7 ONLINE 0 0 0 c4t5000C500625B886Fd0 ONLINE 0 0 0 c4t5000C500625BB773d0 ONLINE 0 0 0 mirror-8 ONLINE 0 0 0 c4t5000C500625BC2C3d0 ONLINE 0 0 0 c4t5000C500625BD3EBd0 ONLINE 0 0 0 mirror-9 ONLINE 0 0 0 c4t5000C50062878C0Bd0 ONLINE 0 0 0 c4t5000C50062878C43d0 ONLINE 0 0 0 mirror-10 ONLINE 0 0 0 c4t5000C50062879687d0 ONLINE 0 0 0 c4t5000C50062879707d0 ONLINE 0 0 0 logs c10d0 ONLINE 0 0 0 c11d0 ONLINE 0 0 0 spares c4t5000C50062879723d0 AVAIL c4t5000C50062879787d0 AVAIL extended device statistics r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device 0.0 337.0 0.0 21899.1 0.0 0.0 0.0 0.0 0 0 c10d0 0.0 337.0 0.0 21835.1 0.0 0.0 0.0 0.0 0 0 c11d0 149.0 249.0 7723.2 7032.7 0.0 0.0 0.0 0.0 0 0 c12d0 144.0 295.0 6766.2 6955.7 0.0 0.0 0.0 0.0 0 0 c14d0 154.0 311.0 9241.6 6445.7 0.0 0.0 0.0 0.0 0 0 c15d0 150.0 263.0 7939.2 6096.7 0.0 0.0 0.0 0.0 0 0 c13d0 597.0 1118.0 31670.2 26530.9 5.8 0.4 3.4 0.2 5 8 pool03 0.0 1382.9 0.0 78460.7 11.4 1.7 8.3 1.3 5 97 pool04 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 rpool 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c7t0d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c7t1d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c7t4d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c7t5d0 0.0 25.0 0.0 1043.5 0.0 0.0 0.0 1.3 0 3 c4t5000C50055FE6A63d0 0.0 69.0 0.0 1478.4 0.0 0.0 0.0 0.7 0 4 c4t5000C5005708351Bd0 0.0 41.0 0.0 1261.9 0.0 0.0 0.0 0.7 0 3 c4t5000C50057086B67d0 0.0 37.0 0.0 2146.4 0.0 0.0 0.0 0.8 0 3 c4t5000C500570858EFd0 0.0 37.0 0.0 2146.4 0.0 0.0 0.0 1.2 0 4 c4t5000C50057085A6Bd0 0.0 69.0 0.0 1478.4 0.0 0.1 0.0 0.8 0 5 c4t5000C5005708296Fd0 0.0 26.0 0.0 1868.9 0.0 0.0 0.0 1.2 0 3 c4t5000C50057089753d0 0.0 27.0 0.0 1868.9 0.0 0.0 0.0 1.0 0 3 c4t5000C500570870D3d0 0.0 39.0 0.0 1261.9 0.0 0.0 0.0 0.7 0 3 c4t5000C50057086307d0 0.0 25.0 0.0 1043.5 0.0 0.0 0.0 1.3 0 3 c4t5000C50055FC9533d0 0.0 26.0 0.0 2103.9 0.0 0.0 0.0 0.8 0 2 c4t5000C50062878C43d0 0.0 15.0 0.0 1074.0 0.0 0.0 0.0 1.9 0 3 c4t5000C500625B886Fd0 0.0 15.0 0.0 1044.0 0.0 0.0 0.0 1.6 0 2 c4t5000C500625BD3EBd0 0.0 15.0 0.0 1044.0 0.0 0.0 0.0 2.6 0 4 c4t5000C500625BC2C3d0 0.0 15.0 0.0 1074.0 0.0 0.0 0.0 1.3 0 2 c4t5000C500625BB773d0 0.0 59.0 0.0 1388.4 0.0 0.0 0.0 0.6 0 3 c4t5000C500625B8427d0 0.0 44.0 0.0 1849.9 0.0 0.0 0.0 0.8 0 3 c4t5000C500625B8137d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c4t5000C50062879723d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c4t5000C50062879787d0 0.0 28.0 0.0 2103.9 0.0 0.0 0.0 1.0 0 3 c4t5000C50062879707d0 0.0 56.0 0.0 1388.4 0.0 0.0 0.0 0.6 0 4 c4t5000C500625B86E3d0 0.0 43.0 0.0 1849.9 0.0 0.0 0.0 0.8 0 3 c4t5000C500625B7EA7d0 0.0 26.0 0.0 2103.9 0.0 0.0 0.0 1.1 0 3 c4t5000C50062878C0Bd0 0.0 28.0 0.0 2103.9 0.0 0.0 0.0 1.2 0 3 c4t5000C50062879687d0 NAME PROPERTY VALUE SOURCE pool04 type filesystem - pool04 creation Fri Oct 31 6:31 2014 - pool04 used 358G - pool04 available 38.5T - pool04 referenced 175G - pool04 compressratio 1.71x - pool04 mounted yes - pool04 quota none default pool04 reservation none default pool04 recordsize 128K default pool04 mountpoint /pool04 default pool04 sharenfs off default pool04 checksum on default pool04 compression lz4 local pool04 atime off local pool04 devices on default pool04 exec on default pool04 setuid on default pool04 readonly off default pool04 zoned off default pool04 snapdir hidden default pool04 aclmode discard default pool04 aclinherit restricted default pool04 canmount on default pool04 xattr on default pool04 copies 1 default pool04 version 5 - pool04 utf8only off - pool04 normalization none - pool04 casesensitivity sensitive - pool04 vscan off default pool04 nbmand off default pool04 sharesmb off default pool04 refquota none default pool04 refreservation 358G local pool04 primarycache all default pool04 secondarycache all default pool04 usedbysnapshots 0 - pool04 usedbydataset 175G - pool04 usedbychildren 2.53M - pool04 usedbyrefreservation 183G - pool04 logbias latency default pool04 dedup off default pool04 mlslabel none default pool04 sync always local pool04 refcompressratio 1.71x - pool04 written 175G - pool04 logicalused 300G - pool04 logicalreferenced 300G - pool04 filesystem_limit none default pool04 snapshot_limit none default pool04 filesystem_count none default pool04 snapshot_count none default pool04 redundant_metadata all default Br, Rune -------------- next part -------------- An HTML attachment was scrubbed... URL: From filip.marvan at aira.cz Fri Oct 31 11:25:01 2014 From: filip.marvan at aira.cz (Filip Marvan) Date: Fri, 31 Oct 2014 12:25:01 +0100 Subject: [OmniOS-discuss] zfs pool 100% busy, disks less than 10% Message-ID: <3BE0DEED8863E5429BAE4CAEDF62456503804DA92C95@AIRA-SRV.aira.local> Hi, what devices are you using as SLOG? As you have "sync=always" on that pool, performance of SLOG device will be crucial for write speed on your pool. I can see that one of your disk is very busy 0.0 1382.9 0.0 78460.7 11.4 1.7 8.3 1.3 5 97 pool04 Is that SLOG? Maybe that is the problem. Filip Marvan ------------------ Date: Fri, 31 Oct 2014 06:33:32 +0000 From: Rune Tipsmark To: "omnios-discuss at lists.omniti.com" Subject: [OmniOS-discuss] zfs pool 100% busy, disks less than 10% Message-ID: <4153d29f2b3a42349a5b701edfb06e92 at EX1301.steait.net> Content-Type: text/plain; charset="us-ascii" Hi all, Hope someone can help me get this pool running as it should, I am seeing something like 200-300 MB/sec max which is much much less than I want to see... 11 mirrored vdevs... 2 spares and 2 SLOG devices, 192gb ram in host... Why is this pool showing near 100% busy when the underlying disks are doing nothing at all.... pool: pool04 state: ONLINE scan: scrub repaired 0 in 0h0m with 0 errors on Fri Oct 31 06:37:30 2014 config: NAME STATE READ WRITE CKSUM pool04 ONLINE 0 0 0 mirror-0 ONLINE 0 0 0 c4t5000C50055FC9533d0 ONLINE 0 0 0 c4t5000C50055FE6A63d0 ONLINE 0 0 0 mirror-1 ONLINE 0 0 0 c4t5000C5005708296Fd0 ONLINE 0 0 0 c4t5000C5005708351Bd0 ONLINE 0 0 0 mirror-2 ONLINE 0 0 0 c4t5000C500570858EFd0 ONLINE 0 0 0 c4t5000C50057085A6Bd0 ONLINE 0 0 0 mirror-3 ONLINE 0 0 0 c4t5000C50057086307d0 ONLINE 0 0 0 c4t5000C50057086B67d0 ONLINE 0 0 0 mirror-4 ONLINE 0 0 0 c4t5000C500570870D3d0 ONLINE 0 0 0 c4t5000C50057089753d0 ONLINE 0 0 0 mirror-5 ONLINE 0 0 0 c4t5000C500625B7EA7d0 ONLINE 0 0 0 c4t5000C500625B8137d0 ONLINE 0 0 0 mirror-6 ONLINE 0 0 0 c4t5000C500625B8427d0 ONLINE 0 0 0 c4t5000C500625B86E3d0 ONLINE 0 0 0 mirror-7 ONLINE 0 0 0 c4t5000C500625B886Fd0 ONLINE 0 0 0 c4t5000C500625BB773d0 ONLINE 0 0 0 mirror-8 ONLINE 0 0 0 c4t5000C500625BC2C3d0 ONLINE 0 0 0 c4t5000C500625BD3EBd0 ONLINE 0 0 0 mirror-9 ONLINE 0 0 0 c4t5000C50062878C0Bd0 ONLINE 0 0 0 c4t5000C50062878C43d0 ONLINE 0 0 0 mirror-10 ONLINE 0 0 0 c4t5000C50062879687d0 ONLINE 0 0 0 c4t5000C50062879707d0 ONLINE 0 0 0 logs c10d0 ONLINE 0 0 0 c11d0 ONLINE 0 0 0 spares c4t5000C50062879723d0 AVAIL c4t5000C50062879787d0 AVAIL extended device statistics r/s w/s kr/s kw/s wait actv wsvc_t asvc_t %w %b device 0.0 337.0 0.0 21899.1 0.0 0.0 0.0 0.0 0 0 c10d0 0.0 337.0 0.0 21835.1 0.0 0.0 0.0 0.0 0 0 c11d0 149.0 249.0 7723.2 7032.7 0.0 0.0 0.0 0.0 0 0 c12d0 144.0 295.0 6766.2 6955.7 0.0 0.0 0.0 0.0 0 0 c14d0 154.0 311.0 9241.6 6445.7 0.0 0.0 0.0 0.0 0 0 c15d0 150.0 263.0 7939.2 6096.7 0.0 0.0 0.0 0.0 0 0 c13d0 597.0 1118.0 31670.2 26530.9 5.8 0.4 3.4 0.2 5 8 pool03 0.0 1382.9 0.0 78460.7 11.4 1.7 8.3 1.3 5 97 pool04 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 rpool 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c7t0d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c7t1d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c7t4d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c7t5d0 0.0 25.0 0.0 1043.5 0.0 0.0 0.0 1.3 0 3 c4t5000C50055FE6A63d0 0.0 69.0 0.0 1478.4 0.0 0.0 0.0 0.7 0 4 c4t5000C5005708351Bd0 0.0 41.0 0.0 1261.9 0.0 0.0 0.0 0.7 0 3 c4t5000C50057086B67d0 0.0 37.0 0.0 2146.4 0.0 0.0 0.0 0.8 0 3 c4t5000C500570858EFd0 0.0 37.0 0.0 2146.4 0.0 0.0 0.0 1.2 0 4 c4t5000C50057085A6Bd0 0.0 69.0 0.0 1478.4 0.0 0.1 0.0 0.8 0 5 c4t5000C5005708296Fd0 0.0 26.0 0.0 1868.9 0.0 0.0 0.0 1.2 0 3 c4t5000C50057089753d0 0.0 27.0 0.0 1868.9 0.0 0.0 0.0 1.0 0 3 c4t5000C500570870D3d0 0.0 39.0 0.0 1261.9 0.0 0.0 0.0 0.7 0 3 c4t5000C50057086307d0 0.0 25.0 0.0 1043.5 0.0 0.0 0.0 1.3 0 3 c4t5000C50055FC9533d0 0.0 26.0 0.0 2103.9 0.0 0.0 0.0 0.8 0 2 c4t5000C50062878C43d0 0.0 15.0 0.0 1074.0 0.0 0.0 0.0 1.9 0 3 c4t5000C500625B886Fd0 0.0 15.0 0.0 1044.0 0.0 0.0 0.0 1.6 0 2 c4t5000C500625BD3EBd0 0.0 15.0 0.0 1044.0 0.0 0.0 0.0 2.6 0 4 c4t5000C500625BC2C3d0 0.0 15.0 0.0 1074.0 0.0 0.0 0.0 1.3 0 2 c4t5000C500625BB773d0 0.0 59.0 0.0 1388.4 0.0 0.0 0.0 0.6 0 3 c4t5000C500625B8427d0 0.0 44.0 0.0 1849.9 0.0 0.0 0.0 0.8 0 3 c4t5000C500625B8137d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c4t5000C50062879723d0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 c4t5000C50062879787d0 0.0 28.0 0.0 2103.9 0.0 0.0 0.0 1.0 0 3 c4t5000C50062879707d0 0.0 56.0 0.0 1388.4 0.0 0.0 0.0 0.6 0 4 c4t5000C500625B86E3d0 0.0 43.0 0.0 1849.9 0.0 0.0 0.0 0.8 0 3 c4t5000C500625B7EA7d0 0.0 26.0 0.0 2103.9 0.0 0.0 0.0 1.1 0 3 c4t5000C50062878C0Bd0 0.0 28.0 0.0 2103.9 0.0 0.0 0.0 1.2 0 3 c4t5000C50062879687d0 NAME PROPERTY VALUE SOURCE pool04 type filesystem - pool04 creation Fri Oct 31 6:31 2014 - pool04 used 358G - pool04 available 38.5T - pool04 referenced 175G - pool04 compressratio 1.71x - pool04 mounted yes - pool04 quota none default pool04 reservation none default pool04 recordsize 128K default pool04 mountpoint /pool04 default pool04 sharenfs off default pool04 checksum on default pool04 compression lz4 local pool04 atime off local pool04 devices on default pool04 exec on default pool04 setuid on default pool04 readonly off default pool04 zoned off default pool04 snapdir hidden default pool04 aclmode discard default pool04 aclinherit restricted default pool04 canmount on default pool04 xattr on default pool04 copies 1 default pool04 version 5 - pool04 utf8only off - pool04 normalization none - pool04 casesensitivity sensitive - pool04 vscan off default pool04 nbmand off default pool04 sharesmb off default pool04 refquota none default pool04 refreservation 358G local pool04 primarycache all default pool04 secondarycache all default pool04 usedbysnapshots 0 - pool04 usedbydataset 175G - pool04 usedbychildren 2.53M - pool04 usedbyrefreservation 183G - pool04 logbias latency default pool04 dedup off default pool04 mlslabel none default pool04 sync always local pool04 refcompressratio 1.71x - pool04 written 175G - pool04 logicalused 300G - pool04 logicalreferenced 300G - pool04 filesystem_limit none default pool04 snapshot_limit none default pool04 filesystem_count none default pool04 snapshot_count none default pool04 redundant_metadata all default Br, Rune -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Subject: Digest Footer _______________________________________________ OmniOS-discuss mailing list OmniOS-discuss at lists.omniti.com http://lists.omniti.com/mailman/listinfo/omnios-discuss ------------------------------ End of OmniOS-discuss Digest, Vol 31, Issue 39 ********************************************** -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6247 bytes Desc: not available URL: From eric.sproul at circonus.com Fri Oct 31 14:14:18 2014 From: eric.sproul at circonus.com (Eric Sproul) Date: Fri, 31 Oct 2014 10:14:18 -0400 Subject: [OmniOS-discuss] zfs pool 100% busy, disks less than 10% In-Reply-To: <4153d29f2b3a42349a5b701edfb06e92@EX1301.steait.net> References: <4153d29f2b3a42349a5b701edfb06e92@EX1301.steait.net> Message-ID: On Fri, Oct 31, 2014 at 2:33 AM, Rune Tipsmark wrote: > Why is this pool showing near 100% busy when the underlying disks are doing > nothing at all?. Simply put, it's just how the accounting works in iostat. It treats the pool like any other device, so if there is even one outstanding request to the pool, it counts towards the busy%. Keith W. from Joyent explained this recently on the illumos-zfs list: http://www.listbox.com/member/archive/182191/2014/10/sort/time_rev/page/3/entry/18:93/20141017161955:F3E11AB2-563A-11E4-8EDC-D0C677981E2F/ The TL;DR is: if your pool has more than one disk in it, the pool-wide busy% is useless. Eric From richard.elling at richardelling.com Fri Oct 31 16:02:56 2014 From: richard.elling at richardelling.com (Richard Elling) Date: Fri, 31 Oct 2014 09:02:56 -0700 Subject: [OmniOS-discuss] zfs pool 100% busy, disks less than 10% In-Reply-To: References: <4153d29f2b3a42349a5b701edfb06e92@EX1301.steait.net> Message-ID: <8FD2525B-70E3-4227-B9D5-8D4399D31335@richardelling.com> On Oct 31, 2014, at 7:14 AM, Eric Sproul wrote: > On Fri, Oct 31, 2014 at 2:33 AM, Rune Tipsmark wrote: > >> Why is this pool showing near 100% busy when the underlying disks are doing >> nothing at all?. > > Simply put, it's just how the accounting works in iostat. It treats > the pool like any other device, so if there is even one outstanding > request to the pool, it counts towards the busy%. Keith W. from > Joyent explained this recently on the illumos-zfs list: > http://www.listbox.com/member/archive/182191/2014/10/sort/time_rev/page/3/entry/18:93/20141017161955:F3E11AB2-563A-11E4-8EDC-D0C677981E2F/ > > The TL;DR is: if your pool has more than one disk in it, the pool-wide > busy% is useless. FWIW, we use %busy as an indicator that we can ignore a device/subsystem when looking for performance problems. We don't use it as an indicator of problems. In other words, if the device isn't > 10% busy, forgetabouddit. If it is more busy, look in more detail at the meaningful performance indicators. -- richard From rt at steait.net Fri Oct 31 19:38:01 2014 From: rt at steait.net (Rune Tipsmark) Date: Fri, 31 Oct 2014 19:38:01 +0000 Subject: [OmniOS-discuss] zfs pool 100% busy, disks less than 10% In-Reply-To: <8FD2525B-70E3-4227-B9D5-8D4399D31335@richardelling.com> References: <4153d29f2b3a42349a5b701edfb06e92@EX1301.steait.net> <8FD2525B-70E3-4227-B9D5-8D4399D31335@richardelling.com> Message-ID: Ok, makes sense. What other kind of indicators can I look at? I get decent results from DD but still feels a bit slow... Compression lz4 should not slow it down right? Cpu is not doing much when copying data over, maybe 15% busy or so... Sync=always, block size 1M 204800000000 bytes (205 GB) copied, 296.379 s, 691 MB/s real 4m56.382s user 0m0.461s sys 3m12.662s Sync=disabled, block size 1M 204800000000 bytes (205 GB) copied, 117.774 s, 1.7 GB/s real 1m57.777s user 0m0.237s sys 1m57.466s ... while doing this I was looking at my FIO cards, I think the reason is that the SLC's need more power to deliver higher performance, they are supposed to deliver 1.5GB/sec but only delivers around 350MB/sec each.... Now looking for aux power cables and will retest... Br, Rune -----Original Message----- From: Richard Elling [mailto:richard.elling at richardelling.com] Sent: Friday, October 31, 2014 9:03 AM To: Eric Sproul Cc: Rune Tipsmark; omnios-discuss at lists.omniti.com Subject: Re: [OmniOS-discuss] zfs pool 100% busy, disks less than 10% On Oct 31, 2014, at 7:14 AM, Eric Sproul wrote: > On Fri, Oct 31, 2014 at 2:33 AM, Rune Tipsmark wrote: > >> Why is this pool showing near 100% busy when the underlying disks are >> doing nothing at all.... > > Simply put, it's just how the accounting works in iostat. It treats > the pool like any other device, so if there is even one outstanding > request to the pool, it counts towards the busy%. Keith W. from > Joyent explained this recently on the illumos-zfs list: > http://www.listbox.com/member/archive/182191/2014/10/sort/time_rev/pag > e/3/entry/18:93/20141017161955:F3E11AB2-563A-11E4-8EDC-D0C677981E2F/ > > The TL;DR is: if your pool has more than one disk in it, the pool-wide > busy% is useless. FWIW, we use %busy as an indicator that we can ignore a device/subsystem when looking for performance problems. We don't use it as an indicator of problems. In other words, if the device isn't > 10% busy, forgetabouddit. If it is more busy, look in more detail at the meaningful performance indicators. -- richard From alexander.eremin at nexenta.com Fri Oct 31 14:09:18 2014 From: alexander.eremin at nexenta.com (Alexander Eremin) Date: Fri, 31 Oct 2014 18:09:18 +0400 Subject: [OmniOS-discuss] pkg from oracle upstream References: Message-ID: <6DB328AD-8E78-4E2E-9F23-516BB62B016A@nexenta.com> Hello Dan, since summer we use newest pkg (rev 3111:52c4611b7306 - later there was implemented python2.7 support and this requires more fixes) in our test builds. So far it looks good and much faster, so would be very nice to update omnios version too: if you are interested in this, I'm pretty sure you could look at attached patch for this revision and prepare more accurate fix for omnios build version. Cheers, Alex -------------- next part -------------- A non-text attachment was scrubbed... Name: pkg.patch Type: application/octet-stream Size: 54481 bytes Desc: not available URL: