From geoffn at gnaa.net Tue Nov 1 05:32:34 2016 From: geoffn at gnaa.net (Geoff Nordli) Date: Mon, 31 Oct 2016 22:32:34 -0700 Subject: [OmniOS-discuss] upgrading pkgsrc Message-ID: <78442e72-db59-ca05-f499-15dc58620e93@gnaa.net> Hi. I am running OmniOS v11 r151016 I tried to upgrade pkgsrc using the steps outlined here: https://pkgsrc.joyent.com/install-on-illumos/ when I tried to upgrade pkgin it gives me this error: root at pc-oi1:~# /opt/local/sbin/pkg_add -U pkgin SSL support disabled pkg_add: Can't process https://pkgsrc.joyent.com:443/packages/SmartOS/2016Q3/x86_64/All/pkgin*: Not owner pkg_add: no pkg found for 'pkgin', sorry. pkg_add: 1 package addition failed If I try to do a pkgin update: root at pc-oi1:~# /opt/local/bin/pkgin update reading local summary... processing local summary... SSL support disabled SSL support disabled SSL support disabled pkgin: Could not fetch https://pkgsrc.joyent.com/packages/SmartOS/2016Q3/x86_64/All/pkg_summary.gz If I use wget, i can download the gz file OK, so definitely the networking is working fine. Any thoughts? thanks, Geoff From peter.tribble at gmail.com Tue Nov 1 11:09:46 2016 From: peter.tribble at gmail.com (Peter Tribble) Date: Tue, 1 Nov 2016 11:09:46 +0000 Subject: [OmniOS-discuss] repo not showing current packages Message-ID: Hi, The web interface for the repo http://pkg.omniti.com/omnios/r151014/ isn't showing the latest versions in search results. It's showing the 20160603 packages, but not the recently updated ones (20161027). Any chance it could be poked? Thanks, -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From davide.poletto at gmail.com Tue Nov 1 12:41:35 2016 From: davide.poletto at gmail.com (Davide Poletto) Date: Tue, 1 Nov 2016 13:41:35 +0100 Subject: [OmniOS-discuss] repo not showing current packages In-Reply-To: References: Message-ID: Yeah, browsing the web page for the OmniOS r151014 repo looks indeed actually unresponsive...or very very slowly responsive, at best. Browsing other OmniOS repositories' web pages (r151010, r151012, r151016, r151018 and r151020) looks normal. On Tue, Nov 1, 2016 at 12:09 PM, Peter Tribble wrote: > Hi, > > The web interface for the repo > > http://pkg.omniti.com/omnios/r151014/ > > isn't showing the latest versions in search results. It's showing the > 20160603 > packages, but not the recently updated ones (20161027). Any chance it could > be poked? > > Thanks, > > -- > -Peter Tribble > http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ > > _______________________________________________ > 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 Nov 1 14:59:09 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 1 Nov 2016 10:59:09 -0400 Subject: [OmniOS-discuss] upgrading pkgsrc In-Reply-To: <78442e72-db59-ca05-f499-15dc58620e93@gnaa.net> References: <78442e72-db59-ca05-f499-15dc58620e93@gnaa.net> Message-ID: <305CF206-A8A0-4E31-A2EE-994C03A79818@omniti.com> This sounds like a pkgsrc problem of some sort. Dan From danmcd at omniti.com Tue Nov 1 15:02:20 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 1 Nov 2016 11:02:20 -0400 Subject: [OmniOS-discuss] repo not showing current packages In-Reply-To: References: Message-ID: <2117DDB3-BC0D-4394-90B7-97D13A313AB4@omniti.com> > On Nov 1, 2016, at 8:41 AM, Davide Poletto wrote: > > Yeah, browsing the web page for the OmniOS r151014 repo looks indeed actually unresponsive...or very very slowly responsive, at best. '014 has grown very large, being an LTS and all that. I'm rebuilding its indices now. > Browsing other OmniOS repositories' web pages (r151010, r151012, r151016, r151018 and r151020) looks normal. 020 isn't officially out yet. Beware. Dan From davide.poletto at gmail.com Tue Nov 1 15:08:33 2016 From: davide.poletto at gmail.com (Davide Poletto) Date: Tue, 1 Nov 2016 16:08:33 +0100 Subject: [OmniOS-discuss] repo not showing current packages In-Reply-To: <2117DDB3-BC0D-4394-90B7-97D13A313AB4@omniti.com> References: <2117DDB3-BC0D-4394-90B7-97D13A313AB4@omniti.com> Message-ID: Great! about 020 repo: I cited it - as long as the others repos - only to prove that 014 repo browsing was behaving abnormally... On Tue, Nov 1, 2016 at 4:02 PM, Dan McDonald wrote: > > > On Nov 1, 2016, at 8:41 AM, Davide Poletto > wrote: > > > > Yeah, browsing the web page for the OmniOS r151014 repo looks indeed > actually unresponsive...or very very slowly responsive, at best. > > '014 has grown very large, being an LTS and all that. I'm rebuilding its > indices now. > > > Browsing other OmniOS repositories' web pages (r151010, r151012, > r151016, r151018 and r151020) looks normal. > > 020 isn't officially out yet. Beware. > > Dan > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From manuel at oetiker.ch Tue Nov 1 15:51:23 2016 From: manuel at oetiker.ch (Manuel Oetiker) Date: Tue, 1 Nov 2016 16:51:23 +0100 (CET) Subject: [OmniOS-discuss] create-lu /dev/zvol/dsk or /dev/zvol/rdsk Message-ID: <1909735809.4212357.1478015483859.JavaMail.zimbra@oetiker.ch> Hi I noticed, that I not always created my iscsi views on the rdsk device. sometime I used the dsk device for the view 600144f0888c0f000000571fece40008 161061273600 /dev/zvol/dsk/ssd-a/iscsi/xfe/disk-00 600144f0888c0f00000057308b55000c 85899345920 /dev/zvol/rdsk/ssd-a/iscsi/xde/disk-00 does this has some bad implications? should I change all views to rdsk? thanks manuel From danmcd at omniti.com Tue Nov 1 16:44:29 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 1 Nov 2016 12:44:29 -0400 Subject: [OmniOS-discuss] create-lu /dev/zvol/dsk or /dev/zvol/rdsk In-Reply-To: <1909735809.4212357.1478015483859.JavaMail.zimbra@oetiker.ch> References: <1909735809.4212357.1478015483859.JavaMail.zimbra@oetiker.ch> Message-ID: > On Nov 1, 2016, at 11:51 AM, Manuel Oetiker wrote: > > Hi > > I noticed, that I not always created my iscsi views on the rdsk device. > sometime I used the dsk device for the view > > 600144f0888c0f000000571fece40008 161061273600 /dev/zvol/dsk/ssd-a/iscsi/xfe/disk-00 > 600144f0888c0f00000057308b55000c 85899345920 /dev/zvol/rdsk/ssd-a/iscsi/xde/disk-00 > > does this has some bad implications? should > I change all views to rdsk? I'm not sure if for ZVOLs there's an underlying difference between raw and not. I'd be cautious, though, and try and unify whichever one you use. Dan From mir at miras.org Tue Nov 1 17:18:35 2016 From: mir at miras.org (Michael Rasmussen) Date: Tue, 1 Nov 2016 18:18:35 +0100 Subject: [OmniOS-discuss] create-lu /dev/zvol/dsk or /dev/zvol/rdsk In-Reply-To: References: <1909735809.4212357.1478015483859.JavaMail.zimbra@oetiker.ch> Message-ID: <20161101181835.0db5108e@sleipner.datanom.net> On Tue, 1 Nov 2016 12:44:29 -0400 Dan McDonald wrote: > > I'm not sure if for ZVOLs there's an underlying difference between raw and not. I'd be cautious, though, and try and unify whichever one you use. > AFAIK rdsk is a character device while dsk is a block device so normally you would do newfs /dev/rsdk/... and then mount /dev/dsk/... but since zfs handles raw devices (rdsk), block devices (dsk), and slices equally well I think it is not important but by convention you always handle zfs raw devices if you follow the recommendation - zfs should always be given the full device so rdsk. -- 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: I was born in a Hostess Cupcake factory before the sexual revolution! -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 163 bytes Desc: OpenPGP digital signature URL: From nshalman at omniti.com Tue Nov 1 17:36:44 2016 From: nshalman at omniti.com (Nahum Shalman) Date: Tue, 1 Nov 2016 13:36:44 -0400 Subject: [OmniOS-discuss] upgrading pkgsrc In-Reply-To: <78442e72-db59-ca05-f499-15dc58620e93@gnaa.net> References: <78442e72-db59-ca05-f499-15dc58620e93@gnaa.net> Message-ID: On Tue, Nov 1, 2016 at 1:32 AM, Geoff Nordli wrote: > I am running OmniOS v11 r151016 > You didn't mention which quarterly release of pkgsrc you were on. I tried to upgrade pkgsrc using the steps outlined here: > > https://pkgsrc.joyent.com/install-on-illumos/ Note that those say: "Upgrades aren't fully supported (we don't yet handle major configuration changes) but this procedure may work fine for many users and avoid having to re-bootstrap. If in doubt ensure you have backed up your data and configuration files. Make sure you use the same set that was originally installed." Upgrades from release to release on pkgsrc have always been tricky. The general advice from Joyent when it comes to e.g. running joyent-brand pkgsrc zones under SmartOS has been "reprovision". That means make a note of which software you use, perhaps with config management, and start each time from scratch, that said, not all hope is lost. when I tried to upgrade pkgin it gives me this error: > > root at pc-oi1:~# /opt/local/sbin/pkg_add -U pkgin > SSL support disabled > pkg_add: Can't process https://pkgsrc.joyent.com:443/ > packages/SmartOS/2016Q3/x86_64/All/pkgin*: Not owner > pkg_add: no pkg found for 'pkgin', sorry. > pkg_add: 1 package addition failed > > > If I try to do a pkgin update: > > root at pc-oi1:~# /opt/local/bin/pkgin update > reading local summary... > processing local summary... > SSL support disabled > SSL support disabled > SSL support disabled > pkgin: Could not fetch https://pkgsrc.joyent.com/pack > ages/SmartOS/2016Q3/x86_64/All/pkg_summary.gz > > > > If I use wget, i can download the gz file OK, so definitely the networking > is working fine. > > Any thoughts? > If you read through https://pkgsrc.joyent.com/changes/ for 2016Q3 you will notice that: "The pkg_install tools now support HTTPS, and both pkgin and pkg_add now fetch packages over HTTPS by default." That's a new feature for 2016Q3. Whatever quarterly release you're currently on doesn't support HTTPS yet. You may be able to use a "http://" URL (rather than "https://") to perform the initial upgrade procedure you linked originally. After pkgin and the pkg_install tools have been upgraded you should be able to switch back to "https://". All the same caveats about the upgrade process still apply. If you're upgrading a Postgres server I hope you've dumped the database in advance of attempting such an upgrade. -Nahum -------------- next part -------------- An HTML attachment was scrubbed... URL: From geoffn at gnaa.net Wed Nov 2 02:56:15 2016 From: geoffn at gnaa.net (Geoff Nordli) Date: Tue, 1 Nov 2016 19:56:15 -0700 Subject: [OmniOS-discuss] upgrading pkgsrc In-Reply-To: References: <78442e72-db59-ca05-f499-15dc58620e93@gnaa.net> Message-ID: <7b3264ba-fb37-3e21-20e6-c437458cec1c@gnaa.net> On 2016-11-01 10:36 AM, Nahum Shalman wrote: > On Tue, Nov 1, 2016 at 1:32 AM, Geoff Nordli > wrote: > > I am running OmniOS v11 r151016 > > > You didn't mention which quarterly release of pkgsrc you were on. > > I tried to upgrade pkgsrc using the steps outlined here: > > https://pkgsrc.joyent.com/install-on-illumos/ > > > > Note that those say: > "Upgrades aren't fully supported (we don't yet handle major > configuration changes) but this procedure may work fine for many users > and avoid having to re-bootstrap. If in doubt ensure you have backed > up your data and configuration files. Make sure you use the same set > that was originally installed." > > Upgrades from release to release on pkgsrc have always been tricky. > The general advice from Joyent when it comes to e.g. running > joyent-brand pkgsrc zones under SmartOS has been "reprovision". That > means make a note of which software you use, perhaps with config > management, and start each time from scratch, that said, not all hope > is lost. > > when I tried to upgrade pkgin it gives me this error: > > root at pc-oi1:~# /opt/local/sbin/pkg_add -U pkgin > SSL support disabled > pkg_add: Can't process > https://pkgsrc.joyent.com:443/packages/SmartOS/2016Q3/x86_64/All/pkgin* > : > Not owner > pkg_add: no pkg found for 'pkgin', sorry. > pkg_add: 1 package addition failed > > > If I try to do a pkgin update: > > root at pc-oi1:~# /opt/local/bin/pkgin update > reading local summary... > processing local summary... > SSL support disabled > SSL support disabled > SSL support disabled > pkgin: Could not fetch > https://pkgsrc.joyent.com/packages/SmartOS/2016Q3/x86_64/All/pkg_summary.gz > > > > > If I use wget, i can download the gz file OK, so definitely the > networking is working fine. > > Any thoughts? > > > If you read through https://pkgsrc.joyent.com/changes/ for 2016Q3 you > will notice that: > "The pkg_install tools now support HTTPS, and both pkgin and pkg_add > now fetch packages over HTTPS by default." > > That's a new feature for 2016Q3. Whatever quarterly release you're > currently on doesn't support HTTPS yet. > > You may be able to use a "http://" URL (rather than "https://") to > perform the initial upgrade procedure you linked originally. After > pkgin and the pkg_install tools have been upgraded you should be able > to switch back to "https://". > > All the same caveats about the upgrade process still apply. > > If you're upgrading a Postgres server I hope you've dumped the > database in advance of attempting such an upgrade. > > -Nahum Hi Nahum. Yes, I had backups. There were only a few packages that needed to get upgrade and I know that Joyents doesn't recommend to upgrade, but to re-provision. Jonathan Perkin gave me some guidance. In order to get back to a working state, first edit this file: /opt/local/etc/pkg_install.conf and change the PKG_PATH URL to use "http://" instead of "https://". You should then be able to follow the upgrade instructions. But I also needed to change http in: /opt/local/etc/pkgin/repositories.conf Have a great day!! Geoff -------------- next part -------------- An HTML attachment was scrubbed... URL: From glush at jet.msk.su Wed Nov 2 12:36:35 2016 From: glush at jet.msk.su (=?utf-8?B?0JTQvNC40YLRgNC40Lkg0JPQu9GD0YjQtdC90L7Qug==?=) Date: Wed, 2 Nov 2016 15:36:35 +0300 Subject: [OmniOS-discuss] smb share disappears after changing folder permissions followed by smb/server restart Message-ID: Hello, SMB server was joined AD domain, then ZFS dataset was shared using smbshare=on property. After changing permissions on dataset folder to something like this (no local users allowed): # /usr/bin/ls -lvd /tzk-data-01 d---rwx---+ 8 Administrator at tzk.local Domain Admins at tzk.local 13 Nov 2 12:12 /tzk-data-01 0:group:Domain Users at tzk.lo:list_directory/read_data/read_xattr/execute /read_attributes/read_acl/synchronize:allow 1:group:Domain Admins at tzk.l:list_directory/read_data/add_file/write_data /add_subdirectory/append_data/read_xattr/write_xattr/execute /delete_child/read_attributes/write_attributes/delete/read_acl /write_acl/write_owner/synchronize:file_inherit/dir_inherit:allow # Everything works fine till smb/server restart. After restart the share disappears from share list (smbutil view shows only c$, IPC$ and vss$). To return it back I have to change dataset folder permissions in such way that local users can access it. For example: - add read/exec permission for user:root - add read/exec permission for everyone@ - create idmap record mapping root to domain admin Is it correct behavior? What prevents smbd (running as root) to share the folder on start? When the dataset is shared i see /tzk-data-01/.zfs/shares/tzk-data-01 file is being created with following attributes: -rwxrwxrwx+ 1 root root 0 Nov 2 15:04 /tzk-data-01/.zfs/shares/tzk-data-01 0:everyone@:read_data/write_data/append_data/read_xattr/write_xattr /execute/delete_child/read_attributes/write_attributes/delete /read_acl/write_acl/write_owner/synchronize:allow When smbd unable to share the dataset after service restart - the file still exists. No id mapping is done, "idmap list" is empty. -- Dmitry Glushenok Jet Infosystems From nshalman at omniti.com Wed Nov 2 13:18:52 2016 From: nshalman at omniti.com (Nahum Shalman) Date: Wed, 2 Nov 2016 09:18:52 -0400 Subject: [OmniOS-discuss] upgrading pkgsrc In-Reply-To: <7b3264ba-fb37-3e21-20e6-c437458cec1c@gnaa.net> References: <78442e72-db59-ca05-f499-15dc58620e93@gnaa.net> <7b3264ba-fb37-3e21-20e6-c437458cec1c@gnaa.net> Message-ID: I'm glad to hear you got it working. :) I had pinged Jonathan just before sending my response and he said he would fix the instructions. He has done so [1]. Hopefully you have saved others from the same pain! -Nahum [1] https://github.com/joyent/pkgsrc.joyent.com/commit/5744de534499a1a0ad0652b4c8f04da231eb509c On Tue, Nov 1, 2016 at 10:56 PM, Geoff Nordli wrote: > > > On 2016-11-01 10:36 AM, Nahum Shalman wrote: > > On Tue, Nov 1, 2016 at 1:32 AM, Geoff Nordli wrote: > >> I am running OmniOS v11 r151016 >> > > You didn't mention which quarterly release of pkgsrc you were on. > > I tried to upgrade pkgsrc using the steps outlined here: >> >> https://pkgsrc.joyent.com/install-on-illumos/ > > > Note that those say: > "Upgrades aren't fully supported (we don't yet handle major configuration > changes) but this procedure may work fine for many users and avoid having > to re-bootstrap. If in doubt ensure you have backed up your data and > configuration files. Make sure you use the same set that was originally > installed." > > Upgrades from release to release on pkgsrc have always been tricky. The > general advice from Joyent when it comes to e.g. running joyent-brand > pkgsrc zones under SmartOS has been "reprovision". That means make a note > of which software you use, perhaps with config management, and start each > time from scratch, that said, not all hope is lost. > > when I tried to upgrade pkgin it gives me this error: >> >> root at pc-oi1:~# /opt/local/sbin/pkg_add -U pkgin >> SSL support disabled >> pkg_add: Can't process https://pkgsrc.joyent.com:443/ >> packages/SmartOS/2016Q3/x86_64/All/pkgin*: Not owner >> pkg_add: no pkg found for 'pkgin', sorry. >> pkg_add: 1 package addition failed >> >> >> If I try to do a pkgin update: >> >> root at pc-oi1:~# /opt/local/bin/pkgin update >> reading local summary... >> processing local summary... >> SSL support disabled >> SSL support disabled >> SSL support disabled >> pkgin: Could not fetch https://pkgsrc.joyent.com/pack >> ages/SmartOS/2016Q3/x86_64/All/pkg_summary.gz >> >> >> >> If I use wget, i can download the gz file OK, so definitely the >> networking is working fine. >> >> Any thoughts? >> > > If you read through https://pkgsrc.joyent.com/changes/ for 2016Q3 you > will notice that: > "The pkg_install tools now support HTTPS, and both pkgin and pkg_add now > fetch packages over HTTPS by default." > > That's a new feature for 2016Q3. Whatever quarterly release you're > currently on doesn't support HTTPS yet. > > You may be able to use a "http://" URL (rather than "https://") to > perform the initial upgrade procedure you linked originally. After pkgin > and the pkg_install tools have been upgraded you should be able to switch > back to "https://". > > All the same caveats about the upgrade process still apply. > > If you're upgrading a Postgres server I hope you've dumped the database in > advance of attempting such an upgrade. > > -Nahum > > > Hi Nahum. > > Yes, I had backups. > > There were only a few packages that needed to get upgrade and I know that > Joyents doesn't recommend to upgrade, but to re-provision. > > Jonathan Perkin gave me some guidance. > > In order to get back to a working state, first edit this file: > > /opt/local/etc/pkg_install.conf > > and change the PKG_PATH URL to use "http://" instead of "https://". > You should then be able to follow the upgrade instructions. > > But I also needed to change http in: /opt/local/etc/pkgin/repositories.conf > > > Have a great day!! > > Geoff > > > > > _______________________________________________ > 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 Nov 2 17:35:47 2016 From: hasslerd at gmx.li (Dominik Hassler) Date: Wed, 2 Nov 2016 18:35:47 +0100 Subject: [OmniOS-discuss] OmniOS/Illumos and entropy Message-ID: Hi, just a general question. Where does Omnios/Illumos get it's entropy from to feed /dev/random. This is just out of curiosity since I can drain any /dev/random on e.g. a linux system (w/o a HWRNG) easily by doing: $ cat /dev/random But on OmniOS it is not possible to drain the entropy pool at all on /dev/random. Thanks for any insights. From danmcd at omniti.com Wed Nov 2 17:53:25 2016 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 2 Nov 2016 13:53:25 -0400 Subject: [OmniOS-discuss] OmniOS/Illumos and entropy In-Reply-To: References: Message-ID: <258E3432-05B8-48D5-92A4-289DFA1C0419@omniti.com> > On Nov 2, 2016, at 1:35 PM, Dominik Hassler wrote: > > Hi, > > just a general question. Where does Omnios/Illumos get it's entropy from to feed /dev/random. > > This is just out of curiosity since I can drain any /dev/random on e.g. a linux system (w/o a HWRNG) easily by doing: > > $ cat /dev/random > > But on OmniOS it is not possible to drain the entropy pool at all on /dev/random. Use the source! http://src.illumos.org/source/xref/illumos-gate/usr/src/uts/common/io/random.c#227 That's the /dev/random read() routine. Note how it calls kcf_rnd_get_bytes()? And note the block() parameter depends on how you open it? You'll get to rnd_get_bytes(), which calls rngprov_getbytes(), which then calls all of the random providers to get more. In this case, it's swrand. Here's the core routine: http://src.illumos.org/source/xref/illumos-gate/usr/src/uts/common/crypto/io/swrand.c#swrand_get_entropy A lot comes from physical memory trolling, e.g.: http://src.illumos.org/source/xref/illumos-gate/usr/src/uts/common/crypto/io/swrand.c#656 One could add another RNG provider using CPU features, but given recent revelations about the some of Intel ones, I'm not so sure how good of an idea that might be. Dan From josh at sysmgr.org Wed Nov 2 18:17:56 2016 From: josh at sysmgr.org (Joshua M. Clulow) Date: Wed, 2 Nov 2016 11:17:56 -0700 Subject: [OmniOS-discuss] OmniOS/Illumos and entropy In-Reply-To: <258E3432-05B8-48D5-92A4-289DFA1C0419@omniti.com> References: <258E3432-05B8-48D5-92A4-289DFA1C0419@omniti.com> Message-ID: On 2 November 2016 at 10:53, Dan McDonald wrote: > One could add another RNG provider using CPU features, but given recent revelations about the some of Intel ones, I'm not so sure how good of an idea that might be. In general, as long as we're combining _multiple_ potentially untrustworthy but independent sources, and using them only to occasionally _reseed_ a high quality PRNG rather than directly using the "random" output, we should be safe. -- Joshua M. Clulow UNIX Admin/Developer http://blog.sysmgr.org From glush at jet.msk.su Thu Nov 3 10:51:42 2016 From: glush at jet.msk.su (=?utf-8?B?0JTQvNC40YLRgNC40Lkg0JPQu9GD0YjQtdC90L7Qug==?=) Date: Thu, 3 Nov 2016 13:51:42 +0300 Subject: [OmniOS-discuss] smb share disappears after changing folder permissions followed by smb/server restart In-Reply-To: References: Message-ID: <5E74C2F1-EAD2-4C47-A64B-A0161F3CAA45@jet.msk.su> It looks like assigning file_dac_read privilege solves the problem. Following command can be used as a workaround: # ppriv -s +file_dac_read $(pgrep -z global smbd) && kill -HUP $(pgrep -z global smbd) -- Dmitry Glushenok Jet Infosystems > 2 ????. 2016 ?., ? 15:36, ??????? ???????? ???????(?): > > Hello, > > SMB server was joined AD domain, then ZFS dataset was shared using smbshare=on property. After changing permissions on dataset folder to something like this (no local users allowed): > > # /usr/bin/ls -lvd /tzk-data-01 > d---rwx---+ 8 Administrator at tzk.local Domain Admins at tzk.local 13 Nov 2 12:12 /tzk-data-01 > 0:group:Domain Users at tzk.lo:list_directory/read_data/read_xattr/execute > /read_attributes/read_acl/synchronize:allow > 1:group:Domain Admins at tzk.l:list_directory/read_data/add_file/write_data > /add_subdirectory/append_data/read_xattr/write_xattr/execute > /delete_child/read_attributes/write_attributes/delete/read_acl > /write_acl/write_owner/synchronize:file_inherit/dir_inherit:allow > # > > Everything works fine till smb/server restart. After restart the share disappears from share list (smbutil view shows only c$, IPC$ and vss$). To return it back I have to change dataset folder permissions in such way that local users can access it. For example: > - add read/exec permission for user:root > - add read/exec permission for everyone@ > - create idmap record mapping root to domain admin > > Is it correct behavior? What prevents smbd (running as root) to share the folder on start? > > When the dataset is shared i see /tzk-data-01/.zfs/shares/tzk-data-01 file is being created with following attributes: > > -rwxrwxrwx+ 1 root root 0 Nov 2 15:04 /tzk-data-01/.zfs/shares/tzk-data-01 > 0:everyone@:read_data/write_data/append_data/read_xattr/write_xattr > /execute/delete_child/read_attributes/write_attributes/delete > /read_acl/write_acl/write_owner/synchronize:allow > > When smbd unable to share the dataset after service restart - the file still exists. No id mapping is done, "idmap list" is empty. > > -- > Dmitry Glushenok > Jet Infosystems > From jimklimov at cos.ru Thu Nov 3 12:58:57 2016 From: jimklimov at cos.ru (Jim Klimov) Date: Thu, 03 Nov 2016 13:58:57 +0100 Subject: [OmniOS-discuss] smb share disappears after changing folder permissions followed by smb/server restart In-Reply-To: <5E74C2F1-EAD2-4C47-A64B-A0161F3CAA45@jet.msk.su> References: <5E74C2F1-EAD2-4C47-A64B-A0161F3CAA45@jet.msk.su> Message-ID: <887CFFEA-124C-47EA-99EB-22F318D816A4@cos.ru> 3 ?????? 2016??. 11:51:42 CET, "??????? ????????" ?????: >It looks like assigning file_dac_read privilege solves the problem. >Following command can be used as a workaround: > ># ppriv -s +file_dac_read $(pgrep -z global smbd) && kill -HUP $(pgrep >-z global smbd) > >-- >Dmitry Glushenok >Jet Infosystems > >> 2 ????. 2016 ?., ? 15:36, ??????? ???????? >???????(?): >> >> Hello, >> >> SMB server was joined AD domain, then ZFS dataset was shared using >smbshare=on property. After changing permissions on dataset folder to >something like this (no local users allowed): >> >> # /usr/bin/ls -lvd /tzk-data-01 >> d---rwx---+ 8 Administrator at tzk.local Domain Admins at tzk.local >13 Nov 2 12:12 /tzk-data-01 >> 0:group:Domain >Users at tzk.lo:list_directory/read_data/read_xattr/execute >> /read_attributes/read_acl/synchronize:allow >> 1:group:Domain >Admins at tzk.l:list_directory/read_data/add_file/write_data >> /add_subdirectory/append_data/read_xattr/write_xattr/execute >> >/delete_child/read_attributes/write_attributes/delete/read_acl >> >/write_acl/write_owner/synchronize:file_inherit/dir_inherit:allow >> # >> >> Everything works fine till smb/server restart. After restart the >share disappears from share list (smbutil view shows only c$, IPC$ and >vss$). To return it back I have to change dataset folder permissions in >such way that local users can access it. For example: >> - add read/exec permission for user:root >> - add read/exec permission for everyone@ >> - create idmap record mapping root to domain admin >> >> Is it correct behavior? What prevents smbd (running as root) to share >the folder on start? >> >> When the dataset is shared i see /tzk-data-01/.zfs/shares/tzk-data-01 >file is being created with following attributes: >> >> -rwxrwxrwx+ 1 root root 0 Nov 2 15:04 >/tzk-data-01/.zfs/shares/tzk-data-01 >> >0:everyone@:read_data/write_data/append_data/read_xattr/write_xattr >> /execute/delete_child/read_attributes/write_attributes/delete >> /read_acl/write_acl/write_owner/synchronize:allow >> >> When smbd unable to share the dataset after service restart - the >file still exists. No id mapping is done, "idmap list" is empty. >> >> -- >> Dmitry Glushenok >> Jet Infosystems >> > >_______________________________________________ >OmniOS-discuss mailing list >OmniOS-discuss at lists.omniti.com >http://lists.omniti.com/mailman/listinfo/omnios-discuss If that is the case, consider fixing up the method_context privileges in the SMF service for the smb/server. Good luck and thanks for sharing ;) -- Typos courtesy of K-9 Mail on my Samsung Android From glush at jet.msk.su Thu Nov 3 13:23:27 2016 From: glush at jet.msk.su (=?utf-8?B?0JTQvNC40YLRgNC40Lkg0JPQu9GD0YjQtdC90L7Qug==?=) Date: Thu, 3 Nov 2016 16:23:27 +0300 Subject: [OmniOS-discuss] smb share disappears after changing folder permissions followed by smb/server restart In-Reply-To: <887CFFEA-124C-47EA-99EB-22F318D816A4@cos.ru> References: <5E74C2F1-EAD2-4C47-A64B-A0161F3CAA45@jet.msk.su> <887CFFEA-124C-47EA-99EB-22F318D816A4@cos.ru> Message-ID: smbd makes it's own privilege set at start: smbd_daemonize_fini(int fd, int exit_status) ... priv_basicset(pset); /* list of privileges for smbd */ (void) priv_addset(pset, PRIV_NET_MAC_AWARE); (void) priv_addset(pset, PRIV_NET_PRIVADDR); (void) priv_addset(pset, PRIV_PROC_AUDIT); (void) priv_addset(pset, PRIV_SYS_DEVICES); (void) priv_addset(pset, PRIV_SYS_SMB); (void) priv_addset(pset, PRIV_SYS_MOUNT); priv_inverse(pset); /* turn off unneeded privileges */ (void) setppriv(PRIV_OFF, PRIV_EFFECTIVE, pset); ... So, SMF context will not work. For now I've set "ppriv && SIGHUP" in a separate service, which is launched just after smbd. -- Dmitry Glushenok Jet Infosystems > 3 ????. 2016 ?., ? 15:58, Jim Klimov ???????(?): > > If that is the case, consider fixing up the method_context privileges in the SMF service for the smb/server. > Good luck and thanks for sharing ;) From glush at jet.msk.su Thu Nov 3 17:23:50 2016 From: glush at jet.msk.su (=?koi8-r?B?5M3J1NLJyiDnzNXbxc7Pyw==?=) Date: Thu, 3 Nov 2016 20:23:50 +0300 Subject: [OmniOS-discuss] smb share disappears after changing folder permissions followed by smb/server restart In-Reply-To: References: <5E74C2F1-EAD2-4C47-A64B-A0161F3CAA45@jet.msk.su> <887CFFEA-124C-47EA-99EB-22F318D816A4@cos.ru> Message-ID: There is a patch: https://www.illumos.org/issues/7529 -- Dmitry Glushenok Jet Infosystems > 3 ????. 2016 ?., ? 16:23, ??????? ???????? ???????(?): > > smbd makes it's own privilege set at start: > > smbd_daemonize_fini(int fd, int exit_status) > ... > priv_basicset(pset); > > /* list of privileges for smbd */ > (void) priv_addset(pset, PRIV_NET_MAC_AWARE); > (void) priv_addset(pset, PRIV_NET_PRIVADDR); > (void) priv_addset(pset, PRIV_PROC_AUDIT); > (void) priv_addset(pset, PRIV_SYS_DEVICES); > (void) priv_addset(pset, PRIV_SYS_SMB); > (void) priv_addset(pset, PRIV_SYS_MOUNT); > > priv_inverse(pset); > > /* turn off unneeded privileges */ > (void) setppriv(PRIV_OFF, PRIV_EFFECTIVE, pset); > ... > > So, SMF context will not work. For now I've set "ppriv && SIGHUP" in a separate service, which is launched just after smbd. > > -- > Dmitry Glushenok > Jet Infosystems > >> 3 ????. 2016 ?., ? 15:58, Jim Klimov ???????(?): >> >> If that is the case, consider fixing up the method_context privileges in the SMF service for the smb/server. >> Good luck and thanks for sharing ;) > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jboren at drakecooper.com Thu Nov 3 22:52:20 2016 From: jboren at drakecooper.com (Joseph Boren) Date: Thu, 3 Nov 2016 16:52:20 -0600 Subject: [OmniOS-discuss] replacing in-kernel CIFS server with samba Message-ID: Greetings, For several reasons, I'd like to experiment with using samba4 instead of the built in cifs server. I know I can build samba from source and have found some resources for doing that, but I wondered what I'm supposed to do about the built in cifs server, is it sufficient just to disable that service? Or do I need to remove it, and if so, any pointers on how to do that? Thanks a ton, Joseph Boren IT Specialist +c: 208.891.2128 + 416 S 8th St., Boise, ID 83702 + drakecooper.com *Read our minds *Blog Facebook , Twitter , Linkedin , Instagram -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Fri Nov 4 02:02:15 2016 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 3 Nov 2016 22:02:15 -0400 Subject: [OmniOS-discuss] replacing in-kernel CIFS server with samba In-Reply-To: References: Message-ID: Disabling will be enough. Dan From jboren at drakecooper.com Fri Nov 4 03:25:25 2016 From: jboren at drakecooper.com (Joseph Boren) Date: Thu, 3 Nov 2016 21:25:25 -0600 Subject: [OmniOS-discuss] replacing in-kernel CIFS server with samba In-Reply-To: References: Message-ID: Perfect. Thank you, Dan. Best regards, Joe On Thursday, November 3, 2016, Dan McDonald > wrote: > Disabling will be enough. > > Dan > > -- Joseph Boren IT Specialist +c: 208.891.2128 + 416 S 8th St., Boise, ID 83702 + drakecooper.com *Read our minds *Blog Facebook , Twitter , Linkedin , Instagram -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Fri Nov 4 15:26:09 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 4 Nov 2016 11:26:09 -0400 Subject: [OmniOS-discuss] OmniOS r151020 is now out! Message-ID: Start with the release notes: http://omnios.omniti.com/wiki.php/ReleaseNotes/r151020 The big highlights include: - LX Zones Beta --> LX Zones can now run in OmniOS. They have their own wiki page, off the Release Notes one above. https://omnios.omniti.com/wiki.php/LXZones Not all LX interfaces and methods are frozen, and are subject to change between now and r151022. - SunSSH end-of-life (NOTE the Upgrading section below). - ACPI Version 6 - Initial NVMe 1.1 Support - VIM 8.0 NOTE that we froze r151020 prior to the BSD Loader push in illumos-gate. We did so to not delay this release. This does mean that the next release, r151022, which is also the next LTS, will be late, and may have other substantial changes. OmniOS r151020 is the new stable release. OmniOS r151018 becomes the "old stable" and will receive critical bugfixes and/or security patches for another six months. OmniOS r151014 continues to be the long-term-support (LTS) release. OmniOS r151016 has officially reached END OF SERVICE LIFE, having finished its time as "old stable". With SunSSH being end-of-lifed, upgrading to r151020 may be difficult for people still using SunSSH. There's a new upgrading page one can get to off the Release Notes: https://omnios.omniti.com/wiki.php/Upgrade_to_r151020 If you've already made the switch to SunSSH, upgrading to r151020 should be no different than any other post-r151014 upgrade. If you are still using SunSSH, read the above link very carefully. We recommend you make the switch to OpenSSH before upgrading, but we understand if you can't. Thanks as always to the OmniOS community, and OmniTI colleagues including (but not limited to): Dale Ghent, Marissa Murphy, and Robert Treat (who convinced me to port LX). Thanks to Joyent staffers Jerry Jelinek and Patrick Mooney, whose LX work AND LX-porting-assitance were invaluable. Happy upgrading! Dan McDonald -- OmniOS Engineering From pasztor at sagv5.gyakg.u-szeged.hu Fri Nov 4 15:58:45 2016 From: pasztor at sagv5.gyakg.u-szeged.hu (=?iso-8859-2?Q?P=C1SZTOR_Gy=F6rgy?=) Date: Fri, 4 Nov 2016 16:58:45 +0100 Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: References: Message-ID: <20161104155845.GB4277@linux.gyakg.u-szeged.hu> Hi, "Dan McDonald" ?rta 2016-11-04 11:26-kor: > NOTE that we froze r151020 prior to the BSD Loader push in illumos-gate. We did so to not delay this release. This does mean that the next release, r151022, which is also the next LTS, will be late, and may have other substantial changes. How big will be that change? Can grub still work in 151022, or only the BSD Loader will be supported from that point? Cheers, Gyu From danmcd at omniti.com Fri Nov 4 16:25:37 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 4 Nov 2016 12:25:37 -0400 Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: <20161104155845.GB4277@linux.gyakg.u-szeged.hu> References: <20161104155845.GB4277@linux.gyakg.u-szeged.hu> Message-ID: I'm not 100% sure yet, but I'm leaning toward not supporting grub at all in 022. Dan Sent from my iPhone (typos, autocorrect, and all) > On Nov 4, 2016, at 11:58 AM, P?SZTOR Gy?rgy wrote: > > Hi, > > "Dan McDonald" ?rta 2016-11-04 11:26-kor: >> NOTE that we froze r151020 prior to the BSD Loader push in illumos-gate. We did so to not delay this release. This does mean that the next release, r151022, which is also the next LTS, will be late, and may have other substantial changes. > > How big will be that change? > Can grub still work in 151022, > or only the BSD Loader will be supported from that point? > > Cheers, > Gyu From gary at genashor.com Fri Nov 4 17:23:12 2016 From: gary at genashor.com (Gary Gendel) Date: Fri, 04 Nov 2016 13:23:12 -0400 Subject: [OmniOS-discuss] Can't bring up r151020 Message-ID: <4e4acde5-8751-88df-314b-558655a8a602@genashor.com> I haven't been able to upgrade since the July patches. I believe it has something to do with ? missing MCFG table should lead to I/O PCIe config access (illumos 6859) Which seems to be tripping up on my machine. The symptom is that the console displays (this is a capture before r151020 which behaves the same): SunOS Release 5.11 Version omnios-r151818-95eaa7e 64-bit Copyright (c) 1983, 2010, Oracle and/or its affiliates. All rights reserved. WARNING: illegal PCI request: offset = f48, size = 2 WARNING: illegal PCI request: offset = 100, size = 4 WARNING: illegal PCI request: offset = f48, size = 2 WARNING: illegal PCI request: offset = 100, size = 4 WARNING: illegal PCI request: offset = f48, size = 2 WARNING: illegal PCI request: offset = 100, size = 4 WARNING: illegal PCI request: offset = f48, size = 2 WARNING: illegal PCI request: offset = 100, size = 4 WARNING: illegal PCI request: offset = 100, size = 4 WARNING: illegal PCI request: offset = 100, size = 4 and then hangs. The system logs are empty so it happens early in the boot process. If anyone has a suggestion on how to get past this, I'm all ears. I'm also open to picking up a new machine like the HP Gen8. I'm using this as a (very) small SOHO server to do email, web, and file services. The current machine is a 1U unit which has a motherboard similar to the SunFire v20, 4 disks: 2 mirrored 5TB(archive), 2 mirrored ~300GB drives (rpool), and 8 GB of RAM which works well for my purpose. The goal is to configure the new machine similarly. I'm looking for recommendations on what configuration I should get and any gotchas to be concerned about. I'm looking forward to the reduced noise and power consumption. :) Thanks in advance, Gary -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3703 bytes Desc: S/MIME Cryptographic Signature URL: From danmcd at omniti.com Fri Nov 4 17:34:44 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 4 Nov 2016 13:34:44 -0400 Subject: [OmniOS-discuss] Can't bring up r151020 In-Reply-To: <4e4acde5-8751-88df-314b-558655a8a602@genashor.com> References: <4e4acde5-8751-88df-314b-558655a8a602@genashor.com> Message-ID: <48F4E4CE-2F36-459E-8FC6-F0267074B48F@omniti.com> > On Nov 4, 2016, at 1:23 PM, Gary Gendel wrote: > > I haven't been able to upgrade since the July patches. I believe it has something to do with > > ? missing MCFG table should lead to I/O PCIe config access (illumos 6859) > > Which seems to be tripping up on my machine. The symptom is that the console displays (this is a capture before r151020 which behaves the same): > > SunOS Release 5.11 Version omnios-r151818-95eaa7e 64-bit > Copyright (c) 1983, 2010, Oracle and/or its affiliates. All rights reserved. > WARNING: illegal PCI request: offset = f48, size = 2 > WARNING: illegal PCI request: offset = 100, size = 4 > WARNING: illegal PCI request: offset = f48, size = 2 > WARNING: illegal PCI request: offset = 100, size = 4 > WARNING: illegal PCI request: offset = f48, size = 2 > WARNING: illegal PCI request: offset = 100, size = 4 > WARNING: illegal PCI request: offset = f48, size = 2 > WARNING: illegal PCI request: offset = 100, size = 4 > WARNING: illegal PCI request: offset = 100, size = 4 > WARNING: illegal PCI request: offset = 100, size = 4 > > and then hangs. The system logs are empty so it happens early in the boot process. If anyone has a suggestion on how to get past this, I'm all ears. IIRC, you have a potentially problematic bge on this guy. Can you boot without bge temporarily? I know it'll kill networking on your box, but can you edit the "unix" grub line and add: -B disable-bge after "/platform/..../unix ". IF you can get there, we may be able to proceed. Dan From mailinglists at qutic.com Fri Nov 4 19:44:44 2016 From: mailinglists at qutic.com (qutic development) Date: Fri, 4 Nov 2016 20:44:44 +0100 Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: References: Message-ID: <35FEF30B-8CD3-4992-BFB1-174C1B80F8FC@qutic.com> > Am 04.11.2016 um 16:26 schrieb Dan McDonald : > > Happy upgrading! Thanks Dan for this release and all the good work! Do the LX Zones need an intel processor? Or will it work with amd too? - Stefan From danmcd at omniti.com Fri Nov 4 19:45:55 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 4 Nov 2016 15:45:55 -0400 Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: <35FEF30B-8CD3-4992-BFB1-174C1B80F8FC@qutic.com> References: <35FEF30B-8CD3-4992-BFB1-174C1B80F8FC@qutic.com> Message-ID: <32A49E76-1B1D-43F1-9139-F341D304EC3B@omniti.com> > On Nov 4, 2016, at 3:44 PM, qutic development wrote: > > >> Am 04.11.2016 um 16:26 schrieb Dan McDonald : >> >> Happy upgrading! > > Thanks Dan for this release and all the good work! > > Do the LX Zones need an intel processor? Or will it work with amd too? LX Zones should work with any processor stock OmniOS runs on -- it's a zone, not dependent on HW like KVM. Dan From danmcd at omniti.com Fri Nov 4 19:45:55 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 4 Nov 2016 15:45:55 -0400 Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: <35FEF30B-8CD3-4992-BFB1-174C1B80F8FC@qutic.com> References: <35FEF30B-8CD3-4992-BFB1-174C1B80F8FC@qutic.com> Message-ID: <32A49E76-1B1D-43F1-9139-F341D304EC3B@omniti.com> > On Nov 4, 2016, at 3:44 PM, qutic development wrote: > > >> Am 04.11.2016 um 16:26 schrieb Dan McDonald : >> >> Happy upgrading! > > Thanks Dan for this release and all the good work! > > Do the LX Zones need an intel processor? Or will it work with amd too? LX Zones should work with any processor stock OmniOS runs on -- it's a zone, not dependent on HW like KVM. Dan From jimklimov at cos.ru Fri Nov 4 21:41:23 2016 From: jimklimov at cos.ru (Jim Klimov) Date: Fri, 04 Nov 2016 22:41:23 +0100 Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: <20161104155845.GB4277@linux.gyakg.u-szeged.hu> References: <20161104155845.GB4277@linux.gyakg.u-szeged.hu> Message-ID: 4 ?????? 2016??. 16:58:45 CET, "P?SZTOR Gy?rgy" ?????: >Hi, > >"Dan McDonald" ?rta 2016-11-04 11:26-kor: >> NOTE that we froze r151020 prior to the BSD Loader push in >illumos-gate. We did so to not delay this release. This does mean >that the next release, r151022, which is also the next LTS, will be >late, and may have other substantial changes. > >How big will be that change? >Can grub still work in 151022, >or only the BSD Loader will be supported from that point? > >Cheers, >Gyu >_______________________________________________ >OmniOS-discuss mailing list >OmniOS-discuss at lists.omniti.com >http://lists.omniti.com/mailman/listinfo/omnios-discuss Can't speak for OmniOS how they'd integrate the change, but in OI/Hipster the Loader is added as default recently, although with a specially configurable option to still use grub for now. It is expected that grub will be removed roughly in a year, though. Jim -- Typos courtesy of K-9 Mail on my Samsung Android From bfriesen at simple.dallas.tx.us Fri Nov 4 22:35:44 2016 From: bfriesen at simple.dallas.tx.us (Bob Friesenhahn) Date: Fri, 4 Nov 2016 17:35:44 -0500 (CDT) Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: References: <20161104155845.GB4277@linux.gyakg.u-szeged.hu> Message-ID: On Fri, 4 Nov 2016, Jim Klimov wrote: > > Can't speak for OmniOS how they'd integrate the change, but in > OI/Hipster the Loader is added as default recently, although with a > specially configurable option to still use grub for now. It is > expected that grub will be removed roughly in a year, though. The main problem with switching to the new loader seems to be that if anything goes wrong and one wants to use a previous boot environment, special incantations are needed to boot that previous boot environment since only grub knows how to automatically boot from the previous boot environment (the boot environment is tailored for the loader). This adds a risk factor that we were insulated from up to now. Bob -- Bob Friesenhahn bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ From omnios at citrus-it.net Sat Nov 5 11:10:47 2016 From: omnios at citrus-it.net (Andy Fiddaman) Date: Sat, 5 Nov 2016 11:10:47 +0000 (UTC) Subject: [OmniOS-discuss] [developer] OmniOS r151020 is now out! In-Reply-To: References: Message-ID: On Fri, 4 Nov 2016, Dan McDonald wrote: ; - VIM 8.0 Thanks Dan - we've now upgraded half of our servers without zones on them without any problems. One note for anyone who has created their own system-wide vimrc file - /usr/share/vim/vimrc - vim8 will override options set in there with the defaults in /usr/share/vim/vim80/defaults.vim. That's a change in behaviour from earlier vim versions. To stop the defaults.vim file from being used at all, add this to the system vimrc: let skip_defaults_vim=1 Another solution for anyone who wants to load the system defaults but then override specific options is the following. In this example I want to turn off the 'ruler' option for everyone by default but still take advantage of the defaults. # cat /usr/share/vim/vimrc unlet! skip_defaults_vim source $VIMRUNTIME/defaults.vim let skip_defaults_vim=1 set noruler Regards, Andy -- Citrus IT Limited | +44 (0)870 199 8000 | enquiries at citrus-it.co.uk Rock House Farm | Green Moor | Wortley | Sheffield | S35 7DQ Registered in England and Wales | Company number 4899123 From danmcd at omniti.com Sat Nov 5 17:11:54 2016 From: danmcd at omniti.com (Dan McDonald) Date: Sat, 5 Nov 2016 13:11:54 -0400 Subject: [OmniOS-discuss] [developer] OmniOS r151020 is now out! In-Reply-To: References: Message-ID: Narrowing to OmniOS-discuss... I'm glad you shared this. I'm an emacs person for big things, but vi for small, and was wondering if vi/vim people had advice for the new world of VIM 8.0. Thanks! Dan Sent from my iPhone (typos, autocorrect, and all) > On Nov 5, 2016, at 7:10 AM, Andy Fiddaman wrote: > > > On Fri, 4 Nov 2016, Dan McDonald wrote: > ; - VIM 8.0 > > Thanks Dan - we've now upgraded half of our servers without zones on them > without any problems. > > One note for anyone who has created their own system-wide vimrc file - > /usr/share/vim/vimrc - vim8 will override options set in there with > the defaults in /usr/share/vim/vim80/defaults.vim. That's a change > in behaviour from earlier vim versions. > > To stop the defaults.vim file from being used at all, add this to the system > vimrc: > > let skip_defaults_vim=1 > > Another solution for anyone who wants to load the system defaults but then > override specific options is the following. In this example I want to > turn off the 'ruler' option for everyone by default but still take > advantage of the defaults. > > # cat /usr/share/vim/vimrc > unlet! skip_defaults_vim > source $VIMRUNTIME/defaults.vim > let skip_defaults_vim=1 > set noruler > > Regards, > > Andy > -- > Citrus IT Limited | +44 (0)870 199 8000 | enquiries at citrus-it.co.uk > Rock House Farm | Green Moor | Wortley | Sheffield | S35 7DQ > Registered in England and Wales | Company number 4899123 > From lists at marzocchi.net Sun Nov 6 00:02:39 2016 From: lists at marzocchi.net (Olaf Marzocchi) Date: Sun, 6 Nov 2016 01:02:39 +0100 Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: References: Message-ID: <79d26d32-e0a3-c473-7443-46a358c2a1f6@marzocchi.net> On 04/11/2016 16:26, Dan McDonald wrote: > With SunSSH being end-of-lifed, upgrading to r151020 may be difficult for people still using SunSSH. There's a new upgrading page one can get to off the Release Notes: > > https://omnios.omniti.com/wiki.php/Upgrade_to_r151020 Hi, I started updating my server, therefore OpenSSH first. After that one I restarted the server. I can login with my pass but two factor authentication doesn't work anymore, I'm never asked for it. What I did: I used the sshd_config.new as starting point with these changes: PasswordAuthentication no UsePAM yes #ChallengeResponseAuthentication yes (untouched, it's the default) I also tried AuthenticationMethods publickey keyboard-interactive but it didn't help. Then I went to pam.conf and did s/sshd-kbdint/sshd/ as written in the guide. This is what I have now at the very end of said file: # SSH keyboard interactive, so that keys don't ask for verification code sshd auth requisite pam_authtok_get.so.1 sshd auth required pam_dhkeys.so.1 sshd auth required pam_google_authenticator.so nullok sshd auth required pam_unix_cred.so.1 sshd auth required pam_unix_auth.so.1 When I login via SSH, I'm not asked for the verification code. I searched online for similar guides but it looks like I'm doing it right so I don't know how to proceed. Anyone able to help? thanks Cheers, Olaf From omnios at citrus-it.net Sun Nov 6 13:15:11 2016 From: omnios at citrus-it.net (Andy Fiddaman) Date: Sun, 6 Nov 2016 13:15:11 +0000 (UTC) Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: <79d26d32-e0a3-c473-7443-46a358c2a1f6@marzocchi.net> References: <79d26d32-e0a3-c473-7443-46a358c2a1f6@marzocchi.net> Message-ID: On Sun, 6 Nov 2016, Olaf Marzocchi wrote: ; Hi, ; I started updating my server, therefore OpenSSH first. After that one I ; restarted the server. ; I can login with my pass but two factor authentication doesn't work anymore, ; I'm never asked for it. I've just checked one of my servers that uses 2FA and I have both sshd-kbdint and sshd at the end of pam.conf. Perhaps you need both? sshd-kbdint auth requisite pam_authtok_get.so.1 sshd-kbdint auth required pam_dhkeys.so.1 sshd-kbdint auth required pam_unix_cred.so.1 sshd-kbdint auth required pam_unix_auth.so.1 sshd-kbdint auth required /opt/CITotp/lib/pam_mobile_otp.so sshd auth requisite pam_authtok_get.so.1 sshd auth required pam_dhkeys.so.1 sshd auth required pam_unix_cred.so.1 sshd auth required pam_unix_auth.so.1 sshd auth required /opt/CITotp/lib/pam_mobile_otp.so Debug output from sshd says: debug3: /etc/ssh/sshd_config:100 setting UsePAM yes /etc/ssh/sshd_config line 100: ignoring UsePAM option value. This option is always on. debug3: PAM service is sshd-none debug3: PAM service is sshd-pubkey debug3: PAM service is sshd-kbdint -- Citrus IT Limited | +44 (0)870 199 8000 | enquiries at citrus-it.co.uk Rock House Farm | Green Moor | Wortley | Sheffield | S35 7DQ Registered in England and Wales | Company number 4899123 From lists at marzocchi.net Sun Nov 6 15:38:38 2016 From: lists at marzocchi.net (Olaf Marzocchi) Date: Sun, 6 Nov 2016 16:38:38 +0100 Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: References: <79d26d32-e0a3-c473-7443-46a358c2a1f6@marzocchi.net> Message-ID: <5f8e6b9e-54f8-bc3a-b39c-24ad75feecfb@marzocchi.net> Hi Andy, thanks for the help, especially on Sunday! I blindly followed the instructions provided on https://omnios.omniti.com/wiki.php/Upgrade_to_r151020 but yes, the recommendation of s/sshd-kbdint/sshd/ was (at least in my case) wrong and by reverting to sshd-kbdint I got the 2FA back. Thanks! Olaf On 06/11/2016 14:15, Andy Fiddaman wrote: > > > On Sun, 6 Nov 2016, Olaf Marzocchi wrote: > > ; Hi, > ; I started updating my server, therefore OpenSSH first. After that one I > ; restarted the server. > ; I can login with my pass but two factor authentication doesn't work anymore, > ; I'm never asked for it. > > I've just checked one of my servers that uses 2FA and I have both > sshd-kbdint and sshd at the end of pam.conf. Perhaps you need both? > > sshd-kbdint auth requisite pam_authtok_get.so.1 > sshd-kbdint auth required pam_dhkeys.so.1 > sshd-kbdint auth required pam_unix_cred.so.1 > sshd-kbdint auth required pam_unix_auth.so.1 > sshd-kbdint auth required /opt/CITotp/lib/pam_mobile_otp.so > > sshd auth requisite pam_authtok_get.so.1 > sshd auth required pam_dhkeys.so.1 > sshd auth required pam_unix_cred.so.1 > sshd auth required pam_unix_auth.so.1 > sshd auth required /opt/CITotp/lib/pam_mobile_otp.so > > > Debug output from sshd says: > > debug3: /etc/ssh/sshd_config:100 setting UsePAM yes > /etc/ssh/sshd_config line 100: ignoring UsePAM option value. This option is always on. > debug3: PAM service is sshd-none > debug3: PAM service is sshd-pubkey > debug3: PAM service is sshd-kbdint > From omnios at citrus-it.net Mon Nov 7 10:23:15 2016 From: omnios at citrus-it.net (Andy Fiddaman) Date: Mon, 7 Nov 2016 10:23:15 +0000 (UTC) Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: <5f8e6b9e-54f8-bc3a-b39c-24ad75feecfb@marzocchi.net> References: <79d26d32-e0a3-c473-7443-46a358c2a1f6@marzocchi.net> <5f8e6b9e-54f8-bc3a-b39c-24ad75feecfb@marzocchi.net> Message-ID: On Sun, 6 Nov 2016, Olaf Marzocchi wrote: ; Hi Andy, ; thanks for the help, especially on Sunday! ; ; I blindly followed the instructions provided on ; https://omnios.omniti.com/wiki.php/Upgrade_to_r151020 but yes, the ; recommendation of s/sshd-kbdint/sshd/ was (at least in my case) wrong and by ; reverting to sshd-kbdint I got the 2FA back. That's strange because the advice in the wiki came from me originally! I definitely needed to update my pam.conf when I first switched to OpenSSH... reaper# (38) pkg history -l | ggrep -B5 -A5 openssh | head -7 New Boot Env. UUID: (None) Snapshot: (None) Start Time: 2015-11-20T12:58:24 End Time: 2015-11-20T12:58:43 Total Time: 0:00:19 Command: /usr/bin/pkg install --no-backup-be --reject pkg:/network/ss h --reject pkg:/network/ssh/ssh-key --reject pkg:/service/network/ssh pkg:/netwo rk/openssh pkg:/network/openssh-server Release Notes: No That was in November 2015 but there's a patch at https://github.com/omniti-labs/omnios-build/blob/master/build/openssh/patches/0011-PAM-enhancements-for-Solaris.patch which looks like it changes the pam service back to sshd-kbdint (assuming default options in sshd_config). That was commited in March 2016 so it seems that the pam.conf update is no longer required. Regards, Andy -- Citrus IT Limited | +44 (0)870 199 8000 | enquiries at citrus-it.co.uk Rock House Farm | Green Moor | Wortley | Sheffield | S35 7DQ Registered in England and Wales | Company number 4899123 From vab at bb-c.de Mon Nov 7 11:05:34 2016 From: vab at bb-c.de (Volker A. Brandt) Date: Mon, 7 Nov 2016 12:05:34 +0100 Subject: [OmniOS-discuss] Can't bring up r151020 In-Reply-To: <4e4acde5-8751-88df-314b-558655a8a602@genashor.com> References: <4e4acde5-8751-88df-314b-558655a8a602@genashor.com> Message-ID: <22560.24574.384702.536840@shelob.bb-c.de> Hi Gary! > I'm also open to picking up a new machine like the HP Gen8. I'm using > this as a (very) small SOHO server to do email, web, and file services. > The current machine is a 1U unit which has a motherboard similar to the > SunFire v20, 4 disks: 2 mirrored 5TB(archive), 2 mirrored ~300GB drives > (rpool), and 8 GB of RAM which works well for my purpose. I can confirm that OmniOS works fine on the HP G8. The only gotcha is that if you want to use the HBA in non-RAID mode (which you do :-) AND you want to install an SSD in place of the DVD, then you will not be able to boot off said SSD. This is a stupid design limitation. I am working on a fix but am not finished yet. I expect to have something in early December. Basically, you need to install an SD card and chainload the "real" OS from there. Regards -- Volker -- ------------------------------------------------------------------------ Volker A. Brandt Consulting and Support for Oracle Solaris Brandt & Brandt Computer GmbH WWW: http://www.bb-c.de/ Am Wiesenpfad 6, 53340 Meckenheim, GERMANY Email: vab at bb-c.de Handelsregister: Amtsgericht Bonn, HRB 10513 Schuhgr??e: 46 Gesch?ftsf?hrer: Rainer J.H. Brandt und Volker A. Brandt "When logic and proportion have fallen sloppy dead" From jaakko.linnosaari at polarshift.fi Mon Nov 7 14:08:07 2016 From: jaakko.linnosaari at polarshift.fi (Jaakko Linnosaari) Date: Mon, 7 Nov 2016 16:08:07 +0200 Subject: [OmniOS-discuss] OmniOS r151020 installation via kayak fails with broken curl Message-ID: <0E0DBC26-806C-48D6-91C5-0C98E588C17F@polarshift.fi> Hi! I?ve upgraded couple of my home servers (with some challenges with SunSSH/OpenSSH transition) and now tried kayak with our test server environment. The installation fails when the script tries to fetch configuration from the kayak server. The correct configuration file is there, and installation succeeds with r151014. Trying curl manually in the maintenance shell results in linking error with libnghttp2... Has anyone had success installing servers with kayak yet? cheers, Jaakko From daleg at omniti.com Mon Nov 7 14:50:19 2016 From: daleg at omniti.com (Dale Ghent) Date: Mon, 7 Nov 2016 09:50:19 -0500 Subject: [OmniOS-discuss] OmniOS r151020 installation via kayak fails with broken curl In-Reply-To: <0E0DBC26-806C-48D6-91C5-0C98E588C17F@polarshift.fi> References: <0E0DBC26-806C-48D6-91C5-0C98E588C17F@polarshift.fi> Message-ID: > On Nov 7, 2016, at 9:08 AM, Jaakko Linnosaari wrote: > > Hi! > > I?ve upgraded couple of my home servers (with some challenges with SunSSH/OpenSSH transition) and now tried kayak with our test server environment. The installation fails when the script tries to fetch configuration from the kayak server. The correct configuration file is there, and installation succeeds with r151014. Trying curl manually in the maintenance shell results in linking error with libnghttp2... > > Has anyone had success installing servers with kayak yet? What is the source of your miniroot.gz ? https://omnios.omniti.com/media/miniroot-r151020.gz Should work. /dale From jimklimov at cos.ru Sun Nov 6 14:02:06 2016 From: jimklimov at cos.ru (Jim Klimov) Date: Sun, 06 Nov 2016 15:02:06 +0100 Subject: [OmniOS-discuss] OmniOS r151020 is now out! In-Reply-To: <79d26d32-e0a3-c473-7443-46a358c2a1f6@marzocchi.net> References: <79d26d32-e0a3-c473-7443-46a358c2a1f6@marzocchi.net> Message-ID: <33DA9CCE-89BE-4FE4-A8AF-8A6649499823@cos.ru> 6 ?????? 2016??. 1:02:39 CET, Olaf Marzocchi ?????: >On 04/11/2016 16:26, Dan McDonald wrote: >> With SunSSH being end-of-lifed, upgrading to r151020 may be difficult >for people still using SunSSH. There's a new upgrading page one can >get to off the Release Notes: >> >> https://omnios.omniti.com/wiki.php/Upgrade_to_r151020 > >Hi, >I started updating my server, therefore OpenSSH first. After that one I > >restarted the server. >I can login with my pass but two factor authentication doesn't work >anymore, I'm never asked for it. > >What I did: >I used the sshd_config.new as starting point with these changes: > >PasswordAuthentication no >UsePAM yes >#ChallengeResponseAuthentication yes (untouched, it's the default) > >I also tried > >AuthenticationMethods publickey keyboard-interactive > >but it didn't help. > >Then I went to pam.conf and did s/sshd-kbdint/sshd/ as written in the >guide. This is what I have now at the very end of said file: > ># SSH keyboard interactive, so that keys don't ask for verification >code >sshd auth requisite pam_authtok_get.so.1 >sshd auth required pam_dhkeys.so.1 >sshd auth required pam_google_authenticator.so nullok >sshd auth required pam_unix_cred.so.1 >sshd auth required pam_unix_auth.so.1 > >When I login via SSH, I'm not asked for the verification code. >I searched online for similar guides but it looks like I'm doing it >right so I don't know how to proceed. Anyone able to help? thanks > >Cheers, >Olaf >_______________________________________________ >OmniOS-discuss mailing list >OmniOS-discuss at lists.omniti.com >http://lists.omniti.com/mailman/listinfo/omnios-discuss As a tangential help in troubleshooting: you can run the client with 'ssh -v' (or several -v's) to boost debugging, and the server too perhaps (might run a server on nonstandard port and in foreground, so as to not fiddle with standard SMF manifest, at least initially). This can help uncover what methods it tries and fails and why. Jim -- Typos courtesy of K-9 Mail on my Samsung Android From jaakko.linnosaari at polarshift.fi Mon Nov 7 16:01:16 2016 From: jaakko.linnosaari at polarshift.fi (Jaakko Linnosaari) Date: Mon, 7 Nov 2016 18:01:16 +0200 Subject: [OmniOS-discuss] OmniOS r151020 installation via kayak fails with broken curl In-Reply-To: References: <0E0DBC26-806C-48D6-91C5-0C98E588C17F@polarshift.fi> Message-ID: <36DA3226-6320-4A4C-9BB2-0FFF07FD38EE@polarshift.fi> > On 7 Nov 2016, at 16.50, Dale Ghent wrote: >> On Nov 7, 2016, at 9:08 AM, Jaakko Linnosaari wrote: >> >> I?ve upgraded couple of my home servers (with some challenges with SunSSH/OpenSSH transition) and now tried kayak with our test server environment. The installation fails when the script tries to fetch configuration from the kayak server. The correct configuration file is there, and installation succeeds with r151014. Trying curl manually in the maintenance shell results in linking error with libnghttp2... >> >> Has anyone had success installing servers with kayak yet? > > What is the source of your miniroot.gz ? I think it?s from this package: pkg://omnios/system/install/kayak-kernel at 1.1-0.151020:20161102T022316Z > https://omnios.omniti.com/media/miniroot-r151020.gz > > Should work. Thanks. I?ll give it a try. -- Jaakko -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Mon Nov 7 16:27:01 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 7 Nov 2016 11:27:01 -0500 Subject: [OmniOS-discuss] OmniOS r151020 installation via kayak fails with broken curl In-Reply-To: <36DA3226-6320-4A4C-9BB2-0FFF07FD38EE@polarshift.fi> References: <0E0DBC26-806C-48D6-91C5-0C98E588C17F@polarshift.fi> <36DA3226-6320-4A4C-9BB2-0FFF07FD38EE@polarshift.fi> Message-ID: <6A09C98F-A000-4890-97EC-4765491956D7@omniti.com> > On Nov 7, 2016, at 11:01 AM, Jaakko Linnosaari wrote: > > I think it?s from this package: pkg://omnios/system/install/kayak-kernel at 1.1-0.151020:20161102T022316Z > Yeah, we need to update that package. We discovered this, fixed it in the download area, but didn't fix it on the IPS server. Sorry about that, Dan From jaakko.linnosaari at polarshift.fi Mon Nov 7 17:54:14 2016 From: jaakko.linnosaari at polarshift.fi (Jaakko Linnosaari) Date: Mon, 7 Nov 2016 19:54:14 +0200 Subject: [OmniOS-discuss] OmniOS r151020 installation via kayak fails with broken curl In-Reply-To: <6A09C98F-A000-4890-97EC-4765491956D7@omniti.com> References: <0E0DBC26-806C-48D6-91C5-0C98E588C17F@polarshift.fi> <36DA3226-6320-4A4C-9BB2-0FFF07FD38EE@polarshift.fi> <6A09C98F-A000-4890-97EC-4765491956D7@omniti.com> Message-ID: <888DC7D8-9864-475D-BC72-9D5400AA13E0@polarshift.fi> > On 7 Nov 2016, at 18.27, Dan McDonald wrote: >> On Nov 7, 2016, at 11:01 AM, Jaakko Linnosaari wrote: >> >> I think it?s from this package: pkg://omnios/system/install/kayak-kernel at 1.1-0.151020:20161102T022316Z > > Yeah, we need to update that package. We discovered this, fixed it in the download area, but didn't fix it on the IPS server. Ok, good to know. I got the install to work with the miniroot Dale linked to. ? Jaakko From robert.fantini at protonmail.com Mon Nov 7 21:57:33 2016 From: robert.fantini at protonmail.com (Robert.fantini) Date: Mon, 07 Nov 2016 16:57:33 -0500 Subject: [OmniOS-discuss] using proxmox, after upgrade KVM does not start Message-ID: Hello My issue may have nothing to do with the omnios upgrade. It could be due to a mistake I made somewhere else. After upgrading to : OmniOS 5.11 omnios-cac2b76 October 2016 SunOS sys4 5.11 omnios-cac2b76 i86pc i386 i86pc our KVM's using iscsi will not start. kvm: -drive file=iscsi://10.2.2.41/iqn.2010-09.org.napp-it:1459891666/2,if=none,id=drive-virtio0,format=raw,cache=none,aio=native,detect-zeroes=on: iSCSI: Failed to connect to LUN : iscsi_service failed with : iscsi_service_reconnect_if_loggedin. Can not reconnect right now. Has anyone seen that? Or suggest iscsi debugging/testing commands to get more information? thanks , Robert Fantini -------------- next part -------------- An HTML attachment was scrubbed... URL: From tobi at oetiker.ch Mon Nov 7 22:33:51 2016 From: tobi at oetiker.ch (Tobias Oetiker) Date: Mon, 7 Nov 2016 23:33:51 +0100 (CET) Subject: [OmniOS-discuss] using proxmox, after upgrade KVM does not start In-Reply-To: References: Message-ID: Robert, Today Robert.fantini wrote: > Hello > > My issue may have nothing to do with the omnios upgrade. It could be due to a mistake I made somewhere else. > > After upgrading to : > OmniOS 5.11 omnios-cac2b76 October 2016 > SunOS sys4 5.11 omnios-cac2b76 i86pc i386 i86pc > > our KVM's using iscsi will not start. > kvm: -drive file=iscsi://10.2.2.41/iqn.2010-09.org.napp-it:1459891666/2,if=none,id=drive-virtio0,format=raw,cache=none,aio=native,detect-zeroes=on: > > iSCSI: Failed to connect to LUN : iscsi_service failed with : iscsi_service_reconnect_if_loggedin. Can not reconnect right now. > > > Has anyone seen that? > > Or suggest iscsi debugging/testing commands to get more information? we found that after the upgrade a few of our luns had disaperaed ... since onone else reported this on the list we thought that we may have made a mistake ... but maybe there is something more just recreate the missing luns ... (we had to pick new lun numbers since omni complained that the old luns were already in use) then rescan iscsi on the proxmox host and all should be well. cheers tobi > > > thanks , Robert Fantini -- Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten, Switzerland www.oetiker.ch tobi at oetiker.ch +41 62 775 9902 From mir at miras.org Mon Nov 7 22:52:45 2016 From: mir at miras.org (Michael Rasmussen) Date: Mon, 7 Nov 2016 23:52:45 +0100 Subject: [OmniOS-discuss] using proxmox, after upgrade KVM does not start In-Reply-To: References: Message-ID: <20161107235245.760e3339@sleipner.datanom.net> On Mon, 7 Nov 2016 23:33:51 +0100 (CET) Tobias Oetiker wrote: > > we found that after the upgrade a few of our luns had disaperaed > ... since onone else reported this on the list we thought that we > may have made a mistake ... but maybe there is something more > Strange, had same experience but thought is was a human error. In my case is was not the Lun gone missing but a few views which was lost. > just recreate the missing luns ... (we had to pick new lun numbers > since omni complained that the old luns were already in use) > Did not have to do this; only recreate the views. > then rescan iscsi on the proxmox host and all should be well. > In proxmox case this is not needed since iscsi connections is made on demand using libiscsi. -- 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: I distinctly remember forgetting that. -Clara Barton -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 163 bytes Desc: OpenPGP digital signature URL: From bauernfeind at ipk-gatersleben.de Tue Nov 8 06:13:41 2016 From: bauernfeind at ipk-gatersleben.de (Jens Bauernfeind) Date: Tue, 8 Nov 2016 06:13:41 +0000 Subject: [OmniOS-discuss] updating ms-omniti repo fails Message-ID: Hello all, the problem discussed in this old thread http://lists.omniti.com/pipermail/omnios-discuss/2015-June/005193.html is still there. I have running a r151014 machine, the first clone of the ms-omniti repo is working fine, but updating is impossible. steps i have done: zfs create rpool/export/pkgrepo/omnios/ms.omniti.com pkgrepo create /export/pkgrepo/omnios/ms.omniti.com pkgrepo set -s /export/pkgrepo/omnios/ms.omniti.com publisher/prefix=ms.omniti.com pkgrecv -v -s http://pkg.omniti.com/omniti-ms/ -d /export/pkgrepo/omnios/ms.omniti.com/ --clone -p ms.omniti.com pkgrepo refresh -s /export/pkgrepo/omnios/ms.omniti.com updating: export PKG_SRC=http://pkg.omniti.com/omniti-ms/ export PKG_DEST=/export/pkgrepo/omnios/ms.omniti.com/ pkgrecv -v '*' --clone -p ms.omniti.com while updating, the following error message appears: PROCESS ITEMS GET (MB) SEND (MB) network/iftop 0/2016 0/9765 0/30694 pkgrecv: 'open' failed for transaction ID '1374267709_pkg%3A%2F%2Fms.omniti.com%2Fnetwork%2Fiftop%401.0.2%2C5.11-0.151 006%3A20130719T210149Z': The specified FMRI, 'pkg://ms.omniti.com/network/iftop at 1.0.2,5.11-0.151006:20130719T210149Z', already exists or has been restricted. Thanks for help Jens -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 6023 bytes Desc: not available URL: From pasztor at sagv5.gyakg.u-szeged.hu Tue Nov 8 20:03:30 2016 From: pasztor at sagv5.gyakg.u-szeged.hu (=?iso-8859-2?Q?P=C1SZTOR_Gy=F6rgy?=) Date: Tue, 8 Nov 2016 21:03:30 +0100 Subject: [OmniOS-discuss] Install onto usb stick Message-ID: <20161108200330.GB24038@linux.gyakg.u-szeged.hu> Hi, I barely remember, if it was a topic here, or on the openindiana-discuss list, if there are some best practice to install omnios/openindiana/illumos* onto a pendrive. My current home nas is installed onto a pendrive. But my previous pendrive died after 1 or 2 years, so I had to reinstall it lately. Since in the last couple of month my nas was turned off, due to other reasons, it just came into my mind, that I should ask that here, if sy has any good advice on that. As I said, it's working now, I just want to prevent the next pendrive damage. The "practice": the base system is on the pendrive, but the most of the services runs in zones. So when my first pendrive died, practically I just had to import my pool from the spinning disks, and create the zone configurations again, based on the prevoius exports. The zones are on the spinning disks. So, the config: rpool: pendrive tank: raidz from spinning disks, and I have a samsung 840 evo, what is partitioned into 2 half, the first partition is smaller, that is the zil for "tank", the second partition is around half of the ssd, and that is the l2arc for "tank". Any advice, to make it better / prevent pendrive / ssd failures? Thanks, Gyu From danmcd at omniti.com Wed Nov 9 20:03:32 2016 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 9 Nov 2016 15:03:32 -0500 Subject: [OmniOS-discuss] New OmniOS bloody Message-ID: <4B4EBBD9-3063-45AB-9AA2-0F590556018F@omniti.com> Say hello to OmniOS r151021, the next bloody cycle! This bloody update unfreezes illumos-omnios:master from upstream (and now sidestream LX) merges. Included with this new bloody update: - Rich Lowe's Address-Space Layout Randomization (ASLR). See here for details: https://www.listbox.com/member/archive/182179/2016/10/search/SEVBRFMgVVA/sort/time_rev/page/1/entry/1:2/20161015183813:0CAE0A88-9328-11E6-B5D5-B7004FCF1B41/ - BSD Loader integration, but NOT SUPPORTED YET. Updates to r151021 will include an /etc/default/be file with "BE_HAS_GRUB=true" in it. This allows us to move forward until we finalize what OmniOS + Loader will look like. - More post-020 ZFS bugfixes from upstream. - Some side-pulls of LX from illumos-joyent, catching us up. If you're an existing bloody user, the existing publisher now contains r151021 bits, and a full upgrade would happen. ISO, USB, and Kayak .zfs.bz2 images are available on the OmniOS Installation page: https://omnios.omniti.com/wiki.php/Installation Note that this bloody cycle will likely be longer, as r151022 is an LTS release and we have a LOT of changes now merged, with more on the way! Dan From mtalbott at lji.org Thu Nov 10 22:27:20 2016 From: mtalbott at lji.org (Michael Talbott) Date: Thu, 10 Nov 2016 14:27:20 -0800 Subject: [OmniOS-discuss] r151020 zone clone failure Message-ID: Hi, Since upgrading to the latest stable (from r151018 to r151020) it appears zone cloning is broken. This worked before the upgrade, but no such luck after the upgrade. I've tried uninstalling, removing the zone and all remnants, but, even after creating a fresh zone from scratch, configuring the new zone and then attempting the clone, this is failing as you can see below :( Any fixes or work arounds out there for this? # cat /etc/release OmniOS v11 r151020 Copyright 2016 OmniTI Computer Consulting, Inc. All rights reserved. Use is subject to license terms. # zonecfg -z sbp11 sbp11: No such zone configured Use 'create' to begin configuring a new zone. zonecfg:sbp11> create -b zonecfg:sbp11> set zonepath=/zones/sbp11 zonecfg:sbp11> set brand=ipkg zonecfg:sbp11> set autoboot=true zonecfg:sbp11> set ip-type=shared zonecfg:sbp11> add net zonecfg:sbp11:net> set address=10.0.5.11/16 zonecfg:sbp11:net> set physical=vlan1 zonecfg:sbp11:net> set defrouter=10.0.3.2 zonecfg:sbp11:net> end zonecfg:sbp11> commit zonecfg:sbp11> exit # zoneadm list -c global sbp11 # zoneadm -z sbp11 install A ZFS file system has been created for this zone. Image: Preparing at /zones/sbp11/root. Publisher: Using omnios (http://pkg.omniti.com/omnios/r151020/). Publisher: Using ms.omniti.com (http://pkg.omniti.com/omniti-ms/). Cache: Using /var/pkg/publisher. Installing: Packages (output follows) Packages to install: 92 Mediators to change: 1 Services to change: 4 DOWNLOAD PKGS FILES XFER (MB) SPEED Completed 92/92 26031/26031 166.2/166.2 804k/s PHASE ITEMS Installing new actions 39567/39567 Updating package state database Done Updating package cache 0/0 Updating image state Done Creating fast lookup database Done Note: Man pages can be obtained by installing pkg:/system/manual Postinstall: Copying SMF seed repository ... done. Done: Installation completed in 121.684 seconds. Next Steps: Boot the zone, then log into the zone console (zlogin -C) to complete the configuration process. # zonecfg -z sbp12 sbp12: No such zone configured Use 'create' to begin configuring a new zone. zonecfg:sbp12> create -b zonecfg:sbp12> set zonepath=/zones/sbp12 zonecfg:sbp12> set brand=ipkg zonecfg:sbp12> set autoboot=true zonecfg:sbp12> set ip-type=shared zonecfg:sbp12> add net zonecfg:sbp12:net> set address=10.0.5.12/16 zonecfg:sbp12:net> set physical=vlan1 zonecfg:sbp12:net> set defrouter=10.0.3.2 zonecfg:sbp12:net> end zonecfg:sbp12> commit zonecfg:sbp12> exit # zoneadm -z sbp12 clone sbp11 zone 'sbp12': "/usr/lib/fs/dev/mount -o attrdir=/zones/sbp12/root/dev /dev /zones/sbp12/lu/dev" failed with exit code 1 zoneadm: zone 'sbp12': call to zoneadmd failed ERROR: Zone mount failed ERROR: Unmounting zone zoneadm: zone 'sbp12': must be mounted before unmount. Thanks, Michael From danmcd at omniti.com Thu Nov 10 22:37:39 2016 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 10 Nov 2016 17:37:39 -0500 Subject: [OmniOS-discuss] r151020 zone clone failure In-Reply-To: References: Message-ID: <3723BE87-4D2E-4AD7-ADE3-C6DF2F4EDAD2@omniti.com> > On Nov 10, 2016, at 5:27 PM, Michael Talbott wrote: > > # zoneadm -z sbp12 clone sbp11 > zone 'sbp12': "/usr/lib/fs/dev/mount -o attrdir=/zones/sbp12/root/dev /dev /zones/sbp12/lu/dev" failed with exit code 1 > zoneadm: zone 'sbp12': call to zoneadmd failed > ERROR: Zone mount failed > ERROR: Unmounting zone > zoneadm: zone 'sbp12': must be mounted before unmount. > Damn. It is possible that the LX bits have broken ipkg/lipkg cloning. I would do the following myself now, but I likelywon't be able to until morning. If you're feeling adventurous: 1.) Edit /usr/lib/brand/ipkg/clone and change "#!/bin/ksh -p" to "#!/bin/ksh -px". 2.) Repeat your attempt to clone a zone. NOTE, you will have far more output this time. 3.) Share the more verbose output with me and/or the list. 4.) Reset /usr/lib/brand/ipkg/clone's first line to "#!/bin/ksh -p". I didn't change the ipkg/lipkg brand code, but maybe I need to, OR it's assuming zoneadm/zoneadmd behavior. And if you don't get additional output, that's also useful data. Thanks, Dan From danmcd at omniti.com Thu Nov 10 22:40:33 2016 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 10 Nov 2016 17:40:33 -0500 Subject: [OmniOS-discuss] r151020 zone clone failure In-Reply-To: References: Message-ID: > On Nov 10, 2016, at 5:27 PM, Michael Talbott wrote: > > zonecfg:sbp12> set ip-type=shared Oh, I didn't notice this was a shared-IP zone. The LX stuff (Joyent's code in general) has been very abusive toward shared-stack IP zones. I may have clobbered shared-stack code by accident as part of the LX port. (Not all of the LX port was confined to the brand/lx package.) Please do what I said earlier if you've the time, though. Dan From mtalbott at lji.org Thu Nov 10 22:47:29 2016 From: mtalbott at lji.org (Michael Talbott) Date: Thu, 10 Nov 2016 14:47:29 -0800 Subject: [OmniOS-discuss] r151020 zone clone failure In-Reply-To: References: Message-ID: <03DAD08B-CEE7-4CA3-9B9B-19ECCF81BCFC@lji.org> Here's a clone attempt after the change: # zoneadm -z sbp11 clone sbp10 + . /usr/lib/brand/ipkg/common.ksh + unset LD_LIBRARY_PATH + PATH=/usr/bin:/usr/sbin + export PATH + . /usr/lib/brand/shared/common.ksh + TEXTDOMAIN=SUNW_OST_OSCMD + export TEXTDOMAIN + gettext '%s: error: wrapper file already exists' + e_cannot_wrap='%s: error: wrapper file already exists' + gettext $'Invalid \'%s\' directory within the zone' + e_baddir=$'Invalid \'%s\' directory within the zone' + gettext $'Invalid \'%s\' file within the zone' + e_badfile=$'Invalid \'%s\' file within the zone' + gettext $'Pathname specified to -a \'%s\' must be absolute.' + e_path_abs=$'Pathname specified to -a \'%s\' must be absolute.' + gettext '%s: error: file or directory not found.' + e_not_found='%s: error: file or directory not found.' + gettext 'Installation aborted.' + e_install_abort='Installation aborted.' + gettext $'Cannot read directory \'%s\'' + e_not_readable=$'Cannot read directory \'%s\'' + gettext 'Error: must be a directory' + e_not_dir='Error: must be a directory' + gettext 'Error: Unknown archive format. Must be a flash archive, a cpio archive (can also be gzipped or bzipped), a pax XUSTAR archive, or a level 0 ufsdump archive.' + e_unknown_archive='Error: Unknown archive format. Must be a flash archive, a cpio archive (can also be gzipped or bzipped), a pax XUSTAR archive, or a level 0 ufsdump archive.' + gettext 'Error: archive contains absolute paths instead of relative paths.' + e_absolute_archive='Error: archive contains absolute paths instead of relative paths.' + gettext 'Error: the archive top-level directory (%s) does not match the zonepath (%s).' + e_mismatch_archive='Error: the archive top-level directory (%s) does not match the zonepath (%s).' + gettext 'Unable to create temporary file' + e_tmpfile='Unable to create temporary file' + gettext 'Zonepath root %s exists and contains data; remove or move aside prior to install.' + e_root_full='Zonepath root %s exists and contains data; remove or move aside prior to install.' + gettext 'Unable to create directory %s.' + f_mkdir='Unable to create directory %s.' + gettext 'Unable to chmod directory %s.' + f_chmod='Unable to chmod directory %s.' + gettext 'Unable to chown directory %s.' + f_chown='Unable to chown directory %s.' + gettext 'HWCAP: %s\n' + f_hwcap_info='HWCAP: %s' + gettext $'The image was created with an incompatible libc.so.1 hwcap lofs mount.\\n The zone will not boot on this platform. See the zone\'s\\n documentation for the recommended way to create the archive.' + f_sanity_hwcap=$'The image was created with an incompatible libc.so.1 hwcap lofs mount.\n The zone will not boot on this platform. See the zone\'s\n documentation for the recommended way to create the archive.' + gettext 'Analysing the archive' + m_analyse_archive='Analysing the archive' + gettext $'Cannot read file \'%s\'' + not_readable=$'Cannot read file \'%s\'' + gettext 'Input is not a flash archive' + not_flar='Input is not a flash archive' + gettext 'Flash archive is a corrupt' + bad_flar='Flash archive is a corrupt' + gettext 'Flash archive contains a ZFS send stream.\n\tRecreate the flar using the -L option with cpio or pax.' + bad_zfs_flar=$'Flash archive contains a ZFS send stream.\n\tRecreate the flar using the -L option with cpio or pax.' + gettext 'Unpacking the archive failed' + f_unpack_failed='Unpacking the archive failed' + gettext 'Archiver %s is not supported' + unknown_archiver='Archiver %s is not supported' + gettext $'Required command \'%s\' not executable!' + cmd_not_exec=$'Required command \'%s\' not executable!' + ZONE_SUBPROC_OK=0 + ZONE_SUBPROC_USAGE=253 + ZONE_SUBPROC_NOTCOMPLETE=254 + ZONE_SUBPROC_FATAL=255 + PROP_PARENT=org.opensolaris.libbe:parentbe + PROP_ACTIVE=org.opensolaris.libbe:active + gettext 'cannot specify both %s and %s options' + f_incompat_options='cannot specify both %s and %s options' + gettext 'Missing %s at %s' + f_sanity_detail='Missing %s at %s' + gettext 'Is this a sparse zone image? The image must be whole-root.' + f_sanity_sparse='Is this a sparse zone image? The image must be whole-root.' + gettext ' Sanity Check: Passed. Looks like an OpenSolaris system.' + sanity_ok=' Sanity Check: Passed. Looks like an OpenSolaris system.' + gettext ' Sanity Check: FAILED (see log for details).' + sanity_fail=' Sanity Check: FAILED (see log for details).' + gettext ' Sanity Check: the Solaris image (release %s) is not an OpenSolaris image and cannot be installed in this type of branded zone.' + sanity_fail_vers=' Sanity Check: the Solaris image (release %s) is not an OpenSolaris image and cannot be installed in this type of branded zone.' + gettext ' Result: *** Installation FAILED ***' + install_fail=' Result: *** Installation FAILED ***' + gettext $'Installing a zone inside of the root pool\'s \'ROOT\' dataset is unsupported.' + f_zfs_in_root=$'Installing a zone inside of the root pool\'s \'ROOT\' dataset is unsupported.' + gettext $'Unable to create the zone\'s ZFS dataset.' + f_zfs_create=$'Unable to create the zone\'s ZFS dataset.' + gettext $'Unable to create the zone\'s ZFS dataset mountpoint.' + f_root_create=$'Unable to create the zone\'s ZFS dataset mountpoint.' + gettext 'unable to determine global zone boot environment.' + f_no_gzbe='unable to determine global zone boot environment.' + gettext 'the zonepath must be a ZFS dataset.\nThe parent directory of the zonepath must be a ZFS dataset so that the\nzonepath ZFS dataset can be created properly.' + f_no_ds=$'the zonepath must be a ZFS dataset.\nThe parent directory of the zonepath must be a ZFS dataset so that the\nzonepath ZFS dataset can be created properly.' + gettext 'multiple active datasets.' + f_multiple_ds='multiple active datasets.' + gettext 'no active dataset.' + f_no_active_ds='no active dataset.' + gettext $'Unable to unmount the zone\'s root ZFS dataset (%s).\\nIs there a global zone process inside the zone root?\\nThe current zone boot environment will remain mounted.\\n' + f_zfs_unmount=$'Unable to unmount the zone\'s root ZFS dataset (%s).\nIs there a global zone process inside the zone root?\nThe current zone boot environment will remain mounted.' + gettext $'Unable to mount the zone\'s ZFS dataset.' + f_zfs_mount=$'Unable to mount the zone\'s ZFS dataset.' + gettext 'Expected %s to be a directory.' + f_safedir='Expected %s to be a directory.' + gettext 'Failed to cp %s %s.' + f_cp='Failed to cp %s %s.' + gettext 'Failed to safely copy %s to %s.' + f_cp_unsafe='Failed to safely copy %s to %s.' + gettext 'brand-specific usage: ' + m_brnd_usage='brand-specific usage: ' + gettext 'Performing zone unconfiguration' + v_unconfig='Performing zone unconfiguration' + gettext 'Zone unconfiguration failed' + e_unconfig='Zone unconfiguration failed' + gettext 'Mounting the zone' + v_mounting='Mounting the zone' + gettext 'Zone mount failed' + e_badmount='Zone mount failed' + gettext 'Unmounting zone' + v_unmount='Unmounting zone' + gettext 'Zone unmount failed' + e_badunmount='Zone unmount failed' + gettext 'Postprocessing failed.' + e_exitfail='Postprocessing failed.' + gettext ' Done: Installation completed in %s seconds.' + m_complete=' Done: Installation completed in %s seconds.' + gettext ' Next Steps: Boot the zone, then log into the zone console (zlogin -C)' + m_postnote=' Next Steps: Boot the zone, then log into the zone console (zlogin -C)' + gettext ' to complete the configuration process.' + m_postnote2=' to complete the configuration process.' + gettext 'clone {sourcezone}' + m_usage='clone {sourcezone}' + gettext 'Error: unable to determine source zone dataset.' + f_nosource='Error: unable to determine source zone dataset.' + gettext 'Error: specified snapshot is invalid for this source zone.' + f_badsource='Error: specified snapshot is invalid for this source zone.' + ROOT=rpool/ROOT + getopts R:s:Xz: opt + ZONENAME=sbp11 + getopts R:s:Xz: opt + ZONEPATH=/zones/sbp11 + getopts R:s:Xz: opt + shift 4 + [ 1 -ne 1 ] + sourcezone=sbp10 + get_current_gzbe + [ -x /usr/sbin/beadm ] + /usr/bin/nawk -v alt= '-F;' $'{\n\t\t\tif (length(alt) == 0) {\n\t\t\t # Field 3 is the BE status. N is the active BE.\n\t\t\t if ($3 !~ "N")\n\t\t\t\tnext\n\t\t\t} else {\n\t\t\t # Field 4 is the BE mountpoint.\n\t\t\t if ($4 != alt)\n\t\t\t\tnext\n\t\t\t}\n\t\t\t# Field 2 is the BE UUID\n\t\t\tprint $2\n\t\t}' + /usr/sbin/beadm list -H + CURRENT_GZBE=0e7a1eea-12fd-4ec5-8daa-b70a31aa59df + [ -z 0e7a1eea-12fd-4ec5-8daa-b70a31aa59df ] + [ -z '' ] + awk -F: '{print $4}' + /usr/sbin/zoneadm -z sbp10 list -p + sourcezonepath=/zones/sbp10 + [ -z /zones/sbp10 ] + get_zonepath_ds /zones/sbp10 + /usr/bin/nawk -v zonepath=/zones/sbp10 $'{\n\t\tif ($2 == zonepath)\n\t\t\tprint $1\n\t}' + /usr/sbin/zfs list -H -t filesystem -o name,mountpoint + ZONEPATH_DS=rpool/zones/sbp10 + [ -z rpool/zones/sbp10 ] + get_active_ds 0e7a1eea-12fd-4ec5-8daa-b70a31aa59df rpool/zones/sbp10 + /usr/bin/nawk -v gzbe=0e7a1eea-12fd-4ec5-8daa-b70a31aa59df $' {\n\t\tif ($1 ~ /ROOT\\/[^\\/]+$/ && $2 == gzbe && $3 == "on") {\n\t\t\tprint $1\n\t\t\tif (found == 1)\n\t\t\t\texit 1\n\t\t\tfound = 1\n\t\t}\n\t }' + /usr/sbin/zfs list -H -r -t filesystem -o name,org.opensolaris.libbe:parentbe,org.opensolaris.libbe:active rpool/zones/sbp10/ROOT + ACTIVE_DS=rpool/zones/sbp10/ROOT/zbe-2 + [ 0 -ne 0 ] + [ -z rpool/zones/sbp10/ROOT/zbe-2 ] + [ x != x ] + /usr/bin/dirname /zones/sbp11 + pdir=/zones + /usr/bin/basename /zones/sbp11 + zpname=sbp11 + get_zonepath_ds /zones + /usr/bin/nawk -v zonepath=/zones $'{\n\t\tif ($2 == zonepath)\n\t\t\tprint $1\n\t}' + /usr/sbin/zfs list -H -t filesystem -o name,mountpoint + ZONEPATH_DS=rpool/zones + [ -z rpool/zones ] + zpds=rpool/zones + fail_zonepath_in_rootds rpool/zones + break + /usr/sbin/zfs create rpool/zones/sbp11 + /usr/sbin/zfs create -o mountpoint=legacy -o zoned=on rpool/zones/sbp11/ROOT + [ x = x ] + SNAPNAME=sbp11_snap + SNAPNUM=0 + [ 0 -lt 100 ] + /usr/sbin/zfs snapshot rpool/zones/sbp10/ROOT/zbe-2 at sbp11_snap + [ 0 = 0 ] + break + [ 0 -ge 100 ] + BENAME=zbe + BENUM=0 + [ 0 -lt 100 ] + /usr/sbin/zfs clone rpool/zones/sbp10/ROOT/zbe-2 at sbp11_snap rpool/zones/sbp11/ROOT/zbe + [ 0 = 0 ] + break + [ 0 -ge 100 ] + /usr/sbin/zfs set org.opensolaris.libbe:active=on rpool/zones/sbp11/ROOT/zbe + /usr/sbin/zfs set org.opensolaris.libbe:parentbe=0e7a1eea-12fd-4ec5-8daa-b70a31aa59df rpool/zones/sbp11/ROOT/zbe + /usr/sbin/zfs set canmount=noauto rpool/zones/sbp11/ROOT/zbe + [ ! -d /zones/sbp11/root ] + /usr/bin/mkdir -p /zones/sbp11/root + /usr/bin/chmod 700 /zones/sbp11 + ZONE_IS_MOUNTED=0 + trap trap_exit EXIT + is_brand_labeled + [ -z ] + AR_OPTIONS='' + awk -F: '{print $6}' + /usr/sbin/zoneadm -z sbp11 list -p + brand=ipkg + [[ ipkg == labeled ]] + return 0 + (( 0 == 0 )) + [ x = xyes ] + unconfigure_zone + vlog 'Performing zone unconfiguration' + fmt='Performing zone unconfiguration' + typeset fmt + shift + [[ -n '' ]] + [[ -n '' ]] + vlog 'Mounting the zone' + fmt='Mounting the zone' + typeset fmt + shift + [[ -n '' ]] + [[ -n '' ]] + ZONE_IS_MOUNTED=1 + zoneadm -z sbp11 mount -f zone 'sbp11': "/usr/lib/fs/dev/mount -o attrdir=/zones/sbp11/root/dev /dev /zones/sbp11/lu/dev" failed with exit code 1 zoneadm: zone 'sbp11': call to zoneadmd failed + fatal 'Zone mount failed' + fmt='Zone mount failed' + typeset fmt + shift + error 'Zone mount failed' + fmt='Zone mount failed' + typeset fmt + shift + printf 'ERROR: Zone mount failed\n' ERROR: Zone mount failed + [[ -n '' ]] + exit + trap_exit + (( 1 != 0 )) + error 'Unmounting zone' + fmt='Unmounting zone' + typeset fmt + shift + printf 'ERROR: Unmounting zone\n' ERROR: Unmounting zone + [[ -n '' ]] + zoneadm -z sbp11 unmount zoneadm: zone 'sbp11': must be mounted before unmount. + exit > On Nov 10, 2016, at 2:40 PM, Dan McDonald wrote: > >> On Nov 10, 2016, at 5:27 PM, Michael Talbott wrote: >> >> zonecfg:sbp12> set ip-type=shared > > > Oh, I didn't notice this was a shared-IP zone. The LX stuff (Joyent's code in general) has been very abusive toward shared-stack IP zones. I may have clobbered shared-stack code by accident as part of the LX port. (Not all of the LX port was confined to the brand/lx package.) > > Please do what I said earlier if you've the time, though. > > Dan > From danmcd at omniti.com Thu Nov 10 22:54:06 2016 From: danmcd at omniti.com (Dan McDonald) Date: Thu, 10 Nov 2016 17:54:06 -0500 Subject: [OmniOS-discuss] r151020 zone clone failure In-Reply-To: <03DAD08B-CEE7-4CA3-9B9B-19ECCF81BCFC@lji.org> References: <03DAD08B-CEE7-4CA3-9B9B-19ECCF81BCFC@lji.org> Message-ID: <73EA4CCE-0A04-4D61-98D0-61AB5A798F25@omniti.com> > On Nov 10, 2016, at 5:47 PM, Michael Talbott wrote: > > Here's a clone attempt after the change: Thank you. This is a lot of useful information. I will dive into this first thing tomorrow. Thanks, Dan From danmcd at omniti.com Fri Nov 11 20:12:57 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 11 Nov 2016 15:12:57 -0500 Subject: [OmniOS-discuss] r151020 zone clone failure In-Reply-To: <73EA4CCE-0A04-4D61-98D0-61AB5A798F25@omniti.com> References: <03DAD08B-CEE7-4CA3-9B9B-19ECCF81BCFC@lji.org> <73EA4CCE-0A04-4D61-98D0-61AB5A798F25@omniti.com> Message-ID: <4B3E128C-08D5-4E38-92FC-7C5CF7A25077@omniti.com> > On Nov 10, 2016, at 5:54 PM, Dan McDonald wrote: > > Thank you. This is a lot of useful information. I will dive into this first thing tomorrow. I may have found the problem, but I need to test it to make sure that by fixing clone I didn't break something else. If this fix is correct, your bug occurred because I merged something from Joyent w/o fully understanding its consequences to ipkg/lipkg zones. If you're feeling brave, you're welcome to try the first-cut replacement zoneadmd binary I have. If you're not, I'll be testing: ipkg: zoneadm {install,ready,boot,halt,boot-from-halt,clone} lipkg: zoneadm {install,ready,boot,halt,boot-from-halt,clone} lx: zoneadm {install,ready,boot,halt,boot-from-halt, (optional clone)} (BTW, I have "optional clone" for LX because the install method for OmniOS LX has something already like clone.) Thanks for finding this bug, and sorry for breaking ipkg/lipkg cloning, Dan From mtalbott at lji.org Fri Nov 11 20:22:55 2016 From: mtalbott at lji.org (Michael Talbott) Date: Fri, 11 Nov 2016 12:22:55 -0800 Subject: [OmniOS-discuss] r151020 zone clone failure In-Reply-To: <4B3E128C-08D5-4E38-92FC-7C5CF7A25077@omniti.com> References: <03DAD08B-CEE7-4CA3-9B9B-19ECCF81BCFC@lji.org> <73EA4CCE-0A04-4D61-98D0-61AB5A798F25@omniti.com> <4B3E128C-08D5-4E38-92FC-7C5CF7A25077@omniti.com> Message-ID: I'm pretty brave ;) Just let me know where I can snag that binary and where to place it and I'll give it a go. Thanks, Michael > On Nov 11, 2016, at 12:12 PM, Dan McDonald wrote: > > >> On Nov 10, 2016, at 5:54 PM, Dan McDonald wrote: >> >> Thank you. This is a lot of useful information. I will dive into this first thing tomorrow. > > I may have found the problem, but I need to test it to make sure that by fixing clone I didn't break something else. > > If this fix is correct, your bug occurred because I merged something from Joyent w/o fully understanding its consequences to ipkg/lipkg zones. > > If you're feeling brave, you're welcome to try the first-cut replacement zoneadmd binary I have. If you're not, I'll be testing: > > ipkg: zoneadm {install,ready,boot,halt,boot-from-halt,clone} > > lipkg: zoneadm {install,ready,boot,halt,boot-from-halt,clone} > > lx: zoneadm {install,ready,boot,halt,boot-from-halt, (optional clone)} > > (BTW, I have "optional clone" for LX because the install method for OmniOS LX has something already like clone.) > > > Thanks for finding this bug, and sorry for breaking ipkg/lipkg cloning, > Dan > From danmcd at omniti.com Fri Nov 11 20:30:28 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 11 Nov 2016 15:30:28 -0500 Subject: [OmniOS-discuss] r151020 zone clone failure In-Reply-To: References: <03DAD08B-CEE7-4CA3-9B9B-19ECCF81BCFC@lji.org> <73EA4CCE-0A04-4D61-98D0-61AB5A798F25@omniti.com> <4B3E128C-08D5-4E38-92FC-7C5CF7A25077@omniti.com> Message-ID: <063A4E4A-CBBE-4666-A895-5F7B5BC72199@omniti.com> > On Nov 11, 2016, at 3:22 PM, Michael Talbott wrote: > > I'm pretty brave ;) Just let me know where I can snag that binary and where to place it and I'll give it a go. Attached. Replace /usr/lib/zones/zoneadmd with the following binary. Testing so far is acceptable. Dan -------------- next part -------------- A non-text attachment was scrubbed... Name: zoneadmd Type: application/octet-stream Size: 147128 bytes Desc: not available URL: From mtalbott at lji.org Fri Nov 11 20:45:54 2016 From: mtalbott at lji.org (Michael Talbott) Date: Fri, 11 Nov 2016 12:45:54 -0800 Subject: [OmniOS-discuss] r151020 zone clone failure In-Reply-To: <063A4E4A-CBBE-4666-A895-5F7B5BC72199@omniti.com> References: <03DAD08B-CEE7-4CA3-9B9B-19ECCF81BCFC@lji.org> <73EA4CCE-0A04-4D61-98D0-61AB5A798F25@omniti.com> <4B3E128C-08D5-4E38-92FC-7C5CF7A25077@omniti.com> <063A4E4A-CBBE-4666-A895-5F7B5BC72199@omniti.com> Message-ID: That seems to have done the trick. Many thanks. Michael > On Nov 11, 2016, at 12:30 PM, Dan McDonald wrote: > > >> On Nov 11, 2016, at 3:22 PM, Michael Talbott wrote: >> >> I'm pretty brave ;) Just let me know where I can snag that binary and where to place it and I'll give it a go. > > Attached. Replace /usr/lib/zones/zoneadmd with the following binary. Testing so far is acceptable. > > Dan > > From danmcd at omniti.com Fri Nov 11 21:06:15 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 11 Nov 2016 16:06:15 -0500 Subject: [OmniOS-discuss] r151020 zone clone failure In-Reply-To: References: <03DAD08B-CEE7-4CA3-9B9B-19ECCF81BCFC@lji.org> <73EA4CCE-0A04-4D61-98D0-61AB5A798F25@omniti.com> <4B3E128C-08D5-4E38-92FC-7C5CF7A25077@omniti.com> <063A4E4A-CBBE-4666-A895-5F7B5BC72199@omniti.com> Message-ID: <32DD8C51-EC24-4488-8081-17D3C8E8CA85@omniti.com> > On Nov 11, 2016, at 3:45 PM, Michael Talbott wrote: > > That seems to have done the trick. Many thanks. Oh cool! Below is the patch. I merged something in from Joyent without considering consequences. I'm having them look at the fix to make sure I don't shoot LX in the foot, but my test rig's LX zones are behaving normally. And worse, the Joyent change in question was brought up in the illumos developer list as I was deep in pushing 020 out the door, so I didn't see it. When this fix hits the repos, it'll require only one package update, but it's a reboot-your-box update. Sorry for this again, Dan diff --git a/usr/src/cmd/zoneadmd/zoneadmd.c b/usr/src/cmd/zoneadmd/zoneadmd.c index 0a728000..1fb8275 100644 --- a/usr/src/cmd/zoneadmd/zoneadmd.c +++ b/usr/src/cmd/zoneadmd/zoneadmd.c @@ -554,9 +554,16 @@ static int zone_ready(zlog_t *zlogp, zone_mnt_t mount_cmd, int zstate) { int err; + boolean_t do_prestate; - if (!ALT_MOUNT(mount_cmd) && - brand_prestatechg(zlogp, zstate, Z_READY) != 0) + /* + * LX zones don't like prestate/poststate being done on non-booting + * mount commands. Other zones may, so be picky about when/when-not + * to invoke pre/post state change. + */ + do_prestate = (strcmp(brand_name, "lx") != 0 || !ALT_MOUNT(mount_cmd)); + + if (do_prestate && brand_prestatechg(zlogp, zstate, Z_READY) != 0) return (-1); if ((err = zonecfg_create_snapshot(zone_name)) != Z_OK) { @@ -580,8 +587,7 @@ zone_ready(zlog_t *zlogp, zone_mnt_t mount_cmd, int zstate) goto bad; } - if (!ALT_MOUNT(mount_cmd) && - brand_poststatechg(zlogp, zstate, Z_READY) != 0) + if (do_prestate && brand_poststatechg(zlogp, zstate, Z_READY) != 0) goto bad; return (0); From jdg117 at elvis.arl.psu.edu Fri Nov 11 21:53:30 2016 From: jdg117 at elvis.arl.psu.edu (John D Groenveld) Date: Fri, 11 Nov 2016 16:53:30 -0500 Subject: [OmniOS-discuss] r151020 zone update fail Message-ID: <201611112153.uABLrUkE025724@elvis.arl.psu.edu> I'm not sure how I put foo into this state, all of my other zones upgraded fine. John groenveld at acm.org # zoneadm -z foo attach -u Log File: /var/tmp/foo.attach_log.QIaG0l Attach Path: /var/opt/zones/foo/root Attach ZFS Dataset: tank/zones/foo/ROOT/zbe-5 Installing: Using pre-existing data in zonepath Global zone version: entire at 11-0.151020:20161102T012108Z Non-Global zone version: entire at 11-0.151018:20161027T172952Z Cache: Using /var/pkg/publisher. Updating non-global zone: Output follows Creating Plan (Running solver): /ERROR: Could not update attaching zone Result: Attach Failed. # cat /var/tmp/foo.attach_log.QIaG0l [Fri Nov 11 16:02:25 EST 2016] Log File: /var/tmp/foo.attach_log.QIaG0l [Fri Nov 11 16:02:27 EST 2016] Attach Path: /var/opt/zones/foo/root [Fri Nov 11 16:02:27 EST 2016] Attach ZFS Dataset: tank/zones/foo/ROOT/zbe-5 [Fri Nov 11 16:02:27 EST 2016] existing [Fri Nov 11 16:02:27 EST 2016] Installing: Using pre-existing data in zonepath [Fri Nov 11 16:02:27 EST 2016] Sanity Check: Passed. Looks like an OpenSolaris system. [Fri Nov 11 16:02:32 EST 2016] Global zone version: entire at 11-0.151020:20161102T012108Z [Fri Nov 11 16:02:32 EST 2016] Non-Global zone version: entire at 11-0.151018:20161027T172952Z [Fri Nov 11 16:02:32 EST 2016] Cache: Using /var/pkg/publisher. [Fri Nov 11 16:02:32 EST 2016] Updating non-global zone: Output follows pkg install: No solution was found to satisfy constraints maintained incorporations: None Plan Creation: dependency error(s) in proposed packages: No suitable version of required package pkg://omnios/entire at 11,5.11-0.151020:20161102T012108Z found: Reject: pkg://omnios/entire at 11,5.11-0.151020:20161102T012108Z Reason: All versions matching 'require' dependency pkg:/package/pkg at 0.5.11,5.11-0.151020 are rejected Reject: pkg://omnios/package/pkg at 0.5.11,5.11-0.151020:20161101T225845Z Reason: Global zone has an older version of package: pkg://omnios/package/pkg at 0.5.11,5.11-0.151018:20160409T070237Z Plan Creation: Errors in installed packages due to proposed changes: No suitable version of installed package pkg://omnios/entire at 11,5.11-0.151018:20161027T172952Z found Reject: pkg://omnios/entire at 11,5.11-0.151018:20161027T172952Z Reason: This version excluded by specified installation version Reject: pkg://omnios/entire at 11,5.11-0.151020:20161102T012108Z Reason: All versions matching 'require' dependency pkg:/package/pkg at 0.5.11,5.11-0.151020 are rejected Reject: pkg://omnios/package/pkg at 0.5.11,5.11-0.151020:20161101T225845Z Reason: Global zone has an older version of package: pkg://omnios/package/pkg at 0.5.11,5.11-0.151018:20160409T070237Z [Fri Nov 11 16:02:40 EST 2016] ERROR: Could not update attaching zone [Fri Nov 11 16:02:41 EST 2016] Result: Attach Failed. # pkg publisher PUBLISHER TYPE STATUS P LOCATION omnios origin online F http://pkg.omniti.com/omnios/r151020/ # pkg -R /var/opt/zones/foo/root/ publisher PUBLISHER TYPE STATUS P LOCATION omnios origin online F http://pkg.omniti.com/omnios/r151020/ From danmcd at omniti.com Fri Nov 11 21:58:07 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 11 Nov 2016 16:58:07 -0500 Subject: [OmniOS-discuss] r151020 zone update fail In-Reply-To: <201611112153.uABLrUkE025724@elvis.arl.psu.edu> References: <201611112153.uABLrUkE025724@elvis.arl.psu.edu> Message-ID: <91B85533-9AAF-4726-8BD2-8508C640579A@omniti.com> Do you have ms.omniti.com packages in that zone? Or does that zone still have SunSSH running in it? Dan From danmcd at omniti.com Fri Nov 11 22:09:06 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 11 Nov 2016 17:09:06 -0500 Subject: [OmniOS-discuss] r151020 zone update fail In-Reply-To: <91B85533-9AAF-4726-8BD2-8508C640579A@omniti.com> References: <201611112153.uABLrUkE025724@elvis.arl.psu.edu> <91B85533-9AAF-4726-8BD2-8508C640579A@omniti.com> Message-ID: <9B47559D-3120-48AC-9BB6-E4892ABEDE83@omniti.com> > On Nov 11, 2016, at 4:58 PM, Dan McDonald wrote: > > Do you have ms.omniti.com packages in that zone? Or does that zone still have SunSSH running in it? You can use the "OOPS" section from Upgrade_to_r151020 to help: zoneadm -z foo attach -F pkg -R /zones/foo/root For various values of until you can get it in line with global. Then: zoneadm -z foo detach zoneadm -z foo attach -u and if you've done all the massaging you need with pkg -R above, this should just work. Using pkg -R gives you more fine-grained control and diagnostics. Dan From jdg117 at elvis.arl.psu.edu Fri Nov 11 22:35:41 2016 From: jdg117 at elvis.arl.psu.edu (John D Groenveld) Date: Fri, 11 Nov 2016 17:35:41 -0500 Subject: [OmniOS-discuss] r151020 zone update fail In-Reply-To: Your message of "Fri, 11 Nov 2016 16:58:07 EST." <91B85533-9AAF-4726-8BD2-8508C640579A@omniti.com> References: <201611112153.uABLrUkE025724@elvis.arl.psu.edu> <91B85533-9AAF-4726-8BD2-8508C640579A@omniti.com> Message-ID: <201611112235.uABMZfea025881@elvis.arl.psu.edu> In message <91B85533-9AAF-4726-8BD2-8508C640579A at omniti.com>, Dan McDonald writ es: >Do you have ms.omniti.com packages in that zone? Or does that zone still have > SunSSH running in it? *blush* Just stock publisher, but SunSSH was still installed in that zone due a bug in my scripting. I fixed that via your documented escape hatch: However, I haven't made it thru the door: # pkg -R /var/opt/zones/foo/root/ publisher PUBLISHER TYPE STATUS P LOCATION omnios origin online F http://pkg.omniti.com/omnios/r151020/ # pkg -R /var/opt/zones/foo/root/ update Creating Plan (Running solver): \ pkg update: No solution was found to satisfy constraints Plan Creation: Package solver has not found a solution to update to latest available versions. This may indicate an overly constrained set of packages are installed. latest incorporations: pkg://omnios/incorporation/jeos/illumos-gate at 11,5.11-0.151020:20161101T221914Z pkg://omnios/incorporation/jeos/omnios-userland at 11,5.11-0.151020:20161102T011117Z pkg://omnios/entire at 11,5.11-0.151020:20161102T012108Z pkg://omnios/consolidation/osnet/osnet-incorporation at 0.5.11,5.11-0.151020:20161101T222452Z The following indicates why the system cannot update to the latest version: No suitable version of required package pkg://omnios/entire at 11,5.11-0.151020:20161102T012108Z found: Reject: pkg://omnios/entire at 11,5.11-0.151020:20161102T012108Z Reason: A version for 'require' dependency on pkg:/package/pkg at 0.5.11,5.11-0.151020 cannot be found John groenveld at acm.org From danmcd at omniti.com Fri Nov 11 23:36:51 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 11 Nov 2016 18:36:51 -0500 Subject: [OmniOS-discuss] Fwd: r151020 zone update fail References: <7C755FC9-1315-49F4-86CB-6233F75167A3@omniti.com> Message-ID: Sent from my iPhone (typos, autocorrect, and all) Begin forwarded message: > From: Dan McDonald > Date: November 11, 2016 at 6:36:27 PM EST > To: John D Groenveld > Subject: Re: [OmniOS-discuss] r151020 zone update fail > > Try update with the --reject of the sunssh packages. Also documented on the upgrade page are those. > > Dan > > Sent from my iPhone (typos, autocorrect, and all) > >> On Nov 11, 2016, at 5:35 PM, John D Groenveld wrote: >> >> In message <91B85533-9AAF-4726-8BD2-8508C640579A at omniti.com>, Dan McDonald writ >> es: >>> Do you have ms.omniti.com packages in that zone? Or does that zone still have >>> SunSSH running in it? >> >> *blush* >> Just stock publisher, but SunSSH was still installed in that zone >> due a bug in my scripting. >> I fixed that via your documented escape hatch: >> >> >> However, I haven't made it thru the door: >> # pkg -R /var/opt/zones/foo/root/ publisher >> PUBLISHER TYPE STATUS P LOCATION >> omnios origin online F http://pkg.omniti.com/omnios/r151020/ >> >> # pkg -R /var/opt/zones/foo/root/ update >> Creating Plan (Running solver): \ >> pkg update: No solution was found to satisfy constraints >> Plan Creation: Package solver has not found a solution to update to latest available versions. >> This may indicate an overly constrained set of packages are installed. >> >> latest incorporations: >> >> pkg://omnios/incorporation/jeos/illumos-gate at 11,5.11-0.151020:20161101T221914Z >> pkg://omnios/incorporation/jeos/omnios-userland at 11,5.11-0.151020:20161102T011117Z >> pkg://omnios/entire at 11,5.11-0.151020:20161102T012108Z >> pkg://omnios/consolidation/osnet/osnet-incorporation at 0.5.11,5.11-0.151020:20161101T222452Z >> >> The following indicates why the system cannot update to the latest version: >> >> No suitable version of required package pkg://omnios/entire at 11,5.11-0.151020:20161102T012108Z found: >> Reject: pkg://omnios/entire at 11,5.11-0.151020:20161102T012108Z >> Reason: A version for 'require' dependency on pkg:/package/pkg at 0.5.11,5.11-0.151020 cannot be found >> >> John >> groenveld at acm.org >> _______________________________________________ >> 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 jboren at drakecooper.com Sat Nov 12 00:44:26 2016 From: jboren at drakecooper.com (Joseph Boren) Date: Fri, 11 Nov 2016 17:44:26 -0700 Subject: [OmniOS-discuss] struggling to get samba 4 installed and working on omnios-r151020-b5b8c75 Message-ID: Greetings, I've been trying to install samba 4 on a test box and failing hard. I need the vfs_fruit module cause I've got nothing but mac clients and SMB over a VPN is unusably slow without it. I started with a brand new install of r151020. Initially I tried the recipe here: https://www.mail-archive.com/omnios-discuss at lists.omniti.com/msg06318.html However I couldn't even get nettle installed via those instructions, so I gave up, installed the Joyent Repos and installed Samba 4 from there. Now I'm struggling to figure out how to get the default SMB stuff disabled and the samba4 enabled. If I disable: svcadm disable network/smb/client:default svcadm disable network/smb/server:default They're right back after a reboot, won't stay disabled, as if I'm using the -t option, but i'm not. I can't figure out how to make it stay disable, and i can't figure out how to enable and start samba. I'm not even sure what to try next. I realized I don't even know if those are the native SMB services or new samba one, or how to figure that out. testparm -smb.conf gets me: rlimit_max: increasing rlimit_max (256) to minimum Windows limit (16384) Error loading services. Has anybody successfully installed samba4 on r151020? Does anyone have any guidance for disabling the native SMB services and enabling/configuring samba 4? I'm pretty much thoroughly stuck now. Thanks and best regards, Joseph Boren IT Specialist +c: 208.891.2128 + 416 S 8th St., Boise, ID 83702 + drakecooper.com *Read our minds *Blog Facebook , Twitter , Linkedin , Instagram -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdg117 at elvis.arl.psu.edu Sat Nov 12 15:29:03 2016 From: jdg117 at elvis.arl.psu.edu (John D Groenveld) Date: Sat, 12 Nov 2016 10:29:03 -0500 Subject: [OmniOS-discuss] r151020 zone update fail In-Reply-To: Your message of "Fri, 11 Nov 2016 17:35:41 EST." <201611112235.uABMZfea025881@elvis.arl.psu.edu> References: <201611112153.uABLrUkE025724@elvis.arl.psu.edu> <91B85533-9AAF-4726-8BD2-8508C640579A@omniti.com> <201611112235.uABMZfea025881@elvis.arl.psu.edu> Message-ID: <201611121529.uACFT3nT029114@elvis.arl.psu.edu> In message <201611112235.uABMZfea025881 at elvis.arl.psu.edu>, John D Groenveld writes: >Just stock publisher, but SunSSH was still installed in that zone >due a bug in my scripting. >I fixed that via your documented escape hatch: > > >However, I haven't made it thru the door: ># pkg -R /var/opt/zones/foo/root/ publisher >PUBLISHER TYPE STATUS P LOCATION >omnios origin online F http://pkg.omniti.com/omnios/r151020/ > ># pkg -R /var/opt/zones/foo/root/ update >Creating Plan (Running solver): \ >pkg update: No solution was found to satisfy constraints Took the beadm(1M) easy way out, bounced back to my r151018 BE, confirmed that all my zones were migrated to openssh and had the correct publisher, and then pkg update to r151020 succeeded including the one I fubar'd. If only all operator error was this easy to fix... John groenveld at acm.org From gearboxes at outlook.com Sun Nov 13 01:45:25 2016 From: gearboxes at outlook.com (Machine Man) Date: Sun, 13 Nov 2016 01:45:25 +0000 Subject: [OmniOS-discuss] pfSense IPv6 DAD messages Message-ID: I am on r14 LTS on all of my systems. Installing pfSense I see the following in the system log; 1000's of them on all interfaces: em0: a looped back NS message is detected during DAD for fe80:1::8:20ee:fed2:125. Another DAD probes are being sent. The only posts I could find date back to 2013 with no resolution that I could find. Would using the virtio instead of e1000 resolve this? -------------- next part -------------- An HTML attachment was scrubbed... URL: From jdg117 at elvis.arl.psu.edu Sun Nov 13 03:04:21 2016 From: jdg117 at elvis.arl.psu.edu (John D Groenveld) Date: Sat, 12 Nov 2016 22:04:21 -0500 Subject: [OmniOS-discuss] struggling to get samba 4 installed and working on omnios-r151020-b5b8c75 In-Reply-To: Your message of "Fri, 11 Nov 2016 17:44:26 MST." References: Message-ID: <201611130304.uAD34LFj000780@elvis.arl.psu.edu> In message , Joseph Boren writes: >Initially I tried the recipe here: >https://www.mail-archive.com/omnios-discuss at lists.omniti.com/msg06318.html >However I couldn't even get nettle installed via those instructions, so I Did you add -shared to the LDFLAGS? samba 4.4.7 with the latest GNU TLS stack builds for me with r151020. John groenveld at acm.org From lists at marzocchi.net Sun Nov 13 16:49:25 2016 From: lists at marzocchi.net (Olaf Marzocchi) Date: Sun, 13 Nov 2016 17:49:25 +0100 Subject: [OmniOS-discuss] no saproc in java.library.path In-Reply-To: References: Message-ID: Hello, I was trying to run hbase 1.2.4 (prerequisite for OpenTSDB) and I found that, contrary to my previous experience with hbase on OS X, it does not run out of the box (+ basic config). Instead, the java process only eats all the CPU. I tried to get a stack trace as suggested in the hbase mailing list (see the discussion at the end of this email), but jstack does not seems to work. This is the output (as also pasted in the link below): $ /usr/java/bin/jstack -m 16177 Attaching to process ID 16177, please wait... Exception in thread "main" java.lang.reflect.InvocationTargetException 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 sun.tools.jstack.JStack.runJStackTool(JStack.java:136) at sun.tools.jstack.JStack.main(JStack.java:102) Caused by: java.lang.UnsatisfiedLinkError: no saproc in java.library.path at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1889) at java.lang.Runtime.loadLibrary0(Runtime.java:849) at java.lang.System.loadLibrary(System.java:1088) at sun.jvm.hotspot.debugger.proc.ProcDebuggerLocal.(ProcDebuggerLocal.java:660) at sun.jvm.hotspot.bugspot.BugSpotAgent.setupDebuggerSolaris(BugSpotAgent.java:634) at sun.jvm.hotspot.bugspot.BugSpotAgent.setupDebugger(BugSpotAgent.java:513) at sun.jvm.hotspot.bugspot.BugSpotAgent.go(BugSpotAgent.java:492) at sun.jvm.hotspot.bugspot.BugSpotAgent.attach(BugSpotAgent.java:331) at sun.jvm.hotspot.tools.Tool.start(Tool.java:163) at sun.jvm.hotspot.tools.JStack.main(JStack.java:86) ... 6 more Both java and jdk packages are installed from main OmniOS repo and compiled for r151020, the OS version I'm running right now. I would appreciate help in the solution of this problem so that the guys from the hbase mailing list can pinpoint the issue about hbase under OmniOS. Of course, if anyone has experience with hbase on OmniOS and has seen similar issues, or is successfully running previous versions, feel free to let me know. I haven't checked any previous version, I only tested the current stable. Thanks Olaf -------- Forwarded Message -------- Subject: Re: "Quickstart" on OmniOS/illumos not working, java eats CPU Date: Sun, 13 Nov 2016 08:28:42 -0800 From: Ted Yu Reply-To: user at hbase.apache.org To: user at hbase.apache.org bq. Caused by: java.lang.UnsatisfiedLinkError: no saproc in java.library.path Was the jstack of version 1.7.0_101 ? I used google and found: http://umchee.blogspot.com/p/i-recently-needed-to-investigate.html but it was about Java 1.5 You can use this command to find the commandline for hbase process (replace 16059 with the new process Id): ps aux | grep 16059 Then use the jstack from same path as java to capture stack trace: Thanks On Sun, Nov 13, 2016 at 8:16 AM, Olaf Marzocchi wrote: > I'm not 100% sure about the stack trace (not a developer myself), but I > searched online and used jstack: > http://pastebin.com/AT6pScBv > > Additional info: after turning on DEBUG, the java process did not > immediately jump to 100% cpu as it used to do, it ramped up and it took 3 > minutes to get there. It may however be a coincidence. > > Hope this can help pinpointing the issue. > > Olaf > > > > On 13/11/2016 01:18, Ted Yu wrote: > >> Can you turn on DEBUG logging ? >> In log4j.properties : >> >> log4j.logger.org.apache.hadoop.hbase=DEBUG >> >> Since this is reproducible, can you capture one or two stack traces before >> killing the server ? >> >> Thanks >> >> On Sat, Nov 12, 2016 at 3:09 PM, Olaf Marzocchi >> wrote: >> >> Hello, >>> I tested HBase with OpenTSDB some time ago on OS X locally and the >>> quickstart guide worked perfectly, merged with >>> http://opentsdb.net/setup-hbase.html where necessary. >>> >>> I tried yesterday to do the same on my OmniOS/illumos server and I >>> encountered issues with HBase. >>> >>> First of all I set "JAVA_HOME=/usr/java" in "conf/hbase-env.sh". >>> >>> $ /usr/java/bin/java -version >>> openjdk version "1.7.0_101" >>> OpenJDK Runtime Environment (build 1.7.0_101-b00) >>> OpenJDK Server VM (build 24.101-b00, mixed mode) >>> >>> I edited also "hbase-site.xml": >>> >>> >>> >>> hbase.rootdir >>> file:///export/home/olaf/hbase >>> >>> >>> hbase.zookeeper.property.dataDir >>> /export/home/olaf/zookeeper >>> >>> >>> >>> I then launched "bin/start-hbase.sh" and it quits apparently normally. A >>> directory "/export/home/olaf/zookeeper" appeared, but no >>> /export/home/olaf/hbase. >>> >>> $ ls -l /export/home/olaf/ >>> drwxrwxr-x 3 olaf olaf 3 Nov 12 01:16 zookeeper >>> >>> I also checked with prstat ("top") and I see that java is taking a full >>> CPU. After some minutes it's still crunching something. >>> If I launch "./bin/hbase shell" it never gets to "hbase(main):001:0>". >>> At this point the only way is to kill -9 both java processes: the stop >>> script doesn't work, it outputs dots forever. >>> >>> I uploaded the two log files (.log and .out) to Pastebin: >>> http://pastebin.com/aFMXmWYU >>> >>> I'm not sure where to begin with the debugging, since I did no >>> customisation at all and I don't see any error logged. >>> >>> I would be grateful for some help. >>> >>> Thanks >>> Olaf >>> >>> >> From peter.tribble at gmail.com Sun Nov 13 17:37:31 2016 From: peter.tribble at gmail.com (Peter Tribble) Date: Sun, 13 Nov 2016 17:37:31 +0000 Subject: [OmniOS-discuss] no saproc in java.library.path In-Reply-To: References: Message-ID: On Sun, Nov 13, 2016 at 4:49 PM, Olaf Marzocchi wrote: > Hello, > I was trying to run hbase 1.2.4 (prerequisite for OpenTSDB) and I found > that, contrary to my previous experience with hbase on OS X, it does not > run out of the box (+ basic config). Instead, the java process only eats > all the CPU. > > I tried to get a stack trace as suggested in the hbase mailing list (see > the discussion at the end of this email), but jstack does not seems to work. > > This is the output (as also pasted in the link below): > > $ /usr/java/bin/jstack -m 16177 > Attaching to process ID 16177, please wait... > Exception in thread "main" java.lang.reflect.InvocationTargetException > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAcce > ssorImpl.java:57) > at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMe > thodAccessorImpl.java:43) > at java.lang.reflect.Method.invoke(Method.java:606) > at sun.tools.jstack.JStack.runJStackTool(JStack.java:136) > at sun.tools.jstack.JStack.main(JStack.java:102) > Caused by: java.lang.UnsatisfiedLinkError: no saproc in java.library.path > at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1889) > at java.lang.Runtime.loadLibrary0(Runtime.java:849) > at java.lang.System.loadLibrary(System.java:1088) > at sun.jvm.hotspot.debugger.proc.ProcDebuggerLocal.(Pro > cDebuggerLocal.java:660) > at sun.jvm.hotspot.bugspot.BugSpotAgent.setupDebuggerSolaris(Bu > gSpotAgent.java:634) > at sun.jvm.hotspot.bugspot.BugSpotAgent.setupDebugger(BugSpotAg > ent.java:513) > at sun.jvm.hotspot.bugspot.BugSpotAgent.go(BugSpotAgent.java:492) > at sun.jvm.hotspot.bugspot.BugSpotAgent.attach(BugSpotAgent. > java:331) > at sun.jvm.hotspot.tools.Tool.start(Tool.java:163) > at sun.jvm.hotspot.tools.JStack.main(JStack.java:86) > ... 6 more > > Both java and jdk packages are installed from main OmniOS repo and > compiled for r151020, the OS version I'm running right now > You're using the openjdk 1.7.0_101 that comes with omnios? You don't need hbase to trigger this. I just tested jstack with an arbitrary pid, and it fails the same way. I'm not sure why, libsaproc.so is installed where I would expect it to be, but jstack looks for it somewhere else. However, the one thing I would say is that for any real work, don't use the jdk that omnios supplies. It's old (java 7) and 32-bit only. At $DAYJOB, we use the Oracle jdk8 and it works fine. (Although my own builds of openjdk7 and openjdk8 are also fine, but it's easier to rely on someone else for maintaining something like java.) > I would appreciate help in the solution of this problem so that the guys > from the hbase mailing list can pinpoint the issue about hbase under OmniOS. > Of course, if anyone has experience with hbase on OmniOS and has seen > similar issues, or is successfully running previous versions, feel free to > let me know. I haven't checked any previous version, I only tested the > current stable. > > Thanks > Olaf > > > > -------- Forwarded Message -------- > Subject: Re: "Quickstart" on OmniOS/illumos not working, java eats CPU > Date: Sun, 13 Nov 2016 08:28:42 -0800 > From: Ted Yu > Reply-To: user at hbase.apache.org > To: user at hbase.apache.org > > bq. Caused by: java.lang.UnsatisfiedLinkError: no saproc in > java.library.path > > Was the jstack of version 1.7.0_101 ? > I used google and found: > http://umchee.blogspot.com/p/i-recently-needed-to-investigate.html > > but it was about Java 1.5 > > You can use this command to find the commandline for hbase process > (replace 16059 > with the new process Id): > ps aux | grep 16059 > > Then use the jstack from same path as java to capture stack trace: > > Thanks > > On Sun, Nov 13, 2016 at 8:16 AM, Olaf Marzocchi > wrote: > > I'm not 100% sure about the stack trace (not a developer myself), but I >> searched online and used jstack: >> http://pastebin.com/AT6pScBv >> >> Additional info: after turning on DEBUG, the java process did not >> immediately jump to 100% cpu as it used to do, it ramped up and it took 3 >> minutes to get there. It may however be a coincidence. >> >> Hope this can help pinpointing the issue. >> >> Olaf >> >> >> >> On 13/11/2016 01:18, Ted Yu wrote: >> >> Can you turn on DEBUG logging ? >>> In log4j.properties : >>> >>> log4j.logger.org.apache.hadoop.hbase=DEBUG >>> >>> Since this is reproducible, can you capture one or two stack traces >>> before >>> killing the server ? >>> >>> Thanks >>> >>> On Sat, Nov 12, 2016 at 3:09 PM, Olaf Marzocchi >>> wrote: >>> >>> Hello, >>> >>>> I tested HBase with OpenTSDB some time ago on OS X locally and the >>>> quickstart guide worked perfectly, merged with >>>> http://opentsdb.net/setup-hbase.html where necessary. >>>> >>>> I tried yesterday to do the same on my OmniOS/illumos server and I >>>> encountered issues with HBase. >>>> >>>> First of all I set "JAVA_HOME=/usr/java" in "conf/hbase-env.sh". >>>> >>>> $ /usr/java/bin/java -version >>>> openjdk version "1.7.0_101" >>>> OpenJDK Runtime Environment (build 1.7.0_101-b00) >>>> OpenJDK Server VM (build 24.101-b00, mixed mode) >>>> >>>> I edited also "hbase-site.xml": >>>> >>>> >>>> >>>> hbase.rootdir >>>> file:///export/home/olaf/hbase >>>> >>>> >>>> hbase.zookeeper.property.dataDir >>>> /export/home/olaf/zookeeper >>>> >>>> >>>> >>>> I then launched "bin/start-hbase.sh" and it quits apparently normally. A >>>> directory "/export/home/olaf/zookeeper" appeared, but no >>>> /export/home/olaf/hbase. >>>> >>>> $ ls -l /export/home/olaf/ >>>> drwxrwxr-x 3 olaf olaf 3 Nov 12 01:16 zookeeper >>>> >>>> I also checked with prstat ("top") and I see that java is taking a full >>>> CPU. After some minutes it's still crunching something. >>>> If I launch "./bin/hbase shell" it never gets to "hbase(main):001:0>". >>>> At this point the only way is to kill -9 both java processes: the stop >>>> script doesn't work, it outputs dots forever. >>>> >>>> I uploaded the two log files (.log and .out) to Pastebin: >>>> http://pastebin.com/aFMXmWYU >>>> >>>> I'm not sure where to begin with the debugging, since I did no >>>> customisation at all and I don't see any error logged. >>>> >>>> I would be grateful for some help. >>>> >>>> Thanks >>>> Olaf >>>> >>>> >>>> >>> > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Mon Nov 14 15:33:45 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 14 Nov 2016 10:33:45 -0500 Subject: [OmniOS-discuss] pfSense IPv6 DAD messages In-Reply-To: References: Message-ID: <68108C49-9F97-4DBF-BEB7-DF868B36C576@omniti.com> > On Nov 12, 2016, at 8:45 PM, Machine Man wrote: > > Would using the virtio instead of e1000 resolve this? > You're not providing nearly enough information here. pfSense doesn't work on stock OmniOS (or any illumos distro). Are you installing pfSense on a KVM VM in OmniOS? I've seen the duplicate packet issues on KVM guests like you said. And when you say e1000, do you mean to the guest? Or from the host? Please be a bit more specific. Dan From danmcd at omniti.com Mon Nov 14 16:56:17 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 14 Nov 2016 11:56:17 -0500 Subject: [OmniOS-discuss] FIX IN PLACE NOW - was Re: r151020 zone clone failure In-Reply-To: References: Message-ID: <11D22491-C14E-49F0-BD6D-55E626C632EA@omniti.com> Looping back around on this. 1.) Some people have had success with my zoneadmd binary. NOTE - it's not a completely fixed one, as an error path didn't have the fix. 2.) I'm building a replacement system/zones package which will be pushed onto the r151020 IPS repo as I type this. When you update, you will need to reboot. THIS IS A 'pkg update' TIME! 3.) I will be pushing out release media as well later today. 4.) A detailed explanation of what happened and what I needed to fix is appended. Thank you all for your help, and for catching this. I'm sorry I didn't see it pre-020. Dan ======================== A change from Joyent, which fixed a bug affecting LX zones, was pulled in by my io-lx work: commit 2d8f41b4e012f1fb3de3a6cb6731708ee5c65bcf Author: Ryan Zezeski Date: Sun Sep 25 15:55:02 2016 -0400 OS-5330 zoneadm mounting an lx or joyent branded zone fails Reviewed by: Jerry Jelinek Approved by: Jerry Jelinek (NOTE: Manual port, because of divergence from SmartOS.) PLEASE NOTE that I had to bring this one in by hand, because we don't have ALL of Joyent's zone changes. That's the first clue I had. If one debugs the "zoneadm -z dest clone source" and the /usr/lib/brand/ipkg/clone script (which was what I asked Michael to do with the -x), one sees this subcommand after the ZFS machinations of creating the new zone's filesystem: zoneadm -z dest attach -f This FAILED on lipkg/ipkg zones. Much gnashing of teeth and deep-diving showed me that for ipkg/lipkg zones, the pre-state script for them actually needs to mount the active dataset. A change in OS-5330 did NOT have zoneadmd invoking pre-state anymore for anything other than booting. This is how their LX AND NATIVE zones work, but it is not how ipkg/lipkg work. To fix this, I had to undo bits of OS-5330, to make them LX-brand specific. The actual implementation I made is a bit of a band-aid, and I know there's one other potential place that needs scrutiny going forward, but I wanted this fix out sooner rather than later. The REAL fix will be for this part of zoneadmd to query the brand about what to do. The Joyent folks confirmed my band-aid fix AND were the ones to tell me the long-term right answer is brand-querying. I will be pushing this into illumos-omnios: commit f45647bf9f5f56ad11e447895be7c32a7f3995a4 Author: Dan McDonald Date: Mon Nov 14 10:48:39 2016 -0500 Joyent's OS-5330 accidentally broke ipkg/lipkg clone via breaking "zoneadm -z zone mount -f" diff --git a/usr/src/cmd/zoneadmd/zoneadmd.c b/usr/src/cmd/zoneadmd/zoneadmd.c index 0a728000..1233fb1 100644 --- a/usr/src/cmd/zoneadmd/zoneadmd.c +++ b/usr/src/cmd/zoneadmd/zoneadmd.c @@ -554,9 +554,16 @@ static int zone_ready(zlog_t *zlogp, zone_mnt_t mount_cmd, int zstate) { int err; + boolean_t do_prestate; - if (!ALT_MOUNT(mount_cmd) && - brand_prestatechg(zlogp, zstate, Z_READY) != 0) + /* + * LX zones don't like prestate/poststate being done on non-booting + * mount commands. Other zones may, so be picky about when/when-not + * to invoke pre/post state change. + */ + do_prestate = (strcmp(brand_name, "lx") != 0 || !ALT_MOUNT(mount_cmd)); + + if (do_prestate && brand_prestatechg(zlogp, zstate, Z_READY) != 0) return (-1); if ((err = zonecfg_create_snapshot(zone_name)) != Z_OK) { @@ -580,8 +587,7 @@ zone_ready(zlog_t *zlogp, zone_mnt_t mount_cmd, int zstate) goto bad; } - if (!ALT_MOUNT(mount_cmd) && - brand_poststatechg(zlogp, zstate, Z_READY) != 0) + if (do_prestate && brand_poststatechg(zlogp, zstate, Z_READY) != 0) goto bad; return (0); @@ -591,7 +597,7 @@ bad: * If something goes wrong, we up the zones's state to the target * state, READY, and then invoke the hook as if we're halting. */ - if (!ALT_MOUNT(mount_cmd)) + if (do_prestate) (void) brand_poststatechg(zlogp, ZONE_STATE_READY, Z_HALT); return (-1); } From danmcd at omniti.com Mon Nov 14 19:09:42 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 14 Nov 2016 14:09:42 -0500 Subject: [OmniOS-discuss] r151020 install media now updated Message-ID: <0F14B1A7-D982-4BD3-B45F-BBCC8E9D7187@omniti.com> The install media for r151020 now also has the "zoneadm clone" fix in it. Thank you! Dan From lists at marzocchi.net Mon Nov 14 22:02:53 2016 From: lists at marzocchi.net (Olaf Marzocchi) Date: Mon, 14 Nov 2016 23:02:53 +0100 Subject: [OmniOS-discuss] no saproc in java.library.path In-Reply-To: References: Message-ID: <1f102c51-d66a-4c20-bd9b-eb46364d76ee@marzocchi.net> Switched to jdk1.8 and the issue disappeared. Thanks! However, someone mentioned in the hbase mailing list that the limit of 256 open files may be too low. Do you agree? my system is the result of (update after update) an original r151006, maybe I still have an old value? Thanks Olaf On 13/11/2016 18:37, Peter Tribble wrote: > > > On Sun, Nov 13, 2016 at 4:49 PM, Olaf Marzocchi > wrote: > > Hello, > I was trying to run hbase 1.2.4 (prerequisite for OpenTSDB) and I > found that, contrary to my previous experience with hbase on OS X, > it does not run out of the box (+ basic config). Instead, the java > process only eats all the CPU. > > I tried to get a stack trace as suggested in the hbase mailing list > (see the discussion at the end of this email), but jstack does not > seems to work. > > This is the output (as also pasted in the link below): > > $ /usr/java/bin/jstack -m 16177 > Attaching to process ID 16177, please wait... > Exception in thread "main" java.lang.reflect.InvocationTargetException > 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 sun.tools.jstack.JStack.runJStackTool(JStack.java:136) > at sun.tools.jstack.JStack.main(JStack.java:102) > Caused by: java.lang.UnsatisfiedLinkError: no saproc in > java.library.path > at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1889) > at java.lang.Runtime.loadLibrary0(Runtime.java:849) > at java.lang.System.loadLibrary(System.java:1088) > at > sun.jvm.hotspot.debugger.proc.ProcDebuggerLocal.(ProcDebuggerLocal.java:660) > at > sun.jvm.hotspot.bugspot.BugSpotAgent.setupDebuggerSolaris(BugSpotAgent.java:634) > at > sun.jvm.hotspot.bugspot.BugSpotAgent.setupDebugger(BugSpotAgent.java:513) > at > sun.jvm.hotspot.bugspot.BugSpotAgent.go(BugSpotAgent.java:492) > at > sun.jvm.hotspot.bugspot.BugSpotAgent.attach(BugSpotAgent.java:331) > at sun.jvm.hotspot.tools.Tool.start(Tool.java:163) > at sun.jvm.hotspot.tools.JStack.main(JStack.java:86) > ... 6 more > > Both java and jdk packages are installed from main OmniOS repo and > compiled for r151020, the OS version I'm running right now > > > You're using the openjdk 1.7.0_101 that comes with omnios? > > You don't need hbase to trigger this. I just tested jstack with an > arbitrary pid, and it > fails the same way. I'm not sure why, libsaproc.so is installed where I > would expect it > to be, but jstack looks for it somewhere else. > > However, the one thing I would say is that for any real work, don't use > the jdk that > omnios supplies. It's old (java 7) and 32-bit only. At $DAYJOB, we use > the Oracle > jdk8 and it works fine. (Although my own builds of openjdk7 and openjdk8 are > also fine, but it's easier to rely on someone else for maintaining something > like java.) > > > I would appreciate help in the solution of this problem so that the > guys from the hbase mailing list can pinpoint the issue about hbase > under OmniOS. > Of course, if anyone has experience with hbase on OmniOS and has > seen similar issues, or is successfully running previous versions, > feel free to let me know. I haven't checked any previous version, I > only tested the current stable. > > Thanks > Olaf > > > > -------- Forwarded Message -------- > Subject: Re: "Quickstart" on OmniOS/illumos not working, java eats CPU > Date: Sun, 13 Nov 2016 08:28:42 -0800 > From: Ted Yu > > Reply-To: user at hbase.apache.org > To: user at hbase.apache.org > > > > bq. Caused by: java.lang.UnsatisfiedLinkError: no saproc in > java.library.path > > Was the jstack of version 1.7.0_101 ? > I used google and found: > http://umchee.blogspot.com/p/i-recently-needed-to-investigate.html > > > but it was about Java 1.5 > > You can use this command to find the commandline for hbase process > (replace 16059 > with the new process Id): > ps aux | grep 16059 > > Then use the jstack from same path as java to capture stack trace: > > Thanks > > On Sun, Nov 13, 2016 at 8:16 AM, Olaf Marzocchi > wrote: > > I'm not 100% sure about the stack trace (not a developer > myself), but I > searched online and used jstack: > http://pastebin.com/AT6pScBv > > Additional info: after turning on DEBUG, the java process did not > immediately jump to 100% cpu as it used to do, it ramped up and > it took 3 > minutes to get there. It may however be a coincidence. > > Hope this can help pinpointing the issue. > > Olaf > > > > On 13/11/2016 01:18, Ted Yu wrote: > > Can you turn on DEBUG logging ? > In log4j.properties : > > log4j.logger.org.apache.hadoop.hbase=DEBUG > > Since this is reproducible, can you capture one or two stack > traces before > killing the server ? > > Thanks > > On Sat, Nov 12, 2016 at 3:09 PM, Olaf Marzocchi > > > wrote: > > Hello, > > I tested HBase with OpenTSDB some time ago on OS X > locally and the > quickstart guide worked perfectly, merged with > http://opentsdb.net/setup-hbase.html > where necessary. > > I tried yesterday to do the same on my OmniOS/illumos > server and I > encountered issues with HBase. > > First of all I set "JAVA_HOME=/usr/java" in > "conf/hbase-env.sh". > > $ /usr/java/bin/java -version > openjdk version "1.7.0_101" > OpenJDK Runtime Environment (build 1.7.0_101-b00) > OpenJDK Server VM (build 24.101-b00, mixed mode) > > I edited also "hbase-site.xml": > > > > hbase.rootdir > file:///export/home/olaf/hbase > > > hbase.zookeeper.property.dataDir > /export/home/olaf/zookeeper > > > > I then launched "bin/start-hbase.sh" and it quits > apparently normally. A > directory "/export/home/olaf/zookeeper" appeared, but no > /export/home/olaf/hbase. > > $ ls -l /export/home/olaf/ > drwxrwxr-x 3 olaf olaf 3 Nov 12 01:16 > zookeeper > > I also checked with prstat ("top") and I see that java > is taking a full > CPU. After some minutes it's still crunching something. > If I launch "./bin/hbase shell" it never gets to > "hbase(main):001:0>". > At this point the only way is to kill -9 both java > processes: the stop > script doesn't work, it outputs dots forever. > > I uploaded the two log files (.log and .out) to Pastebin: > http://pastebin.com/aFMXmWYU > > I'm not sure where to begin with the debugging, since I > did no > customisation at all and I don't see any error logged. > > I would be grateful for some help. > > Thanks > Olaf > > > > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > > > > > -- > -Peter Tribble > http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ From danmcd at omniti.com Mon Nov 14 22:09:56 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 14 Nov 2016 17:09:56 -0500 Subject: [OmniOS-discuss] no saproc in java.library.path In-Reply-To: <1f102c51-d66a-4c20-bd9b-eb46364d76ee@marzocchi.net> References: <1f102c51-d66a-4c20-bd9b-eb46364d76ee@marzocchi.net> Message-ID: <811D7CE0-B277-484F-9704-25D0FD0D96A2@omniti.com> > On Nov 14, 2016, at 5:02 PM, Olaf Marzocchi wrote: > > Switched to jdk1.8 and the issue disappeared. > Thanks! How did you get openjdk8 to compile on OmniOS? We thought about moving off of openjdk7, but had Real Problems (TM) making it happen. Did you bootstrap with an Oracle binary? > However, someone mentioned in the hbase mailing list that the limit of 256 open files may be too low. I don't know if we patch openjdk7 to be more than 256 files, that MIGHT be possible in our openjdk7 build. The ONLY reason, BTW, we ship any JDK is because of illumos's use of it for poold. If I could be rid of JDK, I'd eliminate it and tell people to use omniti-ms, pkgsrc, or others. > Do you agree? my system is the result of (update after update) an original r151006, maybe I still have an old value? No... our java gets (re)built with every release. Dan From lists at marzocchi.net Mon Nov 14 22:15:26 2016 From: lists at marzocchi.net (Olaf Marzocchi) Date: Mon, 14 Nov 2016 23:15:26 +0100 Subject: [OmniOS-discuss] no saproc in java.library.path In-Reply-To: <811D7CE0-B277-484F-9704-25D0FD0D96A2@omniti.com> References: <1f102c51-d66a-4c20-bd9b-eb46364d76ee@marzocchi.net> <811D7CE0-B277-484F-9704-25D0FD0D96A2@omniti.com> Message-ID: <8c44ca94-d56a-18b1-b847-c0ce3e182aa4@marzocchi.net> No compilation, too much work for me and you noticed I'm not an expert. I simply downloaded Oracle Java SE JDK 1.8.x straight from their repository. I took the solaris x64 tar.gz package and unpacked it in /opt/java/jdk1.8.x and set JAVA_HOME in hbase configuration file. The limit of 256 files seems to apply also to their JDK: http://pastebin.com/FVWP56b6 (hbase log output). Olaf On 14/11/2016 23:09, Dan McDonald wrote: > >> On Nov 14, 2016, at 5:02 PM, Olaf Marzocchi wrote: >> >> Switched to jdk1.8 and the issue disappeared. >> Thanks! > > How did you get openjdk8 to compile on OmniOS? We thought about moving off of openjdk7, but had Real Problems (TM) making it happen. Did you bootstrap with an Oracle binary? > >> However, someone mentioned in the hbase mailing list that the limit of 256 open files may be too low. > > I don't know if we patch openjdk7 to be more than 256 files, that MIGHT be possible in our openjdk7 build. > > The ONLY reason, BTW, we ship any JDK is because of illumos's use of it for poold. If I could be rid of JDK, I'd eliminate it and tell people to use omniti-ms, pkgsrc, or others. > >> Do you agree? my system is the result of (update after update) an original r151006, maybe I still have an old value? > > No... our java gets (re)built with every release. > > Dan > From manuel at oetiker.ch Wed Nov 16 12:48:58 2016 From: manuel at oetiker.ch (Manuel Oetiker) Date: Wed, 16 Nov 2016 13:48:58 +0100 (CET) Subject: [OmniOS-discuss] shown view are not found with stmfadm Message-ID: <707604323.1622917.1479300538945.JavaMail.zimbra@oetiker.ch> Hi some iscsi views are not available anymore after the upgrade to r151020. #: sbdadm list-lu | grep flex-test 600144f0888c0f000000581d879e0003 161061273600 /dev/zvol/rdsk/ssd-a/iscsi/flex-test/disk-00 #: stmfadm list-view -l 600144f0888c0f000000581d879e0003 stmfadm: 600144f0888c0f000000581d879e0003: no views found I can delete the view and create a new one. but the lun-0 is still used and not possible to assign. If I take a other lun it works. any hint thanks Manuel From danmcd at omniti.com Wed Nov 16 16:00:02 2016 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 16 Nov 2016 11:00:02 -0500 Subject: [OmniOS-discuss] shown view are not found with stmfadm In-Reply-To: <707604323.1622917.1479300538945.JavaMail.zimbra@oetiker.ch> References: <707604323.1622917.1479300538945.JavaMail.zimbra@oetiker.ch> Message-ID: > On Nov 16, 2016, at 7:48 AM, Manuel Oetiker wrote: > > Hi > > some iscsi views are not available anymore after the upgrade to r151020. From r151018? Or from an earlier revision? I need to know this to track revisions to see what may have interrupted your service. > any hint Please confirm which was the "previously working" revision. Until I hear from you, I'll assume it was r151018. Dan From manuel at oetiker.ch Wed Nov 16 16:13:18 2016 From: manuel at oetiker.ch (Manuel Oetiker) Date: Wed, 16 Nov 2016 17:13:18 +0100 (CET) Subject: [OmniOS-discuss] shown view are not found with stmfadm In-Reply-To: References: <707604323.1622917.1479300538945.JavaMail.zimbra@oetiker.ch> Message-ID: <536005310.1652461.1479312798674.JavaMail.zimbra@oetiker.ch> Hi Dan it was a omnios-r151014 updated to omnios-r151020 cheers Manuel ----- Original Message ----- > From: "Dan McDonald" > To: "Manuel Oetiker" > Cc: "omnios-discuss" , "Dan McDonald" > Sent: Wednesday, November 16, 2016 5:00:02 PM > Subject: Re: [OmniOS-discuss] shown view are not found with stmfadm >> On Nov 16, 2016, at 7:48 AM, Manuel Oetiker wrote: >> >> Hi >> >> some iscsi views are not available anymore after the upgrade to r151020. > > From r151018? Or from an earlier revision? I need to know this to track > revisions to see what may have interrupted your service. > >> any hint > > Please confirm which was the "previously working" revision. Until I hear from > you, I'll assume it was r151018. > > Dan From danmcd at omniti.com Wed Nov 16 16:44:42 2016 From: danmcd at omniti.com (Dan McDonald) Date: Wed, 16 Nov 2016 11:44:42 -0500 Subject: [OmniOS-discuss] shown view are not found with stmfadm In-Reply-To: <536005310.1652461.1479312798674.JavaMail.zimbra@oetiker.ch> References: <707604323.1622917.1479300538945.JavaMail.zimbra@oetiker.ch> <536005310.1652461.1479312798674.JavaMail.zimbra@oetiker.ch> Message-ID: > On Nov 16, 2016, at 11:13 AM, Manuel Oetiker wrote: > > it was a omnios-r151014 updated to omnios-r151020 I was hoping you'd say that, because only one COMSTAR change happened between 018 and 020, and it was a bugfix. I'll look at the larger delta set to see what happened. If you've time or patience, I wonder what 016 or 018 would do with your pool? But I don't think you do. Thanks, and stay tuned, Dan From peter.tribble at gmail.com Fri Nov 18 19:22:02 2016 From: peter.tribble at gmail.com (Peter Tribble) Date: Fri, 18 Nov 2016 19:22:02 +0000 Subject: [OmniOS-discuss] no saproc in java.library.path In-Reply-To: <811D7CE0-B277-484F-9704-25D0FD0D96A2@omniti.com> References: <1f102c51-d66a-4c20-bd9b-eb46364d76ee@marzocchi.net> <811D7CE0-B277-484F-9704-25D0FD0D96A2@omniti.com> Message-ID: Dan, How did you get openjdk8 to compile on OmniOS? We thought about moving off > of openjdk7, but had Real Problems (TM) making it happen. Did you > bootstrap with an Oracle binary? > So the following should work. Apart from the beefit in OmniOS having a current JDK, I wanted to be sure we could build from scratch just in case the Oracle binaries stopped working for any reason. This is on a vanilla r151014, updated to current. I'm Primary Administrator and have a private copy of Studio 12u1 under $HOME, so the java bootstrap is the java7 that ships with omnios. To get the sources: wget http://hg.openjdk.java.net/jdk8u/jdk8u/archive/jdk8u112-b15.tar.bz2 foreach file ( corba hotspot jaxp jaxws langtools jdk nashorn ) wget http://hg.openjdk.java.net/jdk8u/jdk8u/${file}/archive/jdk8u112-b15.tar.bz2 -O ${file}-jdk8u112-b15.tar.bz2 end bzcat jdk8u112-b15.tar.bz2 | tar xf - mv jdk8u-jdk8u112-b15 openjdk8u112 foreach file ( corba hotspot jaxp jaxws langtools jdk nashorn ) bzcat ${file}-jdk8u112-b15.tar.bz2 | tar xf - mv ${file}-jdk8u112-b15 openjdk8u112/$file end Install the 64-bit pkgsrc bootstrap https://pkgsrc.joyent.com/install-on-illumos/ Install appropriate pkgsrc and omnios packages. There was a little trial and error to get this list... pfexec pkg install developer/build/gnu-make system/header system/header/header-audio developer/macro/cpp developer/object-file developer/gnu-binutils pfexec /opt/local/bin/pkgin -y update pfexec /opt/local/bin/pkgin -y install libX11 libXrender renderproto libXext libXtst libXt xproto xextproto inputproto kbproto cups freetype2 Fix up some errant stupidities: pfexec mv /usr/bin/gobjcopy /usr/bin/not-gobjcopy-honestly pfexec ln -s /opt/local/include/X11 /usr/include pfexec ln -s /opt/local/lib/libX11.so* /usr/lib/amd64 pfexec ln -s /opt/local/lib/libXext.so* /usr/lib/amd64 pfexec ln -s /opt/local/lib/libXrender.so* /usr/lib/amd64 [Note: the gobjcopy thing is the only true magic here. You don't need the X11 symlinks for configure, but the build will fail as it doesn't seem to pass along the locations of the X11 libs and includes correctly.] cd to the openjdk8u112 directory from your download. env PATH=${HOME}/sunstudio12.1/bin:/usr/bin:/usr/sbin bash ./configure --with-milestone=fcs --with-update-version=112 --with-build-number=b15 --enable-unlimited-crypto --x-includes=/opt/local/include --x-libraries=/opt/local/lib --with-cups=/opt/local --with-freetype-include=/opt/local/include/freetype2 --with-freetype-lib=/opt/local/lib env PATH=${HOME}/sunstudio12.1/bin:/usr/bin:/usr/sbin gmake all and the built image is in build/solaris-x86_64-normal-server-release/images/j2sdk-image Then there's a little tinkering with sunpkcs11-solaris.cfg, and you'll need a full copy of cacerts, but otherwise you should be good to go. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Fri Nov 18 20:30:48 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 18 Nov 2016 15:30:48 -0500 Subject: [OmniOS-discuss] no saproc in java.library.path In-Reply-To: References: <1f102c51-d66a-4c20-bd9b-eb46364d76ee@marzocchi.net> <811D7CE0-B277-484F-9704-25D0FD0D96A2@omniti.com> Message-ID: <6943BAD1-5FE9-4B91-9AF6-3E7A20F4B2B9@omniti.com> > On Nov 18, 2016, at 2:22 PM, Peter Tribble wrote: > > a private copy of Studio 12u1 under $HOME, so the java bootstrap is the java7 that > ships with omnios. > Will the Studio 12.1 that OSol was allowed to ship with (and is the source of lint(1) for illumos builds of all flavors) work? "Private copy" always scares me, as it may not be the one that has the binary license which allows us to ship it. Dan From danmcd at omniti.com Fri Nov 18 20:31:39 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 18 Nov 2016 15:31:39 -0500 Subject: [OmniOS-discuss] no saproc in java.library.path In-Reply-To: References: <1f102c51-d66a-4c20-bd9b-eb46364d76ee@marzocchi.net> <811D7CE0-B277-484F-9704-25D0FD0D96A2@omniti.com> Message-ID: And one more thing. If you can get it to compile, you might be able to alter the build.sh script for openjdk7 in omnios-build and create an openjdk8 one. We do take PRs, after all. :) Dan From danmcd at omniti.com Fri Nov 18 20:43:40 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 18 Nov 2016 15:43:40 -0500 Subject: [OmniOS-discuss] Fwd: USB 3.0 (xHCI) Beta Testing and Review References: <6b0cc9b1-c45d-f3b8-b87b-1c4e63a5bc41@joyent.com> Message-ID: <4042C208-8AF6-4DD8-9EF2-BE2F3945ED8E@omniti.com> Pardon the top post. Robert Mustacchi of Joyent has been hard at work bringing up xhci/USB3 support for illumos. He's far enough along where he would like people to bang on it. As we did with LX during the last bloody, I've put up an alternate bloody repo, which is bloody-repo + a-USB3-illumos-omnios. It may occasionally have additional illumos-omnios changes, but for the most part it should be matched up with whatever the current bloody build is. The repo is: http://pkg.omniti.com/omnios/USB3/ Additionally, I have install media available from my own kebe.com webserver: http://kebe.com/~danmcd/USB3-media/ If you see problems, follow up on the illumos developer list per below. And yes, I expect this to make it into this bloody cycle and for r151022, aka the next LTS. Happy USB3-ing! Dan > Begin forwarded message: > > From: Robert Mustacchi > Subject: USB 3.0 (xHCI) Beta Testing and Review > Date: November 18, 2016 at 3:13:40 PM EST > To: illumos Developer > Cc: Toomas Soome , Dan McDonald , Adam ?tevko > > Hi, > > Thanks to everyone who's helped test some alpha builds of xhci, I think > we have something that's getting pretty close to being a useful initial > integration of USB 3 and xhci support. > > I'd like to thank Toomas Soome, Dan McDonald, and Adam ?tevko I have > images for SmartOS, OmniOS, OI hipster, and OI/Loader for ready testing. > While there are a few known issues with the current code, if you see > panics or hangs of any kind, please reach out to me with more > information and if at all possible, a crash dump. Please note, this is > all *beta* software. While I don't expect folks to encounter problems, > you certainly may. > > So far, we've tried to test as many of the major classes of devices as > possible. Including audio, storage, serial, Ethernet, keyboard, mouse, > hubs, etc. We've also had a lot of luck so far on the following > chipsets: VMware Fusion controller, Fresco Logic FL1100EX, Intel 8 > Series/C220 series chipset, Sunrise Point-H USB 3.0 xHCI Controller, and > Intel Celeron N2000 Series USB xHCI. > > If you do test it, and things are working, it'd help me if you could > share the output of the following two commands for my notes: > > mdb -ke '::prtusb' > prtconf -dD | grep -i xhci > > With that, here are links to builds and images: > > Toomas has put together his loader branch with the xhci changes. There > are OI ISO/USB images here: > > https://us-east.manta.joyent.com/rmustacc/public/preview/xhci/20161110/oi/OI_Loader_xhci_Text_X86.iso > https://us-east.manta.joyent.com/rmustacc/public/preview/xhci/20161110/oi/OI_Loader_xhci_Text_X86.usb > > Dan McDonald has put together OmniOS bloody repo available at: > http://pkg.omniti.com/omnios/USB3/. For more information on how to use > it, please reach out to OmniOS > > Adam ?tevko has put together an OI Hipster repository. It's available at > http://pkg.openindiana.org/hipster-xhci. For more information or > questions on the repo, please reach out to the OI Developer list. > > > SmartOS builds: > > https://us-east.manta.joyent.com/rmustacc/public/preview/xhci/20161110/smartos/platform-20161117T180610Z.tgz > https://us-east.manta.joyent.com/rmustacc/public/preview/xhci/20161110/smartos/platform-20161117T180610Z.iso > https://us-east.manta.joyent.com/rmustacc/public/preview/xhci/20161110/smartos/platform-20161117T180610Z.usb.bz2 > > > SmartOS debug builds: > > https://us-east.manta.joyent.com/rmustacc/public/preview/xhci/20161110/smartos-dbg/platform-20161116T183444Z.tgz > https://us-east.manta.joyent.com/rmustacc/public/preview/xhci/20161110/smartos-dbg/platform-20161116T183444Z.iso > https://us-east.manta.joyent.com/rmustacc/public/preview/xhci/20161110/smartos-dbg/platform-20161116T183444Z.usb.bz2 > > > Finally, if you're interested in review, we've started the process of > our public review at Joyent. It's available at > https://cr.joyent.us/#/c/872/. We will likely land this in > illumos-joyent and get a little bit more soak time and then RTI it for > illumos. > > Once again, I'd like to thank everyone who's helped test, given me > advice, or helped me through while debugging nasty bugs. > > Thanks, > Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From peter.tribble at gmail.com Fri Nov 18 20:48:46 2016 From: peter.tribble at gmail.com (Peter Tribble) Date: Fri, 18 Nov 2016 20:48:46 +0000 Subject: [OmniOS-discuss] no saproc in java.library.path In-Reply-To: <6943BAD1-5FE9-4B91-9AF6-3E7A20F4B2B9@omniti.com> References: <1f102c51-d66a-4c20-bd9b-eb46364d76ee@marzocchi.net> <811D7CE0-B277-484F-9704-25D0FD0D96A2@omniti.com> <6943BAD1-5FE9-4B91-9AF6-3E7A20F4B2B9@omniti.com> Message-ID: On Fri, Nov 18, 2016 at 8:30 PM, Dan McDonald wrote: > > > On Nov 18, 2016, at 2:22 PM, Peter Tribble > wrote: > > > > a private copy of Studio 12u1 under $HOME, so the java bootstrap is the > java7 that > > ships with omnios. > > > > Will the Studio 12.1 that OSol was allowed to ship with (and is the source > of lint(1) for illumos builds of all flavors) work? "Private copy" always > scares me, as it may not be the one that has the binary license which > allows us to ship it. > Sorry, I meant to clarify that. My "private" copy is just a tarball of that same compiler. If you're using studio12u1 for openjdk7, the exact same one will work for openjdk8. Just a lot quicker for me to ship the tarball onto a machine or VM in my home lab than burn a huge amount of bandwidth downloading it from the origin every time, given that it's pretty big. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Fri Nov 18 21:33:14 2016 From: danmcd at omniti.com (Dan McDonald) Date: Fri, 18 Nov 2016 16:33:14 -0500 Subject: [OmniOS-discuss] More for USB3 testers... References: <62dc0f14-bbdb-70fe-d571-e079c4d22a84@joyent.com> Message-ID: <70218912-E182-4184-9D2D-7DE802B535B5@omniti.com> > From: "Robert Mustacchi" > Subject: [smartos-discuss] USB 3 / xHCI beta testing > Date: November 18, 2016 at 4:26:35 PM EST > To: SmartOS discuss > Reply-To: smartos-discuss at lists.smartos.org > > Hi, > > We are about to wrap the initial set of USB 3.x support. If you have > some additional cycles to perform some testing of this (regardless of if > your system has USB 3 or not), we'd appreciate it. > > If you do test it, and things are working, it'd help me if you could > share the output of the following two commands for my notes: > > mdb -ke '::prtusb' > prtconf -dD | grep -i xhci ALSO NOTE: The USB3 repo has, like all OmniOS repos since r151018, the ability to change from a non-DEBUG kernel to a DEBUG kernel: pkg change-variant debug.illumos=true {--be-name } Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: From jboren at drakecooper.com Sat Nov 19 02:02:34 2016 From: jboren at drakecooper.com (Joseph Boren) Date: Fri, 18 Nov 2016 19:02:34 -0700 Subject: [OmniOS-discuss] struggling to get samba 4 installed and working on omnios-r151020-b5b8c75 In-Reply-To: References: Message-ID: In message <201611130304.uAD34LFj000780 at elvis.arl.psu.edu> John D Groenveld writes: >>Initially I tried the recipe here: >>https://www.mail-archive.com/omnios-discuss at lists.omniti.com/msg06318.html >>However I couldn't even get nettle installed via those instructions, so I >Did you add -shared to the LDFLAGS? >samba 4.4.7 with the latest GNU TLS stack builds for me >with r151020. Hi John, Thanks for the response, I appreciate the help. Apologies for taking so long to respond back, I got sick, then way behind at work. That's actually a perfect example of the problems I'm having: I'm a little too ignorant, and when I hit something I don't understand, generally some bit of prerequisite knowledge that I don't have, there aren't a lot of resources out there. For example, in the recipe it says: "Munge config.make link to add -shared". I honestly don't understand what that means. However, in your response you say "add -shared to the LDFLAGS" which seems to make perfect sense to me if I look at the configure line from the recipe. But I don't have the faintest idea how those 2 statements are the same or related, so I didn't add -shared to anything because I couldn't figure out what "munge config.make link" meant. But if I add -shared to the LDFLAGS and use the example configure command in the recipe it throws the warnings: WARNING: unrecognized options: --disable-static, --enable-mini-gmp and then bombs out with: configure: error: cannot run C compiled programs. If you meant to cross compile, use `--host'. So it's entirely possible I still don't understand what I'm supposed to do with -shared. Or i'm doing something else wrong. Or i'm missing some prerequisite I don't know about. The set of things "I don't know about" here is really large... :) I appreciate you taking the time to respond, thank you. Best regards, Joseph Boren IT Specialist +c: 208.891.2128 + 416 S 8th St., Boise, ID 83702 + drakecooper.com *Read our minds *Blog Facebook , Twitter , Linkedin , Instagram -------------- next part -------------- An HTML attachment was scrubbed... URL: From kiv at kiv.pp.ru Sun Nov 20 09:33:15 2016 From: kiv at kiv.pp.ru (=?UTF-8?B?0JjQu9GM0Y8g0JrRg9C70LDQs9C40L0=?=) Date: Sun, 20 Nov 2016 12:33:15 +0300 Subject: [OmniOS-discuss] In release-151020 solaris10 brand is broken Message-ID: After upgrade from 151014 to 151020, Solaris 10 zone does not boot. Console message is rather cryptic: zone 'ora10': could not set zone brand attribute.: Invalid argument zoneadm: zone 'ora10': call to zoneadmd failed But in /var/adm/messages I see (time is GMT+7) the real cause: Nov 20 15:01:22 srv01-omni1 genunix: [ID 819705 kern.notice] /usr/kernel/brand/amd64/s10_brand: undefined symbol Nov 20 15:01:22 srv01-omni1 genunix: [ID 826211 kern.notice] '__P_BRAND_DATA' Nov 20 15:01:22 srv01-omni1 genunix: [ID 472681 kern.notice] WARNING: mod_load: cannot load module 's10_brand' zone detach, pkg uninstall brand/s10, reboot, verify that there are no files in /usr/kernel/brand/amd64/ pkg install brand/s10, zone attach, then zone boot -- did not help, module does not load. Now, I will boot this zone on backup server which is still release-151014. -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3808 bytes Desc: Firma criptogr??fica S/MIME URL: From jdg117 at elvis.arl.psu.edu Sun Nov 20 14:32:47 2016 From: jdg117 at elvis.arl.psu.edu (John D Groenveld) Date: Sun, 20 Nov 2016 09:32:47 -0500 Subject: [OmniOS-discuss] Samba4 recipe In-Reply-To: Your message of "Fri, 06 May 2016 21:44:45 EDT." <201605070144.u471ijko003268@elvis.arl.psu.edu> References: <201605070144.u471ijko003268@elvis.arl.psu.edu> Message-ID: <201611201432.uAKEWlIf013407@elvis.arl.psu.edu> In message <201605070144.u471ijko003268 at elvis.arl.psu.edu>, John D Groenveld wr ites: >Just some notes for building Samba4 on r151018. > >John >groenveld at acm.org > ># pkg install developer/gcc51 developer/pkg-config \ >developer/parser/bison developer/build/gnu-make \ >developer/object-file system/library/math system/header > > > >Munge config.make link to add -shared > >$ env PATH=/usr/bin:/usr/sbin:/usr/ccs/bin:/usr/gnu/bin:/usr/sfw/bin \ >CC=gcc CFLAGS="-m64 -O3" LDFLAGS=-m64 \ >./configure --prefix=/opt/nettle --disable-static --enable-shared \ >--disable-assembler --enable-mini-gmp $ gzip -dc nettle-3.3.tar.gz | tar xf - $ cd nettle-3.3 $ env PATH=/usr/bin:/usr/sbin:/usr/ccs/bin:/usr/gnu/bin:/usr/sfw/bin \ CC=gcc CFLAGS="-m64 -O3" LDFLAGS=-m64 \ ./configure --prefix=/opt/nettle --disable-static --enable-shared \ --disable-assembler --enable-mini-gmp Edit config.make and add -shared to these lines: LIBNETTLE_LINK = $(CC) $(CFLAGS) $(LDFLAGS) -G -shared -h $(LIBNETTLE_SONAME) LIBHOGWEED_LINK = $(CC) $(CFLAGS) $(LDFLAGS) -G -shared -h $(LIBHOGWEED_SONAME) $ env PATH=/usr/bin:/usr/sbin:/usr/ccs/bin:/usr/gnu/bin:/usr/sfw/bin gmake John groenveld at acm.org From danmcd at omniti.com Mon Nov 21 00:23:12 2016 From: danmcd at omniti.com (Dan McDonald) Date: Sun, 20 Nov 2016 19:23:12 -0500 Subject: [OmniOS-discuss] [developer] USB 3.0 (xHCI) Beta Testing and Review In-Reply-To: <3d3edb7d-e87d-470d-82b4-302f8d89a50d@Spark> References: <6b0cc9b1-c45d-f3b8-b87b-1c4e63a5bc41@joyent.com> <4042C208-8AF6-4DD8-9EF2-BE2F3945ED8E@omniti.com> <3d3edb7d-e87d-470d-82b4-302f8d89a50d@Spark> Message-ID: <781B18B3-449F-4DE8-B3F2-5C6944799637@omniti.com> > On Nov 19, 2016, at 4:24 PM, gabriele.bulfon at sonicle.com wrote: > > Great one! > Any way to check the usb3 code on my XStreamOS illumos tree directly? > That would be interesting. You should be able to drop in a xhci patch from any of the above repo or review pointers. Dan From danmcd at omniti.com Mon Nov 21 00:27:20 2016 From: danmcd at omniti.com (Dan McDonald) Date: Sun, 20 Nov 2016 19:27:20 -0500 Subject: [OmniOS-discuss] In release-151020 solaris10 brand is broken In-Reply-To: References: Message-ID: > On Nov 20, 2016, at 4:33 AM, ???? ??????? wrote: > > After upgrade from 151014 to 151020, Solaris 10 zone does not boot. > > Console message is rather cryptic: > > zone 'ora10': could not set zone brand attribute.: Invalid argument > zoneadm: zone 'ora10': call to zoneadmd failed > > But in /var/adm/messages I see (time is GMT+7) the real cause: > > Nov 20 15:01:22 srv01-omni1 genunix: [ID 819705 kern.notice] > /usr/kernel/brand/amd64/s10_brand: undefined symbol > Nov 20 15:01:22 srv01-omni1 genunix: [ID 826211 kern.notice] > '__P_BRAND_DATA' > Nov 20 15:01:22 srv01-omni1 genunix: [ID 472681 kern.notice] WARNING: > mod_load: cannot load module 's10_brand' > > zone detach, pkg uninstall brand/s10, reboot, > verify that there are no files in /usr/kernel/brand/amd64/ > pkg install brand/s10, zone attach, then zone boot -- did not help, > module does not load. > > Now, I will boot this zone on backup server which is still release-151014. I didn't realize anyone was using the S10 brand. This is likely the result of the LX brand work. I see there's a __P_BRAND_DATA symbol that breaks the loading of the s10 brand. I'll look at that source and see what needs to change. Thanks, and sorry, Dan From danmcd at omniti.com Mon Nov 21 00:49:35 2016 From: danmcd at omniti.com (Dan McDonald) Date: Sun, 20 Nov 2016 19:49:35 -0500 Subject: [OmniOS-discuss] In release-151020 solaris10 brand is broken In-Reply-To: References: Message-ID: <0C86EDFB-91C6-406F-A367-C56CF2A8B3CB@omniti.com> I THINK I found the problem. I don't have S10 zones ready like you do, however. Is it possible for me to generate a replacement /usr/kernel/brand/amd64/s10_brand for you to test? Thanks, Dan From kiv at kiv.pp.ru Mon Nov 21 05:23:46 2016 From: kiv at kiv.pp.ru (=?UTF-8?B?0JjQu9GM0Y8g0JrRg9C70LDQs9C40L0=?=) Date: Mon, 21 Nov 2016 08:23:46 +0300 Subject: [OmniOS-discuss] In release-151020 solaris10 brand is broken In-Reply-To: <45EE88F1-2A62-46E2-8703-B77A6113A495@omniti.com> References: <0C86EDFB-91C6-406F-A367-C56CF2A8B3CB@omniti.com> <4C7A9418-8E59-4CA9-A095-37C164BBF54C@omniti.com> <45EE88F1-2A62-46E2-8703-B77A6113A495@omniti.com> Message-ID: Thank you very much Dan: with patched module, zone boots, our legacy system (based on ancient oracle10) starts, database mounts and works. I even did not need modunload -- because release module was not loaded :) By the way. New lx with centos6 - I think - will be a good replacement for old solaris10, for our purpose. >> On Nov 20, 2016, at 11:09 PM, Dan McDonald wrote: >> >> 1.) Backup /usr/kernel/brand/amd64/s10_brand -- that's the binary that is the problem. > Missed a step: > > 1a.) Verify the s10 brand isn't loaded: > > modinfo | grep s10_brand > > that should have empty output. If it's not empty, modunload it: > > modunload -i > > Dan > -- ? ?????????, ???? ??????? -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 3808 bytes Desc: Firma criptogr??fica S/MIME URL: From jimklimov at cos.ru Mon Nov 21 07:54:40 2016 From: jimklimov at cos.ru (Jim Klimov) Date: Mon, 21 Nov 2016 08:54:40 +0100 Subject: [OmniOS-discuss] In release-151020 solaris10 brand is broken In-Reply-To: References: <0C86EDFB-91C6-406F-A367-C56CF2A8B3CB@omniti.com> <4C7A9418-8E59-4CA9-A095-37C164BBF54C@omniti.com> <45EE88F1-2A62-46E2-8703-B77A6113A495@omniti.com> Message-ID: 21 ?????? 2016??. 6:23:46 CET, "???? ???????" ?????: >Thank you very much Dan: with patched module, zone boots, our legacy >system (based on ancient oracle10) starts, database mounts and works. > >I even did not need modunload -- because release module was not loaded >:) > >By the way. New lx with centos6 - I think - will be a good replacement >for old solaris10, for our purpose. > > >>> On Nov 20, 2016, at 11:09 PM, Dan McDonald >wrote: >>> >>> 1.) Backup /usr/kernel/brand/amd64/s10_brand -- that's the binary >that is the problem. >> Missed a step: >> >> 1a.) Verify the s10 brand isn't loaded: >> >> modinfo | grep s10_brand >> >> that should have empty output. If it's not empty, modunload it: >> >> modunload -i >> >> Dan >> But also note that Oracle DB might just work in a native local zone. Maybe the installer might complain about unexpected commands and paths, but if the deployment is concentrated in e.g. /opt/orahome (and maybe 3 scripts under /usr) you can just copy over the deployment. Or place it into a dataset and remount it (or its clone) in the other zone environment. Similarly for data file location. Hope it helps, Jim Klimov -- Typos courtesy of K-9 Mail on my Samsung Android From danmcd at omniti.com Mon Nov 21 15:01:24 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 21 Nov 2016 10:01:24 -0500 Subject: [OmniOS-discuss] In release-151020 solaris10 brand is broken In-Reply-To: References: <0C86EDFB-91C6-406F-A367-C56CF2A8B3CB@omniti.com> <4C7A9418-8E59-4CA9-A095-37C164BBF54C@omniti.com> <45EE88F1-2A62-46E2-8703-B77A6113A495@omniti.com> Message-ID: <6FC55472-9C5B-4C70-8F78-E3406EE0EDB9@omniti.com> > On Nov 21, 2016, at 12:23 AM, ???? ??????? wrote: > > Thank you very much Dan: with patched module, zone boots, our legacy > system (based on ancient oracle10) starts, database mounts and works. > > I even did not need modunload -- because release module was not loaded :) > > By the way. New lx with centos6 - I think - will be a good replacement > for old solaris10, for our purpose. Thanks for the quick test. The PROPER solution is to: 1.) Have Joyent make sure they/I didn't miss anything ELSE. This code was broken by partially backing out a change made mid-stream during Joyent's LX work. The two commits of relevance on the illumos-omnios tree are: commit 5312cd77c26a98281973239359f5c5cd5dd132b5 Author: DJ Hoffman Date: Mon Jul 7 18:06:37 2014 +0000 OS-3149 lx brand always sends SIGCHLD to parent processes, regardless of how clone was invoked OS-2887 lxbrand add WALL, WCLONE, WNOTHREAD support to waitid Reviewed by: Jerry Jelinek Reviewed by: Robert Mustacchi and the backout: commit 3bdacdb34e9adb4e92016ea49a7b9d9d3ca5e305 Author: Patrick Mooney Date: Tue Mar 31 21:57:35 2015 +0000 OS-4129 lxbrand should not abuse p_brand_data for storing exit signal Reviewed by: Jerry Jelinek Reviewed by: Joshua M. Clulow 2.) Once those are confirmed, ship replacement packages. The GOOD news is that I think we only need to ship brand/s10 and brand/sn1. No reboot should be needed for those guys (unless IPS gets something wrong). Thanks for the help! Dan From danmcd at omniti.com Mon Nov 21 16:08:17 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 21 Nov 2016 11:08:17 -0500 Subject: [OmniOS-discuss] S10 and SN1 zones have updates Message-ID: There is at least ONE person using S10 zones in the audience with r151020, because they broke thanks to some LX churn. I've updated the brand/s10 and brand/sn1 packages for those who use them (brand/sn1 is a sample brand, so I doubt that one, but S10 zones are apparently still used). Thanks to ???? for finding this, and testing the fix. The packages are now available for "pkg update" if you have them. If you don't, it's a NOP. Dan From danmcd at omniti.com Mon Nov 21 19:32:46 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 21 Nov 2016 14:32:46 -0500 Subject: [OmniOS-discuss] Fwd: New NTP release [VU#633847] References: <201611211713.uALHD5RO022895@timberline.sei.cmu.edu> Message-ID: <080362C2-454D-4A9A-AF48-A2509DBA5FF8@omniti.com> Repos for all supported releases (r151014, r151018, r151020) have been updated. The next bloody update will have it as well, when it ships. This is a non-rebooting update, but it will create a backup BE by default unless you specify --no-backup-be in pkg(1). Dan > Begin forwarded message: > > From: CERT Coordination Center > Subject: New NTP release [VU#633847] > Date: November 21, 2016 at 12:13:05 PM EST > To: Dan McDonald > Cc: CERT Coordination Center > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > Greetings, > > Today the NTP Project's ntp-4.2.8p9 was released, fixing several vulnerabilities. We have published a Vulnerability Note summary as VU#633847; the full NTP.org release notes are at . > > We are forwarding this information to you as part of our wide-distribution notices. > > If you have any questions or comments, please let us know. Please include VU#633847 in the subject line of your email. > > Best Regards, > > Vulnerability Analysis Team > ========================================================== > CERT(R) Coordination Center | cert at cert.org > Software Engineering Institute | Hotline : +1 412.268.7090 > Carnegie Mellon University | FAX : +1 412.268.6989 > Pittsburgh, PA 15213-3890 | http://www.cert.org/ > ========================================================== > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v2.0.22 (GNU/Linux) > > iQIcBAEBAgAGBQJYMyr4AAoJEOaVDwyMMQJnFVMP/A+p7QETDxfDuFC7gqfIvGnI > KM1tLa1K/p0wH0vHRmNZtt6M9mzJbECt5Ghnx5KUi5Q1DlfxYzR1UsVY01WH8B6l > 47bNxDkLy1uKzOjGgcSKp83FmF9j5ybTBJsT/K2Z5KNwyyZimCYIW/ml4kYNG8MJ > 9c3njVva46bAEWBi1cqZQswUMuOxPDYKFvyHDNHat3fLrSYglNxA0jcNPEQPFfTq > +aYh32kQaHhn7uMj/AhgW+8j+WURxInacISR6qlsXVDof/MCQwqUE/can1zsBZO3 > lmNfzg4xmPga6voBUsDy3cI2Uc6CZcMB3UBXfruqSZ+N+IYBXRwRcs5ybo4LaR2E > pu5JDufvok4SxmgcVHVZXq08OVT5yerzea+eLfF13dQ52NrLi19wV2xPOVB4uQkd > Pbha1wNdv567tGJmt9Zqp7nFn5EGDQZ0BlpeaegFFuZsZsyzVS1ZFRNAZwt6veew > fXz1VBdMkMVsrRjyndeDlYmW+1wI7R9kNvEMgAtY9PziRxoe+mMkWPGwBnEyQdyr > r5t1347OTGbyV9AsnhVVI+rwv2U8MMUp7atNg24GdgIi8ZqK2oHQlHLNCe5jLm3o > xyH4UKMmn062foR+eO86eaLf8FyP+rgVsAPvKXLJiPUKLowqEHAw1cQLkml1PsTx > dZ3jiJAkyGDroFL1a/Pz > =/jIB > -----END PGP SIGNATURE----- -------------- next part -------------- An HTML attachment was scrubbed... URL: From lorban at bitronix.be Wed Nov 23 08:24:07 2016 From: lorban at bitronix.be (Ludovic Orban) Date: Wed, 23 Nov 2016 09:24:07 +0100 Subject: [OmniOS-discuss] LX zone - installation error Message-ID: Hi, I tried following the instructions on https://omnios.omniti.com/wiki.php/LXZones to try out creating a LX zone but got rewarded with a cryptic "ERROR: " message without any further information. I've compiled a list of the information I believe would be useful double-checking what I did over here: https://gist.github.com/lorban/074e9ad5f566ef545825e5dbd99b8701 But basically, each time I try "zoneadm -z lx0 install -s /tank/lx/images/centos6.zss.gz" I only get a single line with the "ERROR: " message. Any hint? Thank you in advance, Ludovic From jaakko.linnosaari at polarshift.fi Wed Nov 23 10:11:16 2016 From: jaakko.linnosaari at polarshift.fi (Jaakko Linnosaari) Date: Wed, 23 Nov 2016 12:11:16 +0200 Subject: [OmniOS-discuss] LX zone - installation error In-Reply-To: References: Message-ID: <1797DBCA-9539-4DE1-AF1A-178E395D7A0F@polarshift.fi> > On 23 Nov 2016, at 10.24, Ludovic Orban wrote: > > I've compiled a list of the information I believe would be useful > double-checking what I did over here: > https://gist.github.com/lorban/074e9ad5f566ef545825e5dbd99b8701 > > But basically, each time I try "zoneadm -z lx0 install -s > /tank/lx/images/centos6.zss.gz" I only get a single line with the > "ERROR: " message. > > Any hint? Try creating zfs filesystems for tank/lx and tank/lx/lx0. Those seem to be missing. eg. ?cd /tank; mv lx lx.orig; zfs create tank/lx; zfs create tank/lx/lx0' ? Jaakko From lorban at bitronix.be Wed Nov 23 11:47:54 2016 From: lorban at bitronix.be (Ludovic Orban) Date: Wed, 23 Nov 2016 12:47:54 +0100 Subject: [OmniOS-discuss] LX zone - installation error In-Reply-To: <1797DBCA-9539-4DE1-AF1A-178E395D7A0F@polarshift.fi> References: <1797DBCA-9539-4DE1-AF1A-178E395D7A0F@polarshift.fi> Message-ID: That was the exact cause of my problem, thanks for your help! Now maybe that should be documented on the LXZones wiki page (I checked, and it doesn't speak about having to create a filesystem) and maybe the zoneadm command could report a more useful error message? Anyway, problem solved! Thanks again, Ludovic On Wed, Nov 23, 2016 at 11:11 AM, Jaakko Linnosaari wrote: > >> On 23 Nov 2016, at 10.24, Ludovic Orban wrote: >> >> I've compiled a list of the information I believe would be useful >> double-checking what I did over here: >> https://gist.github.com/lorban/074e9ad5f566ef545825e5dbd99b8701 >> >> But basically, each time I try "zoneadm -z lx0 install -s >> /tank/lx/images/centos6.zss.gz" I only get a single line with the >> "ERROR: " message. >> >> Any hint? > > Try creating zfs filesystems for tank/lx and tank/lx/lx0. Those seem to be missing. > > eg. ?cd /tank; mv lx lx.orig; zfs create tank/lx; zfs create tank/lx/lx0' > > > ? Jaakko From lorban at bitronix.be Wed Nov 23 12:28:38 2016 From: lorban at bitronix.be (Ludovic Orban) Date: Wed, 23 Nov 2016 13:28:38 +0100 Subject: [OmniOS-discuss] LX zone - installation error In-Reply-To: References: <1797DBCA-9539-4DE1-AF1A-178E395D7A0F@polarshift.fi> Message-ID: May I abuse your time a bit more? Is there some doc somewhere listing all the different zonecfg properties and attributes? I'd like to lofs mount some folder on my global zone in my lx zone but I can't seem to find how to translate the smartos vmadm config into a omnios zonecfg one. Thanks again, Ludovic On Wed, Nov 23, 2016 at 12:47 PM, Ludovic Orban wrote: > That was the exact cause of my problem, thanks for your help! > > Now maybe that should be documented on the LXZones wiki page (I > checked, and it doesn't speak about having to create a filesystem) and > maybe the zoneadm command could report a more useful error message? > > Anyway, problem solved! > > Thanks again, > Ludovic > > On Wed, Nov 23, 2016 at 11:11 AM, Jaakko Linnosaari > wrote: >> >>> On 23 Nov 2016, at 10.24, Ludovic Orban wrote: >>> >>> I've compiled a list of the information I believe would be useful >>> double-checking what I did over here: >>> https://gist.github.com/lorban/074e9ad5f566ef545825e5dbd99b8701 >>> >>> But basically, each time I try "zoneadm -z lx0 install -s >>> /tank/lx/images/centos6.zss.gz" I only get a single line with the >>> "ERROR: " message. >>> >>> Any hint? >> >> Try creating zfs filesystems for tank/lx and tank/lx/lx0. Those seem to be missing. >> >> eg. ?cd /tank; mv lx lx.orig; zfs create tank/lx; zfs create tank/lx/lx0' >> >> >> ? Jaakko From rafapi at gmail.com Wed Nov 23 13:03:23 2016 From: rafapi at gmail.com (Rafael Pardinas) Date: Wed, 23 Nov 2016 13:03:23 +0000 Subject: [OmniOS-discuss] LX zone - installation error In-Reply-To: References: <1797DBCA-9539-4DE1-AF1A-178E395D7A0F@polarshift.fi> Message-ID: It's pretty straight forward, the commands seem to be implemented following the Solaris specs: add fs set dir=/media (zone full path) set special=/tank/media (global zone path) set type=lofs end The Solaris docs I followed are these: https://docs.oracle.com/cd/E19044-01/sol.containers/817-1592/gcziw/index.html On Wed, 23 Nov 2016 at 12:31 Ludovic Orban wrote: > May I abuse your time a bit more? > > Is there some doc somewhere listing all the different zonecfg > properties and attributes? I'd like to lofs mount some folder on my > global zone in my lx zone but I can't seem to find how to translate > the smartos vmadm config into a omnios zonecfg one. > > Thanks again, > Ludovic > > On Wed, Nov 23, 2016 at 12:47 PM, Ludovic Orban > wrote: > > That was the exact cause of my problem, thanks for your help! > > > > Now maybe that should be documented on the LXZones wiki page (I > > checked, and it doesn't speak about having to create a filesystem) and > > maybe the zoneadm command could report a more useful error message? > > > > Anyway, problem solved! > > > > Thanks again, > > Ludovic > > > > On Wed, Nov 23, 2016 at 11:11 AM, Jaakko Linnosaari > > wrote: > >> > >>> On 23 Nov 2016, at 10.24, Ludovic Orban wrote: > >>> > >>> I've compiled a list of the information I believe would be useful > >>> double-checking what I did over here: > >>> https://gist.github.com/lorban/074e9ad5f566ef545825e5dbd99b8701 > >>> > >>> But basically, each time I try "zoneadm -z lx0 install -s > >>> /tank/lx/images/centos6.zss.gz" I only get a single line with the > >>> "ERROR: " message. > >>> > >>> Any hint? > >> > >> Try creating zfs filesystems for tank/lx and tank/lx/lx0. Those seem to > be missing. > >> > >> eg. ?cd /tank; mv lx lx.orig; zfs create tank/lx; zfs create > tank/lx/lx0' > >> > >> > >> ? Jaakko > _______________________________________________ > 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 lorban at bitronix.be Wed Nov 23 15:57:40 2016 From: lorban at bitronix.be (Ludovic Orban) Date: Wed, 23 Nov 2016 16:57:40 +0100 Subject: [OmniOS-discuss] LX zone - installation error In-Reply-To: References: <1797DBCA-9539-4DE1-AF1A-178E395D7A0F@polarshift.fi> Message-ID: Indeed, that works exactly like that. Thanks a ton! Ludovic On Wed, Nov 23, 2016 at 2:03 PM, Rafael Pardinas wrote: > It's pretty straight forward, the commands seem to be implemented following > the Solaris specs: > > add fs > set dir=/media (zone full path) > set special=/tank/media (global zone path) > set type=lofs > end > > The Solaris docs I followed are these: > https://docs.oracle.com/cd/E19044-01/sol.containers/817-1592/gcziw/index.html > > > > On Wed, 23 Nov 2016 at 12:31 Ludovic Orban wrote: >> >> May I abuse your time a bit more? >> >> Is there some doc somewhere listing all the different zonecfg >> properties and attributes? I'd like to lofs mount some folder on my >> global zone in my lx zone but I can't seem to find how to translate >> the smartos vmadm config into a omnios zonecfg one. >> >> Thanks again, >> Ludovic >> >> On Wed, Nov 23, 2016 at 12:47 PM, Ludovic Orban >> wrote: >> > That was the exact cause of my problem, thanks for your help! >> > >> > Now maybe that should be documented on the LXZones wiki page (I >> > checked, and it doesn't speak about having to create a filesystem) and >> > maybe the zoneadm command could report a more useful error message? >> > >> > Anyway, problem solved! >> > >> > Thanks again, >> > Ludovic >> > >> > On Wed, Nov 23, 2016 at 11:11 AM, Jaakko Linnosaari >> > wrote: >> >> >> >>> On 23 Nov 2016, at 10.24, Ludovic Orban wrote: >> >>> >> >>> I've compiled a list of the information I believe would be useful >> >>> double-checking what I did over here: >> >>> https://gist.github.com/lorban/074e9ad5f566ef545825e5dbd99b8701 >> >>> >> >>> But basically, each time I try "zoneadm -z lx0 install -s >> >>> /tank/lx/images/centos6.zss.gz" I only get a single line with the >> >>> "ERROR: " message. >> >>> >> >>> Any hint? >> >> >> >> Try creating zfs filesystems for tank/lx and tank/lx/lx0. Those seem to >> >> be missing. >> >> >> >> eg. ?cd /tank; mv lx lx.orig; zfs create tank/lx; zfs create >> >> tank/lx/lx0' >> >> >> >> >> >> ? Jaakko >> _______________________________________________ >> OmniOS-discuss mailing list >> OmniOS-discuss at lists.omniti.com >> http://lists.omniti.com/mailman/listinfo/omnios-discuss From Dirk.Willems at exitas.be Fri Nov 25 12:49:33 2016 From: Dirk.Willems at exitas.be (Willems Dirk) Date: Fri, 25 Nov 2016 12:49:33 +0000 Subject: [OmniOS-discuss] unable to add second nic on KVM In-Reply-To: <71c6fa3e-aede-2dd8-30e3-53e008c4aa58@exitas.be> References: <71c6fa3e-aede-2dd8-30e3-53e008c4aa58@exitas.be> Message-ID: <642d26ab-6da6-e94b-0a71-e71423dca908@exitas.be> Hello, First of all Thank you for this wonderful OS. I'm using it for Home virtualization and have the same server also running on my Work for testing purpose and want to use it for Production servers by my clients. But I've a small problem. I'm trying to add a second nic to the KVM image but when I start the KVM I can't connect anymore over ssh or ping the global zone and have to kill the KVM process. root at OmniosExitas:/root# pkg info entire Name: entire Summary: Incorporation to constrain core system packages to same build Description: This package constrains core system package versions to the same build and provides a minimal set of additional packages. State: Installed Publisher: omnios Version: 11 Branch: 0.151020 Packaging Date: 2. November 2016 01:21:08 Size: 1.75 kB FMRI: pkg://omnios/entire at 11-0.151020:20161102T012108Z my dladm output root at OmniosExitas:/root# dladm show-link LINK CLASS MTU STATE BRIDGE OVER igb0 phys 1500 up -- -- igb1 phys 1500 up -- -- aggr0 aggr 1500 up -- igb0,igb1 VLJJ01 vnic 1500 up -- aggr0 VLJJ02 vnic 1500 up -- aggr0 VSwitchJJ0 etherstub 9000 up -- -- VSJJ01 vnic 9000 up -- VSwitchJJ0 VSJJ02 vnic 9000 up -- VSwitchJJ0 my script what I use #!/usr/bin/bash NAME=JJ01 # configuration VNIC1=VLJJ01 VNIC2=VSJJ01 # Sample zvol path. HDD=/dev/zvol/rdsk/KVM_pool/JJ01 # NOTE: You can add more... (see NOTE2 below) #HDD2=/dev/zvol/rdsk/data/vm-data-zvol CD=/Downloads/oel6.8.iso VNC=1 # Memory for the KVM instance, in Megabytes (2^20 bytes). MEM=10240 mac1=`dladm show-vnic -po macaddress $VNIC1` mac2=`dladm show-vnic -po macaddress $VNIC2` /usr/bin/qemu-system-x86_64 \ -name "$(basename $NAME)" \ -boot cd \ -enable-kvm \ -vnc 0.0.0.0:$VNC \ -smp 2 \ -m $MEM \ -no-hpet \ -localtime \ -drive file=$HDD,if=ide,index=0 \ -drive file=$CD,media=cdrom,if=ide,index=2 \ -net nic,vlan=0,name=net0,model=e1000,macaddr=$mac1 \ -net vnic,vlan=0,name=net0,ifname=$VNIC1,macaddr=$mac1, \ -net nic,vlan=0,name=net1,model=e1000,macaddr=$mac2 \ -net vnic,vlan=0,name=net1,ifname=$VNIC2,macaddr=$mac2, \ -vga std \ -daemonize # NOTE2: Add an additional -drive file=$HDD2,if=ide,index=X for a unique X # in the big command-line above if you want another drive. Repeat as needed. if [ $? -gt 0 ]; then echo "Failed to start VM" fi # TCP port for VNC connections to the KVM instance. 5900 is added in the command. port=`expr 5900 + $VNC` public_nic=$(dladm show-vnic|grep vnic1|awk '{print $2}') public_ip=$(ifconfig $public_nic|grep inet|awk '{print $2}') echo "Started VM:" echo "Public: ${public_ip}:${port}" I don't see any suspicious in the messages only following /var/adm/messages [cid:part2.353ACBB6.6DAD3CB6 at exitas.be] Thanks in advance. Kind Regards. Dirk Willems -- [cid:part3.D6CE8210.C4479A2D at exitas.be] Dirk Willems System Engineer [cid:part5.6F22BB88.3617876A at exitas.be] +32 (0)3 443 12 38 [cid:part7.7200D0A3.5DEB2526 at exitas.be] dirk.willems at exitas.be [cid:part9.7E20E0B9.94C6903E at exitas.be] Veldkant 31 - 2550 Kontich [cid:part11.7E270D5C.AC2FAB09 at exitas.be] www.exitas.be Disclaimer -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ATT00001.png Type: image/png Size: 42899 bytes Desc: ATT00001.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ATT00002.png Type: image/png Size: 8486 bytes Desc: ATT00002.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ATT00003.png Type: image/png Size: 524 bytes Desc: ATT00003.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ATT00004.png Type: image/png Size: 791 bytes Desc: ATT00004.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ATT00005.png Type: image/png Size: 803 bytes Desc: ATT00005.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ATT00006.png Type: image/png Size: 796 bytes Desc: ATT00006.png URL: From skeltonr at btconnect.com Fri Nov 25 19:47:29 2016 From: skeltonr at btconnect.com (Richard Skelton) Date: Fri, 25 Nov 2016 19:47:29 +0000 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 Message-ID: <58389551.7020305@btconnect.com> Hi All, I upgraded my system to r151020 from r151018 this morning but now I see :- root at myhost:/root# fmadm faulty --------------- ------------------------------------ -------------- --------- TIME EVENT-ID MSG-ID SEVERITY --------------- ------------------------------------ -------------- --------- Nov 25 11:13:16 d0babcfc-cda3-618d-ca1e-e0b97111e6d1 PCIEX-8000-0A Critical Host : myhost Platform : ProLiant-DL385-G7 Chassis_id : CZ214902CB Product_sn : Fault class : fault.io.pciex.device-interr Affects : dev:////pci at 0,0/pci1002,5a1f at b/pci10b5,8616 at 0/pci10b5,8616 at 6/pci8086,3 at 0,1 faulted and taken out of service FRU : "MB/PCI-E Slot 3" (hc://:product-id=ProLiant-DL385-G7:server-id=myhost:chassis-id=CZ214902CB/motherboard=0/hostbridge=6/pciexrc=6/pciexbus=9/pciexdev=0/pciexfn=0/pciexbus=10/pciexdev=6/pciexfn=0/pciexbus=15/pciexdev=0) faulty Description : A problem was detected for a PCIEX device. Refer to http://illumos.org/msg/PCIEX-8000-0A for more information. Response : One or more device instances may be disabled Impact : Loss of services provided by the device instances associated with this fault Action : Schedule a repair procedure to replace the affected device. Use fmadm faulty to identify the device or contact Sun for support. --------------- ------------------------------------ -------------- --------- TIME EVENT-ID MSG-ID SEVERITY --------------- ------------------------------------ -------------- --------- Sep 13 15:59:14 540f3aea-a6ab-e70e-8c6b-8b79bba9d1eb PCIEX-8000-0A Critical Host : myhost Platform : ProLiant-DL385-G7 Chassis_id : CZ214902CB Product_sn : Fault class : fault.io.pciex.device-interr Affects : dev:////pci at 0,0/pci1002,5a1f at b/pci10b5,8616 at 0/pci10b5,8616 at 6/pci8086,3 at 0 faulted and taken out of service FRU : "MB/PCI-E Slot 3" (hc://:product-id=ProLiant-DL385-G7:server-id=myhost:chassis-id=CZ214902CB/motherboard=0/hostbridge=6/pciexrc=6/pciexbus=9/pciexdev=0/pciexfn=0/pciexbus=10/pciexdev=6/pciexfn=0/pciexbus=15/pciexdev=0) faulty Description : A problem was detected for a PCIEX device. Refer to http://illumos.org/msg/PCIEX-8000-0A for more information. Response : One or more device instances may be disabled Impact : Loss of services provided by the device instances associated with this fault Action : Schedule a repair procedure to replace the affected device. Use fmadm faulty to identify the device or contact Sun for support I also see :- root at myhost:/root# fmdump -e Nov 25 12:37:06.0565 ereport.io.service.lost Nov 25 12:37:06.1043 ereport.io.device.inval_state Nov 25 12:37:09.1068 ereport.io.service.lost Nov 25 12:37:27.0156 ereport.io.device.inval_state Nov 25 12:37:30.0181 ereport.io.service.lost Nov 25 12:37:30.0661 ereport.io.device.inval_state Nov 25 12:37:33.0686 ereport.io.service.lost Nov 25 12:37:33.1149 ereport.io.device.inval_state root at myhost:/root# prtdiag -v System Configuration: HP ProLiant DL385 G7 BIOS Configuration: HP A18 02/02/2014 BMC Configuration: IPMI 2.0 (KCS: Keyboard Controller Style) ==== Processor Sockets ==================================== Version Location Tag -------------------------------- -------------------------- AMD Opteron(tm) Processor 6176 SE Proc 1 AMD Opteron(tm) Processor 6176 SE Proc 2 ==== Memory Device Sockets ================================ Type Status Set Device Locator Bank Locator ----------- ------ --- ------------------- ---------------- DDR3 in use 1 Proc 1 DIMM 1A DDR3 empty 2 Proc 1 DIMM 2I DDR3 in use 3 Proc 1 DIMM 3E DDR3 in use 4 Proc 1 DIMM 4C DDR3 empty 5 Proc 1 DIMM 5K DDR3 in use 6 Proc 1 DIMM 6G DDR3 in use 7 Proc 1 DIMM 7B DDR3 empty 8 Proc 1 DIMM 8J DDR3 in use 9 Proc 1 DIMM 9F DDR3 in use 10 Proc 1 DIMM 10D DDR3 empty 11 Proc 1 DIMM 11L DDR3 in use 12 Proc 1 DIMM 12H DDR3 in use 13 Proc 2 DIMM 1A DDR3 empty 14 Proc 2 DIMM 2I DDR3 in use 15 Proc 2 DIMM 3E DDR3 in use 16 Proc 2 DIMM 4C DDR3 empty 17 Proc 2 DIMM 5K DDR3 in use 18 Proc 2 DIMM 6G DDR3 in use 19 Proc 2 DIMM 7B DDR3 empty 20 Proc 2 DIMM 8J DDR3 in use 21 Proc 2 DIMM 9F DDR3 in use 22 Proc 2 DIMM 10D DDR3 empty 23 Proc 2 DIMM 11L DDR3 in use 24 Proc 2 DIMM 12H ==== On-Board Devices ===================================== ==== Upgradeable Slots ==================================== ID Status Type Description --- --------- ---------------- ---------------------------- 1 available PCI Exp. Gen 2 x8 PCI-E Slot 1 2 in use PCI Exp. Gen 2 x4 PCI-E Slot 2, LSI Logic / Symbios Logic SAS2008 PCI-Express Fusion-MPT SAS-2 [Falcon] (mpt_sas) 3 in use PCI Exp. Gen 2 x4 PCI-E Slot 3, Intel Corporation 82599ES 10-Gigabit SFI/SFP+ Network Connection (ixgbe) 4 available PCI Exp. Gen 2 x8 PCI-E Slot 4 5 in use PCI Exp. Gen 2 x4 PCI-E Slot 5, Intel Corporation PCIe Data Center SSD (nvme) 6 in use PCI Exp. Gen 2 x4 PCI-E Slot 6, Intel Corporation PCIe Data Center SSD (nvme) If I run dladm show-phys it just hangs :-) Any ideas? Cheers Richard. -------------- next part -------------- An HTML attachment was scrubbed... URL: From bardishe at gmail.com Sat Nov 26 18:46:43 2016 From: bardishe at gmail.com (Artyom Zhandarovsky) Date: Sun, 27 Nov 2016 05:46:43 +1100 Subject: [OmniOS-discuss] l2arc cache size 16.0E Message-ID: Hi everyone. I have a situation with incorrect size of l2arc cache ssd`s, as for me this is because of compression current compressratio is DR_data/volume_0 compressratio 1.69x - Is there some fix to solve, cause i think it is not good. But if it wont affect system stability well i think i can live with this sometime... size of SSD`s is 200Gb capacity operations bandwidth pool alloc free read write read write ----------- ----- ----- ----- ----- ----- ----- DR_data 2.85T 40.6T 0 80 19.1K 831K raidz1 981G 13.5T 0 21 11.4K 196K c4t1d0 - - 0 6 1.65K 29.8K c4t2d0 - - 0 4 1.56K 28.7K c4t3d0 - - 0 6 1.10K 29.6K c4t4d0 - - 0 5 1.65K 29.0K c4t5d0 - - 0 6 1.56K 30.2K c4t6d0 - - 0 6 1.56K 29.5K c4t7d0 - - 0 6 1.74K 29.9K c4t8d0 - - 0 6 563 29.8K raidz1 949G 13.6T 0 21 0 201K c4t10d0 - - 0 6 0 30.4K c4t11d0 - - 0 5 0 29.5K c4t12d0 - - 0 6 0 30.2K c4t13d0 - - 0 5 0 29.7K c4t14d0 - - 0 6 0 30.7K c4t15d0 - - 0 6 0 30.5K c4t16d0 - - 0 6 0 30.5K c4t9d0 - - 0 6 0 30.5K raidz1 991G 13.5T 0 20 7.71K 159K c4t17d0 - - 0 6 1.10K 24.8K c4t18d0 - - 0 5 563 24.4K c4t19d0 - - 0 5 1.19K 24.5K c4t20d0 - - 0 6 563 24.5K c4t21d0 - - 0 5 1.10K 24.4K c4t22d0 - - 0 4 1.10K 23.1K c4t23d0 - - 0 5 1.10K 24.3K c4t24d0 - - 0 5 1.01K 23.8K logs - - - - - - c1t1d0 4.82M 186G 0 8 0 144K c1t2d0 4.73M 186G 0 8 0 131K cache - - - - - - * c1t3d0 291G 16.0E 0 0 3.30K 0* * c1t4d0 291G 16.0E 0 0 15.8K 0* ----------- ----- ----- ----- ----- ----- ----- oot at kup-san-02:/root# zpool status -v DR_data pool: DR_data state: ONLINE scan: none requested config: NAME STATE READ WRITE CKSUM DR_data ONLINE 0 0 0 raidz1-0 ONLINE 0 0 0 c4t1d0 ONLINE 0 0 0 c4t2d0 ONLINE 0 0 0 c4t3d0 ONLINE 0 0 0 c4t4d0 ONLINE 0 0 0 c4t5d0 ONLINE 0 0 0 c4t6d0 ONLINE 0 0 0 c4t7d0 ONLINE 0 0 0 c4t8d0 ONLINE 0 0 0 raidz1-1 ONLINE 0 0 0 c4t10d0 ONLINE 0 0 0 c4t11d0 ONLINE 0 0 0 c4t12d0 ONLINE 0 0 0 c4t13d0 ONLINE 0 0 0 c4t14d0 ONLINE 0 0 0 c4t15d0 ONLINE 0 0 0 c4t16d0 ONLINE 0 0 0 c4t9d0 ONLINE 0 0 0 raidz1-2 ONLINE 0 0 0 c4t17d0 ONLINE 0 0 0 c4t18d0 ONLINE 0 0 0 c4t19d0 ONLINE 0 0 0 c4t20d0 ONLINE 0 0 0 c4t21d0 ONLINE 0 0 0 c4t22d0 ONLINE 0 0 0 c4t23d0 ONLINE 0 0 0 c4t24d0 ONLINE 0 0 0 logs c1t1d0 ONLINE 0 0 0 c1t2d0 ONLINE 0 0 0 cache c1t3d0 ONLINE 0 0 0 c1t4d0 ONLINE 0 0 0 spares c4t0d0 AVAIL errors: No known data errors root at kup-san-02:/root# root at kup-san-02:/root# zpool list -v DR_data NAME SIZE ALLOC FREE EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT DR_data 43.5T 2.85T 40.6T - 4% 6% 1.00x ONLINE - raidz1 14.5T 981G 13.5T - 4% 6% c4t1d0 - - - - - - c4t2d0 - - - - - - c4t3d0 - - - - - - c4t4d0 - - - - - - c4t5d0 - - - - - - c4t6d0 - - - - - - c4t7d0 - - - - - - c4t8d0 - - - - - - raidz1 14.5T 949G 13.6T - 4% 6% c4t10d0 - - - - - - c4t11d0 - - - - - - c4t12d0 - - - - - - c4t13d0 - - - - - - c4t14d0 - - - - - - c4t15d0 - - - - - - c4t16d0 - - - - - - c4t9d0 - - - - - - raidz1 14.5T 991G 13.5T - 4% 6% c4t17d0 - - - - - - c4t18d0 - - - - - - c4t19d0 - - - - - - c4t20d0 - - - - - - c4t21d0 - - - - - - c4t22d0 - - - - - - c4t23d0 - - - - - - c4t24d0 - - - - - - log - - - - - - c1t1d0 186G 5.13M 186G - 8% 0% c1t2d0 186G 5.13M 186G - 7% 0% cache - - - - - - c1t3d0 186G 291G 16.0E - 0% 156% c1t4d0 186G 291G 16.0E - 0% 156% spare - - - - - - c4t0d0 - - - - - - root at kup-san-02:/root# -------------- next part -------------- An HTML attachment was scrubbed... URL: From natxo.asenjo at gmail.com Sun Nov 27 20:00:56 2016 From: natxo.asenjo at gmail.com (Natxo Asenjo) Date: Sun, 27 Nov 2016 21:00:56 +0100 Subject: [OmniOS-discuss] plex on lx zone (centos 6.8) stops working after 2/3 days Message-ID: hi, the subject says it. I installed a centos lx zone following the fine manual (https://omnios.omniti.com/wiki.php/LXZones) and it works great, but plex disappears from the network every 2 to three days. Rebooting the zone solves the problem until it happens again. How could I troubleshoot this issue? The zone os is available (I can login using ssh, so networking works, only plex does not broadcast). TIA -- Groeten, natxo -------------- next part -------------- An HTML attachment was scrubbed... URL: From mtalbott at lji.org Sun Nov 27 21:04:50 2016 From: mtalbott at lji.org (Michael Talbott) Date: Sun, 27 Nov 2016 13:04:50 -0800 Subject: [OmniOS-discuss] plex on lx zone (centos 6.8) stops working after 2/3 days In-Reply-To: References: Message-ID: <5F8422A7-850F-4008-A4AF-FC9B1A5E4136@lji.org> First thought is dhcp lease time expires, maybe something gets a new IP and the plex service doesn't reregister? Are you using dhcp or a static IP? What's your zone config? Michael Sent from my iPhone > On Nov 27, 2016, at 12:00 PM, Natxo Asenjo wrote: > > hi, > > the subject says it. I installed a centos lx zone following the fine manual (https://omnios.omniti.com/wiki.php/LXZones) and it works great, but plex disappears from the network every 2 to three days. Rebooting the zone solves the problem until it happens again. > > How could I troubleshoot this issue? The zone os is available (I can login using ssh, so networking works, only plex does not broadcast). > > TIA > > -- > Groeten, > natxo > _______________________________________________ > 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 Sun Nov 27 22:00:18 2016 From: danmcd at omniti.com (Dan McDonald) Date: Sun, 27 Nov 2016 17:00:18 -0500 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: <58389551.7020305@btconnect.com> References: <58389551.7020305@btconnect.com> Message-ID: <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> Check for any "ixgbe" output in /var/adm/messages* on your system as well, in case the driver was complaining about something. It's possible the newer ixgbe is detecting something wrong with your board now. ISTR some additional checks added. Dan From danmcd at omniti.com Sun Nov 27 22:02:17 2016 From: danmcd at omniti.com (Dan McDonald) Date: Sun, 27 Nov 2016 17:02:17 -0500 Subject: [OmniOS-discuss] l2arc cache size 16.0E In-Reply-To: References: Message-ID: <7CDEC300-AA33-4931-9CBF-00D4F96BBD80@omniti.com> Which version of OmniOS? "uname -v" along with the /etc/release file if it's older than r151018. Dan From bardishe at gmail.com Sun Nov 27 22:09:00 2016 From: bardishe at gmail.com (Artyom Zhandarovsky) Date: Mon, 28 Nov 2016 09:09:00 +1100 Subject: [OmniOS-discuss] l2arc cache size 16.0E In-Reply-To: <7CDEC300-AA33-4931-9CBF-00D4F96BBD80@omniti.com> References: <7CDEC300-AA33-4931-9CBF-00D4F96BBD80@omniti.com> Message-ID: root at kup-san-02:/root# uname -snrvmapiX SunOS kup-san-02 5.11 omnios-r151020-b5b8c75 i86pc i386 i86pcSystem = SunOS Node = kup-san-02 Release = 5.11 KernelID = omnios-r151020-b5b8c75 Machine = i86pc BusType = Serial = Users = OEM# = 0 Origin# = 1 NumCPU = 12 root at kup-san-02:/root# uname -v omnios-r151020-b5b8c75 root at kup-san-02:/root# cat /etc/release OmniOS v11 r151020 Copyright 2016 OmniTI Computer Consulting, Inc. All rights reserved. Use is subject to license terms. root at kup-san-02:/root# 2016-11-28 9:02 GMT+11:00 Dan McDonald : > Which version of OmniOS? "uname -v" along with the /etc/release file if > it's older than r151018. > > Dan > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From natxo.asenjo at gmail.com Mon Nov 28 05:53:38 2016 From: natxo.asenjo at gmail.com (Natxo Asenjo) Date: Mon, 28 Nov 2016 06:53:38 +0100 Subject: [OmniOS-discuss] plex on lx zone (centos 6.8) stops working after 2/3 days In-Reply-To: <5F8422A7-850F-4008-A4AF-FC9B1A5E4136@lji.org> References: <5F8422A7-850F-4008-A4AF-FC9B1A5E4136@lji.org> Message-ID: hi, this is the zone export info: create -b set zonepath=/tank/zones/plex set brand=lx set autoboot=true set ip-type=exclusive add net set physical=vnic2 add property (name=gateway,value="192.168.0.254") add property (name=ips,value="192.168.0.120") add property (name=primary,value="true") add property (name=resolver,value="192.168.0.1") end add attr set name=kernel-version set type=string set value=2.6.32 end So that is a fixed ip I tried using dhcp but it would not work. I'll double check the dhcp server logs to ensure no double ip addresses have been leased. On Sun, Nov 27, 2016 at 10:04 PM, Michael Talbott wrote: > First thought is dhcp lease time expires, maybe something gets a new IP > and the plex service doesn't reregister? Are you using dhcp or a static IP? > What's your zone config? > > Michael > Sent from my iPhone > > On Nov 27, 2016, at 12:00 PM, Natxo Asenjo wrote: > > hi, > > the subject says it. I installed a centos lx zone following the fine > manual (https://omnios.omniti.com/wiki.php/LXZones) and it works great, > but plex disappears from the network every 2 to three days. Rebooting the > zone solves the problem until it happens again. > > How could I troubleshoot this issue? The zone os is available (I can login > using ssh, so networking works, only plex does not broadcast). > > TIA > > -- > Groeten, > natxo > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > -- -- Groeten, natxo -------------- next part -------------- An HTML attachment was scrubbed... URL: From skeltonr at btconnect.com Mon Nov 28 08:54:51 2016 From: skeltonr at btconnect.com (Richard Skelton) Date: Mon, 28 Nov 2016 08:54:51 +0000 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> References: <58389551.7020305@btconnect.com> <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> Message-ID: <583BF0DB.7030509@btconnect.com> Hi Dan, Sorry yes /var/adm/messages is filling up with ;- Nov 28 08:35:17 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Failed to initialize adapter Nov 28 08:35:17 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 4 Nov 28 08:35:17 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 5 Nov 28 08:35:17 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check Nov 28 08:35:17 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Invalid mac address The initial boot has ;- # grep ixgbe /var/adm/messages.0 Nov 25 11:08:34 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x40 vector 0x60 ioapic 0xff intin 0xff is bound to cpu 21 Nov 25 11:08:34 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 22 Nov 25 11:08:34 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check Nov 25 11:08:35 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe0 registered Nov 25 11:08:35 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe0: Intel 10Gb Ethernet Nov 25 11:08:35 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x4a vector 0x66 ioapic 0xff intin 0xff is bound to cpu 9 Nov 25 11:08:35 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x4b vector 0x67 ioapic 0xff intin 0xff is bound to cpu 10 Nov 25 11:08:35 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check Nov 25 11:08:35 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe1 registered Nov 25 11:08:35 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe1: Intel 10Gb Ethernet Nov 25 11:08:35 myhost mac: [ID 435574 kern.info] NOTICE: ixgbe0 link up, 10000 Mbps, full duplex Nov 25 11:09:07 myhost mac: [ID 736570 kern.info] NOTICE: ixgbe0 unregistered Nov 25 11:09:10 myhost mac: [ID 736570 kern.info] NOTICE: ixgbe1 unregistered Nov 25 11:12:33 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 8 Nov 25 11:12:33 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 9 Nov 25 11:12:33 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check Nov 25 11:12:33 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Invalid mac address Nov 25 11:12:36 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Failed to initialize adapter Nov 25 11:12:36 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 10 Nov 25 11:12:36 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 11 Nov 25 11:12:36 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check Nov 25 11:12:36 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Invalid mac address Nov 25 11:12:39 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Failed to initialize adapter Nov 25 11:13:03 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 12 Nov 25 11:13:03 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 13 Nov 25 11:13:03 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check Nov 25 11:13:03 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Invalid mac address Dan McDonald wrote: > Check for any "ixgbe" output in /var/adm/messages* on your system as well, in case the driver was complaining about something. > > It's possible the newer ixgbe is detecting something wrong with your board now. ISTR some additional checks added. > > Dan > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bfriesen at simple.dallas.tx.us Mon Nov 28 14:45:42 2016 From: bfriesen at simple.dallas.tx.us (Bob Friesenhahn) Date: Mon, 28 Nov 2016 08:45:42 -0600 (CST) Subject: [OmniOS-discuss] l2arc cache size 16.0E In-Reply-To: References: <7CDEC300-AA33-4931-9CBF-00D4F96BBD80@omniti.com> Message-ID: I have seen a similar complaint on the FreeBSD filesystem list recently. Bob -- Bob Friesenhahn bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ From rafapi at gmail.com Mon Nov 28 14:51:18 2016 From: rafapi at gmail.com (Rafael Pardinas) Date: Mon, 28 Nov 2016 14:51:18 +0000 Subject: [OmniOS-discuss] plex on lx zone (centos 6.8) stops working after 2/3 days In-Reply-To: References: <5F8422A7-850F-4008-A4AF-FC9B1A5E4136@lji.org> Message-ID: Perhaps you should add a network mask to avoid network inconsistencies. I did it by just adding the IP using the CIDR notation, in your case it should be: add property (name=ips,value="192.168.0.120/24") <-- or whatever your subnet definition may be. Also, is your /etc/resolv.conf using the correct DNS server IP? I was under the impression that it had to added as a separate attribute in the zone config. On Mon, 28 Nov 2016 at 05:55 Natxo Asenjo wrote: > hi, > > this is the zone export info: > > create -b > set zonepath=/tank/zones/plex > set brand=lx > set autoboot=true > set ip-type=exclusive > add net > set physical=vnic2 > add property (name=gateway,value="192.168.0.254") > add property (name=ips,value="192.168.0.120") > add property (name=primary,value="true") > add property (name=resolver,value="192.168.0.1") > end > add attr > set name=kernel-version > set type=string > set value=2.6.32 > end > > So that is a fixed ip I tried using dhcp but it would not work. > > I'll double check the dhcp server logs to ensure no double ip addresses > have been leased. > > On Sun, Nov 27, 2016 at 10:04 PM, Michael Talbott > wrote: > > First thought is dhcp lease time expires, maybe something gets a new IP > and the plex service doesn't reregister? Are you using dhcp or a static IP? > What's your zone config? > > Michael > Sent from my iPhone > > On Nov 27, 2016, at 12:00 PM, Natxo Asenjo wrote: > > hi, > > the subject says it. I installed a centos lx zone following the fine > manual (https://omnios.omniti.com/wiki.php/LXZones) and it works great, > but plex disappears from the network every 2 to three days. Rebooting the > zone solves the problem until it happens again. > > How could I troubleshoot this issue? The zone os is available (I can login > using ssh, so networking works, only plex does not broadcast). > > TIA > > -- > Groeten, > natxo > > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss > > > > > -- > -- > Groeten, > natxo > _______________________________________________ > 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 daleg at omniti.com Mon Nov 28 15:03:14 2016 From: daleg at omniti.com (Dale Ghent) Date: Mon, 28 Nov 2016 10:03:14 -0500 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: <583BF0DB.7030509@btconnect.com> References: <58389551.7020305@btconnect.com> <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> <583BF0DB.7030509@btconnect.com> Message-ID: <0B188151-EBAD-46C0-B9A2-E923AE3BDDD1@omniti.com> The "invalid MAC address" error is worrisome, but first things first: Is "allow_unsupported_sfp=1" in /kernel/dev/ixgbe.conf ? If not, add that line and reboot. /dale > On Nov 28, 2016, at 3:54 AM, Richard Skelton wrote: > > Hi Dan, > Sorry yes /var/adm/messages is filling up with ;- > Nov 28 08:35:17 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Failed to initialize adapter > Nov 28 08:35:17 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 4 > Nov 28 08:35:17 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 5 > Nov 28 08:35:17 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check > Nov 28 08:35:17 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Invalid mac address > > > > The initial boot has ;- > # grep ixgbe /var/adm/messages.0 > > Nov 25 11:08:34 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x40 vector 0x60 ioapic 0xff intin 0xff is bound to cpu 21 > Nov 25 11:08:34 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 22 > Nov 25 11:08:34 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check > Nov 25 11:08:35 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe0 registered > Nov 25 11:08:35 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe0: Intel 10Gb Ethernet > Nov 25 11:08:35 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x4a vector 0x66 ioapic 0xff intin 0xff is bound to cpu 9 > Nov 25 11:08:35 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x4b vector 0x67 ioapic 0xff intin 0xff is bound to cpu 10 > Nov 25 11:08:35 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check > Nov 25 11:08:35 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe1 registered > Nov 25 11:08:35 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe1: Intel 10Gb Ethernet > Nov 25 11:08:35 myhost mac: [ID 435574 kern.info] NOTICE: ixgbe0 link up, 10000 Mbps, full duplex > Nov 25 11:09:07 myhost mac: [ID 736570 kern.info] NOTICE: ixgbe0 unregistered > Nov 25 11:09:10 myhost mac: [ID 736570 kern.info] NOTICE: ixgbe1 unregistered > Nov 25 11:12:33 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 8 > Nov 25 11:12:33 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 9 > Nov 25 11:12:33 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check > Nov 25 11:12:33 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Invalid mac address > Nov 25 11:12:36 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Failed to initialize adapter > Nov 25 11:12:36 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 10 > Nov 25 11:12:36 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 11 > Nov 25 11:12:36 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check > Nov 25 11:12:36 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Invalid mac address > Nov 25 11:12:39 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Failed to initialize adapter > Nov 25 11:13:03 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 12 > Nov 25 11:13:03 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 13 > Nov 25 11:13:03 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check > Nov 25 11:13:03 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Invalid mac address > > > Dan McDonald wrote: >> Check for any "ixgbe" output in /var/adm/messages* on your system as well, in case the driver was complaining about something. >> >> It's possible the newer ixgbe is detecting something wrong with your board now. ISTR some additional checks added. >> >> Dan >> >> >> >> > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From bardishe at gmail.com Mon Nov 28 15:32:16 2016 From: bardishe at gmail.com (Artyom Zhandarovsky) Date: Tue, 29 Nov 2016 02:32:16 +1100 Subject: [OmniOS-discuss] l2arc cache size 16.0E In-Reply-To: References: <7CDEC300-AA33-4931-9CBF-00D4F96BBD80@omniti.com> Message-ID: I have no issues on Solaris 11.3 2016-11-29 1:45 GMT+11:00 Bob Friesenhahn : > I have seen a similar complaint on the FreeBSD filesystem list recently. > > Bob > -- > Bob Friesenhahn > bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ > GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Mon Nov 28 15:40:05 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 28 Nov 2016 10:40:05 -0500 Subject: [OmniOS-discuss] l2arc cache size 16.0E In-Reply-To: References: <7CDEC300-AA33-4931-9CBF-00D4F96BBD80@omniti.com> Message-ID: <5164E9A1-FE40-4ECE-8F73-FC2F5233A7E9@omniti.com> > On Nov 28, 2016, at 10:32 AM, Artyom Zhandarovsky wrote: > > I have no issues on Solaris 11.3 Oracle Solaris != illumos Oracle Solaris ZFS != OpenZFS Please understand that. Having said that, if the problem exists on FreeBSD as well, it's highly likely an OpenZFS problem. Dan From danmcd at omniti.com Mon Nov 28 15:40:10 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 28 Nov 2016 10:40:10 -0500 Subject: [OmniOS-discuss] unable to add second nic on KVM In-Reply-To: <642d26ab-6da6-e94b-0a71-e71423dca908@exitas.be> References: <71c6fa3e-aede-2dd8-30e3-53e008c4aa58@exitas.be> <642d26ab-6da6-e94b-0a71-e71423dca908@exitas.be> Message-ID: <9DE482A4-ECC7-4D44-BE15-1362AE2FDCE4@omniti.com> > On Nov 25, 2016, at 7:49 AM, Willems Dirk wrote: > >> Hello, >> >> First of all Thank you for this wonderful OS. >> >> I'm using it for Home virtualization and have the same server also running on my Work for testing purpose and want to use it for Production servers by my clients. >> >> But I've a small problem. >> I'm trying to add a second nic to the KVM image but when I start the KVM I can't connect anymore over ssh or ping the global zone and have to kill the KVM process. I think I just reproduced this on an older OmniOS (r151014). I may need to consult Joyent, as they maintain the KVM code, but also have other things in their MAC/GLDv3 layer that upstream illumos doesn't, which may be interacting poorly. Sorry I can't help immediately, Dan From bardishe at gmail.com Mon Nov 28 15:43:50 2016 From: bardishe at gmail.com (Artyom Zhandarovsky) Date: Tue, 29 Nov 2016 02:43:50 +1100 Subject: [OmniOS-discuss] l2arc cache size 16.0E In-Reply-To: <5164E9A1-FE40-4ECE-8F73-FC2F5233A7E9@omniti.com> References: <7CDEC300-AA33-4931-9CBF-00D4F96BBD80@omniti.com> <5164E9A1-FE40-4ECE-8F73-FC2F5233A7E9@omniti.com> Message-ID: I`am using OmniOs because of napp-it -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Mon Nov 28 16:32:00 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 28 Nov 2016 11:32:00 -0500 Subject: [OmniOS-discuss] New OmniOS bloody Message-ID: <93937E8C-746C-42F7-BF5E-5F9585098E74@omniti.com> I've updated the repo, and the release media for OmniOS bloody. The highlights are: * Updates from upstream and from LX-via-illumos-joyent. * Python2.7 is now an optional package, and a required part of the illumos-tools consolidation. This is in anticipation of a full on switch of OmniOS native Python from 2.6 to 2.7. NOTE that /usr/bin/python still points to 2.6, and will continue to do so until such time as we remove Python2.6 support. * BSD's "bmake" is available. Some community members are working with bmake on illumos and/or OmniOS, and I'm honoring their request. * VIM is up to patch level 95. Dan From skeltonr at btconnect.com Mon Nov 28 20:43:04 2016 From: skeltonr at btconnect.com (Richard Skelton) Date: Mon, 28 Nov 2016 20:43:04 +0000 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: <0B188151-EBAD-46C0-B9A2-E923AE3BDDD1@omniti.com> References: <58389551.7020305@btconnect.com> <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> <583BF0DB.7030509@btconnect.com> <0B188151-EBAD-46C0-B9A2-E923AE3BDDD1@omniti.com> Message-ID: <583C96D8.3040808@btconnect.com> hi Dale. I did :- root at myhost:/intel2# echo "allow_unsupported_sfp=1" >> /kernel/drv/ixgbe.conf root at myhost:/intel2# tail /kernel/drv/ixgbe.conf # # For example, if you see, # "/pci at 7b,0/pci10de,5d at e/pci8086,a15f at 0" 0 "ixgbe" # "/pci at 7b,0/pci10de,5d at e/pci8086,a15f at 0,1" 1 "ixgbe" # # name = "pciex8086,10c6" parent = "/pci at 0,0/pci10de,\5d at e" unit-address = "0" # flow_control = 1; # name = "pciex8086,10c6" parent = "/pci at 0,0/\pci10de,5d at e" unit-address = "1" # flow_control = 3; allow_unsupported_sfp=1 Then rebooted I now see;- Nov 28 20:23:21 myhost genunix: [ID 107833 kern.warning] WARNING: Unexpected EOF on line 98 of /kernel/drv/ixgbe.conf Nov 28 20:23:37 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x42 vector 0x63 ioapic 0xff intin 0xff is bound to cpu 17 Nov 28 20:23:37 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x43 vector 0x64 ioapic 0xff intin 0xff is bound to cpu 18 Nov 28 20:23:37 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation,or bypass this check Nov 28 20:23:37 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe0 registered Nov 28 20:23:37 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe0: Intel 10Gb Ethernet Nov 28 20:23:38 myhost mac: [ID 435574 kern.info] NOTICE: ixgbe0 link up, 10000 Mbps, full duplex Nov 28 20:24:09 myhost mac: [ID 736570 kern.info] NOTICE: ixgbe0 unregistered root at myhost:/root# dladm show-phys LINK MEDIA STATE SPEED DUPLEX DEVICE bnx2 Ethernet down 0 half bnx2 bnx0 Ethernet up 1000 full bnx0 bnx3 Ethernet down 0 half bnx3 bnx1 Ethernet down 0 unknown bnx1 Dale Ghent wrote: > The "invalid MAC address" error is worrisome, but first things first: > > Is "allow_unsupported_sfp=1" in /kernel/dev/ixgbe.conf ? If not, add that line and reboot. > > /dale > > >> On Nov 28, 2016, at 3:54 AM, Richard Skelton wrote: >> >> Hi Dan, >> Sorry yes /var/adm/messages is filling up with ;- >> Nov 28 08:35:17 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Failed to initialize adapter >> Nov 28 08:35:17 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 4 >> Nov 28 08:35:17 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 5 >> Nov 28 08:35:17 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check >> Nov 28 08:35:17 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Invalid mac address >> >> >> >> The initial boot has ;- >> # grep ixgbe /var/adm/messages.0 >> >> Nov 25 11:08:34 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x40 vector 0x60 ioapic 0xff intin 0xff is bound to cpu 21 >> Nov 25 11:08:34 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 22 >> Nov 25 11:08:34 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check >> Nov 25 11:08:35 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe0 registered >> Nov 25 11:08:35 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe0: Intel 10Gb Ethernet >> Nov 25 11:08:35 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x4a vector 0x66 ioapic 0xff intin 0xff is bound to cpu 9 >> Nov 25 11:08:35 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x4b vector 0x67 ioapic 0xff intin 0xff is bound to cpu 10 >> Nov 25 11:08:35 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check >> Nov 25 11:08:35 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe1 registered >> Nov 25 11:08:35 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe1: Intel 10Gb Ethernet >> Nov 25 11:08:35 myhost mac: [ID 435574 kern.info] NOTICE: ixgbe0 link up, 10000 Mbps, full duplex >> Nov 25 11:09:07 myhost mac: [ID 736570 kern.info] NOTICE: ixgbe0 unregistered >> Nov 25 11:09:10 myhost mac: [ID 736570 kern.info] NOTICE: ixgbe1 unregistered >> Nov 25 11:12:33 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 8 >> Nov 25 11:12:33 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 9 >> Nov 25 11:12:33 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check >> Nov 25 11:12:33 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Invalid mac address >> Nov 25 11:12:36 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Failed to initialize adapter >> Nov 25 11:12:36 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 10 >> Nov 25 11:12:36 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 1 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 11 >> Nov 25 11:12:36 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check >> Nov 25 11:12:36 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Invalid mac address >> Nov 25 11:12:39 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe1: Failed to initialize adapter >> Nov 25 11:13:03 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x41 vector 0x61 ioapic 0xff intin 0xff is bound to cpu 12 >> Nov 25 11:13:03 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x49 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 13 >> Nov 25 11:13:03 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Unsupported SFP+ module detected. Please replace it with a supported SFP+ module per Intel documentation, or bypass this check >> Nov 25 11:13:03 myhost ixgbe: [ID 611667 kern.warning] WARNING: ixgbe0: Invalid mac address >> >> >> Dan McDonald wrote: >> >>> Check for any "ixgbe" output in /var/adm/messages* on your system as well, in case the driver was complaining about something. >>> >>> It's possible the newer ixgbe is detecting something wrong with your board now. ISTR some additional checks added. >>> >>> 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 Mon Nov 28 20:44:35 2016 From: danmcd at omniti.com (Dan McDonald) Date: Mon, 28 Nov 2016 15:44:35 -0500 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: <583C96D8.3040808@btconnect.com> References: <58389551.7020305@btconnect.com> <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> <583BF0DB.7030509@btconnect.com> <0B188151-EBAD-46C0-B9A2-E923AE3BDDD1@omniti.com> <583C96D8.3040808@btconnect.com> Message-ID: Literally separate things with a space, AND end it with a semicolon: allow_unsupported_sfp = 1; Dan From Dirk.Willems at exitas.be Mon Nov 28 21:40:35 2016 From: Dirk.Willems at exitas.be (Willems Dirk) Date: Mon, 28 Nov 2016 21:40:35 +0000 Subject: [OmniOS-discuss] unable to add second nic on KVM In-Reply-To: <9DE482A4-ECC7-4D44-BE15-1362AE2FDCE4@omniti.com> References: <71c6fa3e-aede-2dd8-30e3-53e008c4aa58@exitas.be> <642d26ab-6da6-e94b-0a71-e71423dca908@exitas.be> <9DE482A4-ECC7-4D44-BE15-1362AE2FDCE4@omniti.com> Message-ID: On 2016-11-28 16:40, Dan McDonald wrote: >> On Nov 25, 2016, at 7:49 AM, Willems Dirk wrote: >> >>> Hello, >>> >>> First of all Thank you for this wonderful OS. >>> >>> I'm using it for Home virtualization and have the same server also running on my Work for testing purpose and want to use it for Production servers by my clients. >>> >>> But I've a small problem. >>> I'm trying to add a second nic to the KVM image but when I start the KVM I can't connect anymore over ssh or ping the global zone and have to kill the KVM process. > I think I just reproduced this on an older OmniOS (r151014). > > I may need to consult Joyent, as they maintain the KVM code, but also have other things in their MAC/GLDv3 layer that upstream illumos doesn't, which may be interacting poorly. > > Sorry I can't help immediately, > Dan > Hello Dan, Thank You very much for your feedback and help, I really appreciate it. Kind Regards, Dirk Willems From paladinemishakal at gmail.com Tue Nov 29 04:16:11 2016 From: paladinemishakal at gmail.com (Lawrence Giam) Date: Tue, 29 Nov 2016 12:16:11 +0800 Subject: [OmniOS-discuss] Large infolog_hival file Message-ID: Hi All, I have a Supermicro box running OmniOS R151014 build 170cea2. I am getting a lot of entries in infolog_hival about resource.sysevent.EC_hba.ESC_sas_hba_port_broadcast lawrence at sgsan1n2:/var/fm/fmd$ fmdump -I Nov 29 08:45:33.0893 resource.sysevent.EC_hba.ESC_sas_hba_port_broadcast Nov 29 08:45:33.0893 resource.sysevent.EC_hba.ESC_sas_hba_port_broadcast Nov 29 08:45:33.0893 resource.sysevent.EC_hba.ESC_sas_hba_port_broadcast lawrence at sgsan1n2:/var/fm/fmd$ fmdump -IV Nov 29 2016 03:12:20.907712302 (absent) nvlist version: 0 driver_instance = 0 port_address = w500304800bfc5c02 devfs_path = /pci at 0,0/pci8086,340c at 5/pci15d9,400 at 0 PhyIdentifier = 0x2 event_type = port_broadcast_ses class = resource.sysevent.EC_hba.ESC_sas_hba_port_broadcast version = 0x0 __ttl = 0x1 __tod = 0x583c8194 0x361a972e Nov 29 2016 03:12:20.907718074 (absent) nvlist version: 0 driver_instance = 0 port_address = w500304800bfc5c03 devfs_path = /pci at 0,0/pci8086,340c at 5/pci15d9,400 at 0 PhyIdentifier = 0x3 event_type = port_broadcast_ses class = resource.sysevent.EC_hba.ESC_sas_hba_port_broadcast version = 0x0 __ttl = 0x1 __tod = 0x583c8194 0x361aadba This is a chassic with 2 motherboard sharing a single backbone, each motherboard accessing it's own set of disks and we have a HA solution to auto-mount the disks to the other motherboard should there be a problem with the motherboard. IS it possible to disable them? Thanks & Regards. -------------- next part -------------- An HTML attachment was scrubbed... URL: From lists at marzocchi.net Tue Nov 29 09:47:10 2016 From: lists at marzocchi.net (Olaf Marzocchi) Date: Tue, 29 Nov 2016 10:47:10 +0100 Subject: [OmniOS-discuss] OpenSSH and reverse DNS Message-ID: Dear all, Since I upgraded to OpenSSH I have the following problem with DNS: reverse mapping checking getaddrinfo for hostxxx.retail.telecomitalia.it [_ip_] failed - POSSIBLE BREAK-IN ATTEMPT! The remote SSH server has always been OenSSH, the issue appeared when the client (OmniOS) got updated. I have no access to the DNS records. I already have a dynamic DNS configured, but the reverse one is out of my reach. I found online possible solutions and I described the issue also here without success: http://superuser.com/questions/1149850/how-to-disable-the-message-reverse-mapping-checking-getaddrinfo-for-xxx-failed "UseDNS no" helped me to be able at least to connect, but still I cannot disable the warning. Since I launch daily rsync backups via cron, I get emails every morning without any real security issue (in my case). Before I pipe the output of rsync through grep to brutally delete that specific warning, is there a better solution I could test? Thanks Olaf -------------- next part -------------- An HTML attachment was scrubbed... URL: From lotheac at iki.fi Tue Nov 29 11:30:17 2016 From: lotheac at iki.fi (Lauri Tirkkonen) Date: Tue, 29 Nov 2016 13:30:17 +0200 Subject: [OmniOS-discuss] OpenSSH and reverse DNS In-Reply-To: References: Message-ID: <20161129113017.GA229@kekkonen.niksula.hut.fi> On Tue, Nov 29 2016 10:47:10 +0100, Olaf Marzocchi wrote: > Dear all, > Since I upgraded to OpenSSH I have the following problem with DNS: > reverse mapping checking getaddrinfo for hostxxx.retail.telecomitalia.it [_ip_] failed - POSSIBLE BREAK-IN ATTEMPT! > The remote SSH server has always been OenSSH, the issue appeared when the client (OmniOS) got updated. Strange - why would the server suddenly start caring about DNS checks if the client was updated? > I have no access to the DNS records. I already have a dynamic DNS configured, but the reverse one is out of my reach. > > I found online possible solutions and I described the issue also here without success: http://superuser.com/questions/1149850/how-to-disable-the-message-reverse-mapping-checking-getaddrinfo-for-xxx-failed > > "UseDNS no" helped me to be able at least to connect, but still I cannot disable the warning. Since I launch daily rsync backups via cron, I get emails every morning without any real security issue (in my case). "UseDNS no" sshd option should indeed resolve this -- my reading of both the manual and code is that it prevents sshd from resolving client addresses. If you have console access to the server try 'sshd -T | grep usedns' to see if it actually is using the correct configuration file. I guess that the sshd you're using could also be doing something weird, but in stock 7.3p1 remote_hostname is only called from auth_get_canonical_hostname, which always seems to get options.use_dns from the caller. -- Lauri Tirkkonen | lotheac @ IRCnet From lotheac at iki.fi Tue Nov 29 11:50:56 2016 From: lotheac at iki.fi (Lauri Tirkkonen) Date: Tue, 29 Nov 2016 13:50:56 +0200 Subject: [OmniOS-discuss] OpenSSH and reverse DNS In-Reply-To: <20161129113017.GA229@kekkonen.niksula.hut.fi> References: <20161129113017.GA229@kekkonen.niksula.hut.fi> Message-ID: <20161129115056.GB229@kekkonen.niksula.hut.fi> On Tue, Nov 29 2016 13:30:17 +0200, Lauri Tirkkonen wrote: > On Tue, Nov 29 2016 10:47:10 +0100, Olaf Marzocchi wrote: > > Dear all, > > Since I upgraded to OpenSSH I have the following problem with DNS: > > reverse mapping checking getaddrinfo for hostxxx.retail.telecomitalia.it [_ip_] failed - POSSIBLE BREAK-IN ATTEMPT! Another interesting thing to note is that this particular log message was changed in 7.3: https://github.com/openssh/openssh-portable/commit/e690fe85750e93fca1fb7c7c8587d4130a4f7aba So it actually may be the *client* that is calling get_canonical_hostname (which is strange, because only sshd should be doing that). And so it is: https://github.com/omniti-labs/omnios-build/blob/master/build/openssh/patches/0015-GSS-API-key-exchange-support.patch#L1652 The culprit is thus the GSSAPI patch (which I personally don't even agree with, but oh well). I think the option you need to disable in the client's ssh_config is GSSAPIKeyExchange. -- Lauri Tirkkonen | lotheac @ IRCnet From skeltonr at btconnect.com Tue Nov 29 14:02:45 2016 From: skeltonr at btconnect.com (Richard Skelton) Date: Tue, 29 Nov 2016 14:02:45 +0000 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: References: <58389551.7020305@btconnect.com> <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> <583BF0DB.7030509@btconnect.com> <0B188151-EBAD-46C0-B9A2-E923AE3BDDD1@omniti.com> <583C96D8.3040808@btconnect.com> Message-ID: <583D8A85.6090604@btconnect.com> Hi, After setting the allow_unsupported_sfp = 1; I now see :- Nov 28 20:59:46 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x43 vector 0x64 ioapic 0xff intin 0xff is bound to cpu 6 Nov 28 20:59:46 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x44 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 7 Nov 28 20:59:46 myhost pseudo: [ID 129642 kern.info] pseudo-device: devinfo0 Nov 28 20:59:46 myhost genunix: [ID 936769 kern.info] devinfo0 is /pseudo/devinfo at 0 Nov 28 20:59:46 myhost ixgbe: [ID 398243 kern.notice] NOTICE: WARNING: Intel (R) Network Connections are quality tested using Intel (R) Ethernet Optics. Using untested modules is not supported and may cause unstable operation or damage to the module or the adapter. Intel Corporation is not responsible for any harm caused by using untested modules. Nov 28 20:59:47 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe0 registered Nov 28 20:59:47 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe0: Intel 10Gb Ethernet Nov 28 20:59:49 myhost mac: [ID 435574 kern.info] NOTICE: ixgbe0 link up, 10000 Mbps, full duplex I think the SFP+ are made by Cisco will get the P/N tomorrow What SFP+ P/N should I use? Dan McDonald wrote: > Literally separate things with a space, AND end it with a semicolon: > > allow_unsupported_sfp = 1; > > Dan > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Tue Nov 29 15:37:19 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 29 Nov 2016 10:37:19 -0500 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: <583D8A85.6090604@btconnect.com> References: <58389551.7020305@btconnect.com> <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> <583BF0DB.7030509@btconnect.com> <0B188151-EBAD-46C0-B9A2-E923AE3BDDD1@omniti.com> <583C96D8.3040808@btconnect.com> <583D8A85.6090604@btconnect.com> Message-ID: > On Nov 29, 2016, at 9:02 AM, Richard Skelton wrote: > > What SFP+ P/N should I use? You can still use your cisco one with the unsupported_sfp check off. If it's been working for you, why not? OTOH, I hear (I don't use much SFP) that they can be tricky to debug when problems DO occur. I hope others with more SFP experience can speak up. You may also wish to consult the illumos developer's list to ask this as well. Dan From danmcd at omniti.com Tue Nov 29 15:52:25 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 29 Nov 2016 10:52:25 -0500 Subject: [OmniOS-discuss] OpenSSH and reverse DNS In-Reply-To: <20161129115056.GB229@kekkonen.niksula.hut.fi> References: <20161129113017.GA229@kekkonen.niksula.hut.fi> <20161129115056.GB229@kekkonen.niksula.hut.fi> Message-ID: <95442F15-412D-432B-A184-DC2C1D2F3358@omniti.com> > On Nov 29, 2016, at 6:50 AM, Lauri Tirkkonen wrote: > > On Tue, Nov 29 2016 13:30:17 +0200, Lauri Tirkkonen wrote: >> On Tue, Nov 29 2016 10:47:10 +0100, Olaf Marzocchi wrote: >>> Dear all, >>> Since I upgraded to OpenSSH I have the following problem with DNS: >>> reverse mapping checking getaddrinfo for hostxxx.retail.telecomitalia.it [_ip_] failed - POSSIBLE BREAK-IN ATTEMPT! > > Another interesting thing to note is that this particular log message > was changed in 7.3: > https://github.com/openssh/openssh-portable/commit/e690fe85750e93fca1fb7c7c8587d4130a4f7aba > > So it actually may be the *client* that is calling > get_canonical_hostname (which is strange, because only sshd should be > doing that). And so it is: > https://github.com/omniti-labs/omnios-build/blob/master/build/openssh/patches/0015-GSS-API-key-exchange-support.patch#L1652 > > The culprit is thus the GSSAPI patch (which I personally don't even > agree with, but oh well). I think the option you need to disable in the > client's ssh_config is GSSAPIKeyExchange. That GSSAPI patch is there for people who needed it with SunSSH. If the patch is problematic, it should be fixed. The illumos community is a good place to discuss this, even if the patch isn't part of illumos per se. Dan From mir at miras.org Tue Nov 29 16:02:36 2016 From: mir at miras.org (Michael Rasmussen) Date: Tue, 29 Nov 2016 17:02:36 +0100 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: References: <58389551.7020305@btconnect.com> <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> <583BF0DB.7030509@btconnect.com> <0B188151-EBAD-46C0-B9A2-E923AE3BDDD1@omniti.com> <583C96D8.3040808@btconnect.com> <583D8A85.6090604@btconnect.com> Message-ID: <20161129170236.416d6bb6@sleipner.datanom.net> On Tue, 29 Nov 2016 10:37:19 -0500 Dan McDonald wrote: > > You can still use your cisco one with the unsupported_sfp check off. If it's been working for you, why not? > > OTOH, I hear (I don't use much SFP) that they can be tricky to debug when problems DO occur. I hope others with more SFP experience can speak up. You may also wish to consult the illumos developer's list to ask this as well. > At work for our Cisco switches we also use noname SFP+'s since one Cisco cost as much as three nonames. On the other hand to be able to request support we have a number of Cisco SFP+'s to replace in the switch we want support for - otherwise the default reply from Cisco is to use Cisco SFP+'s;-) A general observation is that there is no significant difference in performance or lifetime. -- 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: Can you MAIL a BEAN CAKE? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 195 bytes Desc: OpenPGP digital signature URL: From john.barfield at bissinc.com Wed Nov 30 00:51:24 2016 From: john.barfield at bissinc.com (John Barfield) Date: Wed, 30 Nov 2016 00:51:24 +0000 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x Message-ID: Good evening, I?m trying to upgrade an omnios 151012 and I keep getting errors from pkg src: WARNING: The boot environment being modified is not the active one. Changes made will not be reflected on the next boot. If I try to activate the ?omnios? be it says: Unable to activate omnios. Missing boot menu file. If I look at the bootadm list-menu command it shows: -------------- next part -------------- An HTML attachment was scrubbed... URL: From danmcd at omniti.com Wed Nov 30 01:00:30 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 29 Nov 2016 20:00:30 -0500 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x In-Reply-To: References: Message-ID: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> > On Nov 29, 2016, at 7:51 PM, John Barfield wrote: > > Good evening, > > I?m trying to upgrade an omnios 151012 and I keep getting errors from pkg src: > > WARNING: The boot environment being modified is not the active one. Changes > made will not be reflected on the next boot. > > If I try to activate the ?omnios? be it says: > Unable to activate omnios. > Missing boot menu file. > > If I look at the bootadm list-menu command it shows: > pkgsrc? Or pkg(1)? You may wish to try "installgrub" again (I'd read its man page first) to make sure what's there. Also, what does "beadm list" say? To what release are you trying to upgrade? Dan From john.barfield at bissinc.com Wed Nov 30 01:06:15 2016 From: john.barfield at bissinc.com (John Barfield) Date: Wed, 30 Nov 2016 01:06:15 +0000 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x In-Reply-To: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> References: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> Message-ID: <9D220EB5-9DA3-4F01-8802-ED4ACA98FBED@bissinc.com> Hi Dan, I hit a bad key combination and outlook sent me email prematurely. I?d like to update to the latest stable release. The old nfs lockd bug is on my last nerve. This box doesn?t have any active production zones any longer its only acting as SAN/NAS. Thanks for your help and quick response. <---------------? Beadm list BE Active Mountpoint Space Policy Created omnios N / 4.65G static 2015-03-18 12:08 omnios-backup-1 - - 56.0K static 2015-03-18 12:11 omniosvar - - 25.0K static 2015-03-18 12:08 beadm list -H omnios;3c9ed32a-97d6-edef-a5db-940da597c20a;N;/;4989154816;static;1426698520 omnios-backup-1;8e73df93-0162-e502-8a52-f3eb2f87f72f;;;57344;static;1426698697 omniosvar;(null);;;25600;static;1426698520 bootadm list-menu the location for the active GRUB menu is: /tmp/bootadm_mnt_zfs_legacy.1603/boot/grub/menu.lst default 0 timeout 30 0 OmniOS v11 r151012 1 omniosvar 2 omnios-backup-1 On 11/29/16, 7:00 PM, "Dan McDonald" wrote: > On Nov 29, 2016, at 7:51 PM, John Barfield wrote: > > Good evening, > > I?m trying to upgrade an omnios 151012 and I keep getting errors from pkg src: > > WARNING: The boot environment being modified is not the active one. Changes > made will not be reflected on the next boot. > > If I try to activate the ?omnios? be it says: > Unable to activate omnios. > Missing boot menu file. > > If I look at the bootadm list-menu command it shows: > pkgsrc? Or pkg(1)? You may wish to try "installgrub" again (I'd read its man page first) to make sure what's there. Also, what does "beadm list" say? To what release are you trying to upgrade? Dan From john.barfield at bissinc.com Wed Nov 30 01:06:29 2016 From: john.barfield at bissinc.com (John Barfield) Date: Wed, 30 Nov 2016 01:06:29 +0000 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x In-Reply-To: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> References: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> Message-ID: pkg On 11/29/16, 7:00 PM, "Dan McDonald" wrote: > On Nov 29, 2016, at 7:51 PM, John Barfield wrote: > > Good evening, > > I?m trying to upgrade an omnios 151012 and I keep getting errors from pkg src: > > WARNING: The boot environment being modified is not the active one. Changes > made will not be reflected on the next boot. > > If I try to activate the ?omnios? be it says: > Unable to activate omnios. > Missing boot menu file. > > If I look at the bootadm list-menu command it shows: > pkgsrc? Or pkg(1)? You may wish to try "installgrub" again (I'd read its man page first) to make sure what's there. Also, what does "beadm list" say? To what release are you trying to upgrade? Dan From danmcd at omniti.com Wed Nov 30 01:11:46 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 29 Nov 2016 20:11:46 -0500 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x In-Reply-To: <9D220EB5-9DA3-4F01-8802-ED4ACA98FBED@bissinc.com> References: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> <9D220EB5-9DA3-4F01-8802-ED4ACA98FBED@bissinc.com> Message-ID: <8730ACC7-BDAE-4590-AA33-DEB58E3EE157@omniti.com> > On Nov 29, 2016, at 8:06 PM, John Barfield wrote: > > Hi Dan, > > I hit a bad key combination and outlook sent me email prematurely. > > I?d like to update to the latest stable release. The old nfs lockd bug is on my last nerve. Oh my... I also just realized something... you're REALLY (r151012) old. We now require signatures for our packages. I'd HIGHLY recommend as an intermediate step updating to r151014 (still the LTS release) first. Here's the how-to guide which includes rewhacking your package policy: https://omnios.omniti.com/wiki.php/Upgrade_to_r151014#PerformingtheUpgradeallothercasesincludinglipkgifyouwish Start there, including that you back up your current be first: beadm create backup-r151012-be THEN you can upgrade to the latest stable -- NOTE you have to switch your r151014 from SunSSH to OpenSSH first: https://omnios.omniti.com/wiki.php/Upgrade_to_r151020 Dan From john.barfield at bissinc.com Wed Nov 30 01:13:59 2016 From: john.barfield at bissinc.com (John Barfield) Date: Wed, 30 Nov 2016 01:13:59 +0000 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x In-Reply-To: <8730ACC7-BDAE-4590-AA33-DEB58E3EE157@omniti.com> References: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> <9D220EB5-9DA3-4F01-8802-ED4ACA98FBED@bissinc.com> <8730ACC7-BDAE-4590-AA33-DEB58E3EE157@omniti.com> Message-ID: <60406939-051E-49CD-B292-0C6DAE9B94BB@bissinc.com> Thanks Dan, One issue?I can?t backup my BE for a like reason. beadm create omnios-backup2 Unable to create omnios-backup2. Missing boot menu file. On 11/29/16, 7:11 PM, "Dan McDonald" wrote: > On Nov 29, 2016, at 8:06 PM, John Barfield wrote: > > Hi Dan, > > I hit a bad key combination and outlook sent me email prematurely. > > I?d like to update to the latest stable release. The old nfs lockd bug is on my last nerve. Oh my... I also just realized something... you're REALLY (r151012) old. We now require signatures for our packages. I'd HIGHLY recommend as an intermediate step updating to r151014 (still the LTS release) first. Here's the how-to guide which includes rewhacking your package policy: https://omnios.omniti.com/wiki.php/Upgrade_to_r151014#PerformingtheUpgradeallothercasesincludinglipkgifyouwish Start there, including that you back up your current be first: beadm create backup-r151012-be THEN you can upgrade to the latest stable -- NOTE you have to switch your r151014 from SunSSH to OpenSSH first: https://omnios.omniti.com/wiki.php/Upgrade_to_r151020 Dan From danmcd at omniti.com Wed Nov 30 01:20:01 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 29 Nov 2016 20:20:01 -0500 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x In-Reply-To: <60406939-051E-49CD-B292-0C6DAE9B94BB@bissinc.com> References: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> <9D220EB5-9DA3-4F01-8802-ED4ACA98FBED@bissinc.com> <8730ACC7-BDAE-4590-AA33-DEB58E3EE157@omniti.com> <60406939-051E-49CD-B292-0C6DAE9B94BB@bissinc.com> Message-ID: <648473C8-17C1-42C0-A54E-4F8CC8038473@omniti.com> Hmmm, try re-writing your rpool's menu.lst: cp /tmp/bootadm_mnt_zfs_legacy.1603/boot/grub/menu.lst /rpool/boot/grub/. See if that helps out. Dan From john.barfield at bissinc.com Wed Nov 30 01:26:41 2016 From: john.barfield at bissinc.com (John Barfield) Date: Wed, 30 Nov 2016 01:26:41 +0000 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x In-Reply-To: <648473C8-17C1-42C0-A54E-4F8CC8038473@omniti.com> References: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> <9D220EB5-9DA3-4F01-8802-ED4ACA98FBED@bissinc.com> <8730ACC7-BDAE-4590-AA33-DEB58E3EE157@omniti.com> <60406939-051E-49CD-B292-0C6DAE9B94BB@bissinc.com> <648473C8-17C1-42C0-A54E-4F8CC8038473@omniti.com> Message-ID: <46F191E9-9810-4377-96F8-37109093C136@bissinc.com> So that?s the final caveat here lol?bootadm says that the menu file is at that path but that file structure doesn?t actually exist. Been racking my brain on this most of the afternoon. ls /tmp total 87 drwxrwxrwt 21 root sys 1536 Nov 29 18:54 . drwxr-xr-x 27 root root 31 Nov 29 18:20 .. drwx------ 2 root root 117 Nov 29 18:06 .be.55aiuc drwx------ 2 root root 117 Nov 29 18:06 .be.65aiuc drwx------ 2 root root 117 Nov 29 18:54 .be.ETaWmd drwxr-xr-x 27 root root 31 Nov 29 18:20 .be.FTaWmd drwx------ 2 root root 117 Nov 29 18:31 .be.Guaacd drwx------ 2 root root 117 Nov 29 18:31 .be.Huaacd drwx------ 2 root root 117 Nov 29 18:04 .be.MMaWrc drwx------ 2 root root 117 Nov 29 18:04 .be.NMaWrc drwx------ 2 root root 117 Nov 29 18:15 .be.YGayKc drwx------ 2 root root 117 Nov 29 18:15 .be.ZGayKc drwx------ 2 root root 117 Nov 29 18:54 .be.hvaGnd drwx------ 2 root root 117 Nov 29 18:54 .be.ivaGnd drwx------ 2 root root 117 Nov 29 18:13 .be.kdaGIc drwx------ 2 root root 117 Nov 29 18:13 .be.ldaGIc drwx------ 2 jbarfield adm 184 Nov 29 18:31 ssh-XXRhayad drwx------ 2 root root 117 Nov 29 18:06 tmp6I1NQ8 drwx------ 2 root root 117 Nov 29 18:04 tmp7ZYdY_ drwx------ 2 root root 117 Nov 29 18:13 tmpDp50Ty drwx------ 2 root root 117 Nov 29 18:31 tmpVgEUzA -rw-rw-r-- 1 zabbix zabbix 1234 Nov 29 19:03 zabbix_agentd.log -rw-rw-r-- 1 zabbix zabbix 3 Nov 29 17:57 zabbix_agentd.pid On 11/29/16, 7:20 PM, "Dan McDonald" wrote: Hmmm, try re-writing your rpool's menu.lst: cp /tmp/bootadm_mnt_zfs_legacy.1603/boot/grub/menu.lst /rpool/boot/grub/. See if that helps out. Dan From danmcd at omniti.com Wed Nov 30 01:28:19 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 29 Nov 2016 20:28:19 -0500 Subject: [OmniOS-discuss] USB3 repo & media updated! Message-ID: <8CFF555B-E3EF-4D81-849C-39BC4E6226C5@omniti.com> Now that a new bloody is out, so is a new bloody+xhci/USB3: The repo is: http://pkg.omniti.com/omnios/USB3/ The install media path is still here, but with refreshed files: http://kebe.com/~danmcd/USB3-media/ md5 (OmniOS-USB3-initial.iso) = 63db82350619b01a7267aee6a3d0f5ce sha1 (OmniOS-USB3-initial.iso) = cd3a136b902fd8e1f13ce94660fa327dfb381a9d md5 (OmniOS-USB3-initial.usb-dd) = 93a29fc8956a9f8951986ab9faf07ce7 sha1 (OmniOS-USB3-initial.usb-dd) = 16dcaff5dbac3efddaa255a01676947c04c2b47a md5 (USB3-miniroot.gz) = 338ff55dcf5cbd26857030a0b659e3a2 sha1 (USB3-miniroot.gz) = 9e0c6f6a3c235b7675c4d266ec0e900ffbb8f509 md5 (USB3.zfs.bz2) = 350bd16efd4bde0c79056e24dfbc4a76 sha1 (USB3.zfs.bz2) = 84fdc84906112e55b921cd65b0901c5b04a35d21 Thanks again to Robert Mustacchi of Joyent for his work on xhci/USB3. He has this to say about the bits added since last time: "Mostly changes are related to support for xHCI 0.96, fixes a class of hangs/panics on newer controllers that have bios interaction, and some ugen fixes. I also have a D script for debugging device conn issues. It's a big one. https://us-east.manta.joyent.com/rmustacc/public/tmp/connect.d Basically unplug the device. Run: dtrace -qs connect.d -o /var/tmp/connect.out Then plug in the device and wait for the message that it failed to appear in /var/adm/messages. Then you can kill the script." If you find more xhci/USB3 bugs, please let at least Robert know, if not the illumos developer's list. Thanks! Dan From danmcd at omniti.com Wed Nov 30 01:30:04 2016 From: danmcd at omniti.com (Dan McDonald) Date: Tue, 29 Nov 2016 20:30:04 -0500 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x In-Reply-To: <46F191E9-9810-4377-96F8-37109093C136@bissinc.com> References: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> <9D220EB5-9DA3-4F01-8802-ED4ACA98FBED@bissinc.com> <8730ACC7-BDAE-4590-AA33-DEB58E3EE157@omniti.com> <60406939-051E-49CD-B292-0C6DAE9B94BB@bissinc.com> <648473C8-17C1-42C0-A54E-4F8CC8038473@omniti.com> <46F191E9-9810-4377-96F8-37109093C136@bissinc.com> Message-ID: You may need to construct your menu.lst yourself, then. Here's a few sample entries. Note the up-front globals (like default entry), followed by specific entries. HOPEFULLY that's all you have to do. Otherwise, you may need to re-run installgrub(1M). Dan =============== # default menu entry to boot default 1 # # menu timeout in second before default OS is booted # set to -1 to wait for user input timeout 30 # # To enable grub serial console to ttya uncomment the following lines # and comment out the splashimage line below # WARNING: do not enable grub serial console when BIOS console serial # redirection is active. # serial --unit=0 --speed=9600 # terminal serial # # Uncomment the following line to enable GRUB splashimage on console splashimage /boot/grub/splash.xpm.gz foreground F7FBFF background 333333 # # To chainload another OS # # title Another OS # root (hd,) # chainloader +1 # # To chainload a Solaris release not based on GRUB: # # title Solaris 9 # root (hd,) # chainloader +1 # makeactive # # To load a Solaris instance based on GRUB: # # title Solaris # bootfs /ROOT/ # kernel /platform/i86pc/kernel/amd64/unix # module /platform/i86pc/amd64/boot_archive # # To override Solaris boot args (see kernel(1M)), console device and # properties set via eeprom(1M) edit the "kernel" line to: # # kernel /platform/i86pc/kernel/amd64/unix -B prop=value,... # # # title r151021-bootstrap bootfs rpool/ROOT/r151021-bootstrap kernel$ /platform/i86pc/kernel/amd64/unix -B $ZFS-BOOTFS module$ /platform/i86pc/amd64/boot_archive #============ End of LIBBE entry ============= title r151021-bootstrap-backup-1 bootfs rpool/ROOT/r151021-bootstrap-backup-1 kernel$ /platform/i86pc/kernel/amd64/unix -B $ZFS-BOOTFS module$ /platform/i86pc/amd64/boot_archive #============ End of LIBBE entry ============= From john.barfield at bissinc.com Wed Nov 30 01:56:59 2016 From: john.barfield at bissinc.com (John Barfield) Date: Wed, 30 Nov 2016 01:56:59 +0000 Subject: [OmniOS-discuss] boot environment mounting on /tmp/tmp.x In-Reply-To: References: <13E9878C-23F7-4A66-8344-32A46E051E29@omniti.com> <9D220EB5-9DA3-4F01-8802-ED4ACA98FBED@bissinc.com> <8730ACC7-BDAE-4590-AA33-DEB58E3EE157@omniti.com> <60406939-051E-49CD-B292-0C6DAE9B94BB@bissinc.com> <648473C8-17C1-42C0-A54E-4F8CC8038473@omniti.com> <46F191E9-9810-4377-96F8-37109093C136@bissinc.com> Message-ID: Install grub did not fix the issue. Its booting up with temp BE?s everytime. This issue is really weird. I noticed that the zpool thinks that one of its disks are c6t0d0s0 but its actually using c5t0d0s0 could this be related? Thanks for your help this evening. On 11/29/16, 7:30 PM, "Dan McDonald" wrote: You may need to construct your menu.lst yourself, then. Here's a few sample entries. Note the up-front globals (like default entry), followed by specific entries. HOPEFULLY that's all you have to do. Otherwise, you may need to re-run installgrub(1M). Dan =============== # default menu entry to boot default 1 # # menu timeout in second before default OS is booted # set to -1 to wait for user input timeout 30 # # To enable grub serial console to ttya uncomment the following lines # and comment out the splashimage line below # WARNING: do not enable grub serial console when BIOS console serial # redirection is active. # serial --unit=0 --speed=9600 # terminal serial # # Uncomment the following line to enable GRUB splashimage on console splashimage /boot/grub/splash.xpm.gz foreground F7FBFF background 333333 # # To chainload another OS # # title Another OS # root (hd,) # chainloader +1 # # To chainload a Solaris release not based on GRUB: # # title Solaris 9 # root (hd,) # chainloader +1 # makeactive # # To load a Solaris instance based on GRUB: # # title Solaris # bootfs /ROOT/ # kernel /platform/i86pc/kernel/amd64/unix # module /platform/i86pc/amd64/boot_archive # # To override Solaris boot args (see kernel(1M)), console device and # properties set via eeprom(1M) edit the "kernel" line to: # # kernel /platform/i86pc/kernel/amd64/unix -B prop=value,... # # # title r151021-bootstrap bootfs rpool/ROOT/r151021-bootstrap kernel$ /platform/i86pc/kernel/amd64/unix -B $ZFS-BOOTFS module$ /platform/i86pc/amd64/boot_archive #============ End of LIBBE entry ============= title r151021-bootstrap-backup-1 bootfs rpool/ROOT/r151021-bootstrap-backup-1 kernel$ /platform/i86pc/kernel/amd64/unix -B $ZFS-BOOTFS module$ /platform/i86pc/amd64/boot_archive #============ End of LIBBE entry ============= From Fred_Liu at issi.com Wed Nov 30 11:47:32 2016 From: Fred_Liu at issi.com (Fred Liu) Date: Wed, 30 Nov 2016 11:47:32 +0000 Subject: [OmniOS-discuss] any supoorted in Mellanox HBA in illumos? Message-ID: <689f0365a7954eba8e0afc45c72248b1@CNSH-DAG1.ISSI.com> Hi, Please post if you know. Thanks. Fred From wonko at 4amlunch.net Wed Nov 30 12:03:50 2016 From: wonko at 4amlunch.net (wonko at 4amlunch.net) Date: Wed, 30 Nov 2016 07:03:50 -0500 Subject: [OmniOS-discuss] any supoorted in Mellanox HBA in illumos? In-Reply-To: <689f0365a7954eba8e0afc45c72248b1@CNSH-DAG1.ISSI.com> References: <689f0365a7954eba8e0afc45c72248b1@CNSH-DAG1.ISSI.com> Message-ID: I'm using ConnectX-2 cards in my OmniOS fileserver. Everything works perfectly. Including SRP (SCSI RDMA). I use it with ESXi hosts. -brian > On Nov 30, 2016, at 06:47, Fred Liu wrote: > > Hi, > > Please post if you know. > > > Thanks. > > Fred > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From Fred_Liu at issi.com Wed Nov 30 12:21:31 2016 From: Fred_Liu at issi.com (Fred Liu) Date: Wed, 30 Nov 2016 12:21:31 +0000 Subject: [OmniOS-discuss] any supoorted in Mellanox HBA in illumos? In-Reply-To: References: <689f0365a7954eba8e0afc45c72248b1@CNSH-DAG1.ISSI.com> Message-ID: <7fde1b12cf2d4d71b4efa457155c0173@CNSH-DAG1.ISSI.com> This is encouraging! Thanks. Fred > -----Original Message----- > From: wonko at 4amlunch.net [mailto:wonko at 4amlunch.net] > Sent: ???, ??? 30, 2016 20:04 > To: Fred Liu > Cc: omnios-discuss at lists.omniti.com; smartos-discuss at lists.smartos.org; > Discussion list for OpenIndiana; developer at lists.illumos.org > Subject: Re: [OmniOS-discuss] any supoorted in Mellanox HBA in illumos? > > I'm using ConnectX-2 cards in my OmniOS fileserver. Everything works perfectly. > Including SRP (SCSI RDMA). I use it with ESXi hosts. > > -brian > > > On Nov 30, 2016, at 06:47, Fred Liu wrote: > > > > Hi, > > > > Please post if you know. > > > > > > Thanks. > > > > Fred > > _______________________________________________ > > OmniOS-discuss mailing list > > OmniOS-discuss at lists.omniti.com > > http://lists.omniti.com/mailman/listinfo/omnios-discuss From glush at jet.msk.su Wed Nov 30 14:15:14 2016 From: glush at jet.msk.su (=?utf-8?B?0JTQvNC40YLRgNC40Lkg0JPQu9GD0YjQtdC90L7Qug==?=) Date: Wed, 30 Nov 2016 17:15:14 +0300 Subject: [OmniOS-discuss] Host is not being rebooted if it uses ZFS over iSCSI Message-ID: Hi! Does anybody uses ZFS over iSCSI? There is problem with reboots as iscsi-initiator service does not taking care of ZFS while shutting down. It leads to zpool goes into UNAVAIL state and then first sync() issued gets blocked with following stack: stack pointer for thread ffffff03d8d968c0: ffffff00101cfd50 [ ffffff00101cfd50 _resume_from_idle+0x112() ] ffffff00101cfd80 swtch+0x141() ffffff00101cfdc0 cv_wait+0x70(ffffff03df9f9582, ffffff03df9f9540) ffffff00101cfe10 txg_wait_synced+0x83(ffffff03df9f93c0, 0) ffffff00101cfe30 spa_sync_allpools+0x81() ffffff00101cfe70 zfs_sync+0xc5(0, 0, ffffff03d5457cf8) ffffff00101cfeb0 vfs_sync+0x9c(0) ffffff00101cfec0 syssync+0xb() ffffff00101cff10 sys_syscall32+0xff() All subsequent sync() calls are returning immediately. The problem is first sync() is being made by svc.startd just before killing all remaining processes and calling uadmin() (http://src.illumos.org/source/xref/illumos-gate/usr/src/cmd/svc/startd/graph.c#3615). So, uadmin() gets never called and system not being rebooted. I have not came up with any acceptable modifications of /lib/svc/method/iscsi-initiator to workaround the problem. Any suggestions? -- Dmitry Glushenok Jet Infosystems From cks at cs.toronto.edu Wed Nov 30 15:05:22 2016 From: cks at cs.toronto.edu (Chris Siebenmann) Date: Wed, 30 Nov 2016 10:05:22 -0500 Subject: [OmniOS-discuss] Host is not being rebooted if it uses ZFS over iSCSI In-Reply-To: Your message of Wed, 30 Nov 2016 17:15:14 +0300. Message-ID: <20161130150522.43BE25A0044@apps1.cs.toronto.edu> > Does anybody uses ZFS over iSCSI? > > There is problem with reboots as iscsi-initiator service does not > taking care of ZFS while shutting down. It leads to zpool goes into > UNAVAIL state and then first sync() issued gets blocked with following > stack: We have a significant ZFS-over-iSCSI fileserver environment and haven't seen this, but we may be shutting our systems down in a different way than you are. We're on OmniOS r151014 and we haven't seen this behavior when we reboot or shut down our systems via 'reboot' or 'poweroff'. For us it seems to go the other way around; running 'reboot' won't completely shut down the iSCSI initiator before the machine reboots, apparently because it's being held busy by our active ZFS pools. It may be relevant that we're using OmniOS multipathing across two IPs for each target (configured statically); this might change the device stack inside the kernel. I don't think we've ever tried 'shutdown' itself on a live production system, so I don't know how that would work. - cks From glush at jet.msk.su Wed Nov 30 16:45:53 2016 From: glush at jet.msk.su (=?utf-8?B?0JTQvNC40YLRgNC40Lkg0JPQu9GD0YjQtdC90L7Qug==?=) Date: Wed, 30 Nov 2016 19:45:53 +0300 Subject: [OmniOS-discuss] Host is not being rebooted if it uses ZFS over iSCSI In-Reply-To: <20161130150522.43BE25A0044@apps1.cs.toronto.edu> References: <20161130150522.43BE25A0044@apps1.cs.toronto.edu> Message-ID: <326FA912-A3C6-4CC0-8D1A-D62DE81395E0@jet.msk.su> The 'reboot' works, but it kills all processes at once (including iscsid), leaving no chances for ZFS to flush write cache. And I suspect that with current iscsi/initiator SMF service implementation the same happens (but for some reason sync(), which is being called by 'reboot' after killing all processes, is not hangs). > 30 ????. 2016 ?., ? 18:05, Chris Siebenmann ???????(?): > >> Does anybody uses ZFS over iSCSI? >> >> There is problem with reboots as iscsi-initiator service does not >> taking care of ZFS while shutting down. It leads to zpool goes into >> UNAVAIL state and then first sync() issued gets blocked with following >> stack: > > We have a significant ZFS-over-iSCSI fileserver environment and > haven't seen this, but we may be shutting our systems down in a > different way than you are. > > We're on OmniOS r151014 and we haven't seen this behavior when we > reboot or shut down our systems via 'reboot' or 'poweroff'. For us it > seems to go the other way around; running 'reboot' won't completely shut > down the iSCSI initiator before the machine reboots, apparently because > it's being held busy by our active ZFS pools. > > It may be relevant that we're using OmniOS multipathing across two > IPs for each target (configured statically); this might change the > device stack inside the kernel. > > I don't think we've ever tried 'shutdown' itself on a live production > system, so I don't know how that would work. > > - cks -- Dmitry Glushenok Jet Infosystems -------------- next part -------------- An HTML attachment was scrubbed... URL: From skeltonr at btconnect.com Wed Nov 30 18:38:15 2016 From: skeltonr at btconnect.com (Richard Skelton) Date: Wed, 30 Nov 2016 18:38:15 +0000 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: <583D8A85.6090604@btconnect.com> References: <58389551.7020305@btconnect.com> <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> <583BF0DB.7030509@btconnect.com> <0B188151-EBAD-46C0-B9A2-E923AE3BDDD1@omniti.com> <583C96D8.3040808@btconnect.com> <583D8A85.6090604@btconnect.com> Message-ID: <583F1C97.10303@btconnect.com> Hi, The SFP+ is a Cisco P/N 10-3105-01 is just not supported because it's not an Intel part :-) Richard Skelton wrote: > Hi, > After setting the allow_unsupported_sfp = 1; > I now see :- > > Nov 28 20:59:46 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: > pciex8086,10fb (ixgbe) instance 0 irq 0x43 vector 0x64 ioapic 0xff > intin 0xff is bound to cpu 6 > Nov 28 20:59:46 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: > pciex8086,10fb (ixgbe) instance 0 irq 0x44 vector 0x65 ioapic 0xff > intin 0xff is bound to cpu 7 > Nov 28 20:59:46 myhost pseudo: [ID 129642 kern.info] pseudo-device: > devinfo0 > Nov 28 20:59:46 myhost genunix: [ID 936769 kern.info] devinfo0 is > /pseudo/devinfo at 0 > Nov 28 20:59:46 myhost ixgbe: [ID 398243 kern.notice] NOTICE: WARNING: > Intel (R) Network Connections are quality tested using Intel (R) > Ethernet Optics. Using untested modules is not supported and may cause > unstable operation or damage to the module or the adapter. Intel > Corporation is not responsible for any harm caused by using untested > modules. > Nov 28 20:59:47 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe0 > registered > Nov 28 20:59:47 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe0: > Intel 10Gb Ethernet > Nov 28 20:59:49 myhost mac: [ID 435574 kern.info] NOTICE: ixgbe0 link > up, 10000 Mbps, full duplex > > I think the SFP+ are made by Cisco will get the P/N tomorrow > > What SFP+ P/N should I use? > > Dan McDonald wrote: >> Literally separate things with a space, AND end it with a semicolon: >> >> allow_unsupported_sfp = 1; >> >> Dan >> >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From daleg at omniti.com Wed Nov 30 18:52:25 2016 From: daleg at omniti.com (Dale Ghent) Date: Wed, 30 Nov 2016 13:52:25 -0500 Subject: [OmniOS-discuss] Intel Corporation 82599ES 10-Gigabit not working since upgrade to r151020 In-Reply-To: <583F1C97.10303@btconnect.com> References: <58389551.7020305@btconnect.com> <0D731D47-9BA4-4460-B5CD-1084E72CDBF9@omniti.com> <583BF0DB.7030509@btconnect.com> <0B188151-EBAD-46C0-B9A2-E923AE3BDDD1@omniti.com> <583C96D8.3040808@btconnect.com> <583D8A85.6090604@btconnect.com> <583F1C97.10303@btconnect.com> Message-ID: <0EBF16D1-0B93-4409-BC66-EFE3869D3999@omniti.com> The Intel ixgbe driver (and igb/e1000g driver for the 1Gb NICs as well) comes in two parts: The ?Intel Common Code? - a library that?s maintained by Intel and provides the core functions for configuring and operating an Intel NIC, and the OS-specific wrapper code that uses it and provides all the OS-specific logic to interface between the ICC and the OS?s own management and network APIs. When it comes to SFP/SFFP copper and optical parts, Intel tests the ICC using a specific set of these from specific vendors. The ICC code interrogates the SFP module for its make and model, and if it?s not on their list of tested parts, you see the warning that you saw. Unless it?s overridden (in our case, by adding the line in ixgbe.conf) the ICC will refuse to progress during the interface attach phase, or during the setup phase if a module was replaced after boot. Intel maintains a list of these makes and models for each card. For example, here?s the list for the X520 (aka 82599) http://www.intel.com/content/www/us/en/support/network-and-i-o/ethernet-products/000005528.html /dale > On Nov 30, 2016, at 1:38 PM, Richard Skelton wrote: > > Hi, > The SFP+ is a Cisco P/N 10-3105-01 is just not supported because it's not an Intel part :-) > > Richard Skelton wrote: >> Hi, >> After setting the allow_unsupported_sfp = 1; >> I now see :- >> Nov 28 20:59:46 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x43 vector 0x64 ioapic 0xff intin 0xff is bound to cpu 6 >> Nov 28 20:59:46 myhost pcplusmp: [ID 805372 kern.info] pcplusmp: pciex8086,10fb (ixgbe) instance 0 irq 0x44 vector 0x65 ioapic 0xff intin 0xff is bound to cpu 7 >> Nov 28 20:59:46 myhost pseudo: [ID 129642 kern.info] pseudo-device: devinfo0 >> Nov 28 20:59:46 myhost genunix: [ID 936769 kern.info] devinfo0 is /pseudo/devinfo at 0 >> Nov 28 20:59:46 myhost ixgbe: [ID 398243 kern.notice] NOTICE: WARNING: Intel (R) Network Connections are quality tested using Intel (R) Ethernet Optics. Using untested modules is not supported and may cause unstable operation or damage to the module or the adapter. Intel Corporation is not responsible for any harm caused by using untested modules. >> Nov 28 20:59:47 myhost mac: [ID 469746 kern.info] NOTICE: ixgbe0 registered >> Nov 28 20:59:47 myhost ixgbe: [ID 611667 kern.info] NOTICE: ixgbe0: Intel 10Gb Ethernet >> Nov 28 20:59:49 myhost mac: [ID 435574 kern.info] NOTICE: ixgbe0 link up, 10000 Mbps, full duplex >> >> I think the SFP+ are made by Cisco will get the P/N tomorrow >> >> What SFP+ P/N should I use? >> >> Dan McDonald wrote: >>> Literally separate things with a space, AND end it with a semicolon: >>> >>> allow_unsupported_sfp = 1; >>> >>> Dan >>> >>> >>> > _______________________________________________ > OmniOS-discuss mailing list > OmniOS-discuss at lists.omniti.com > http://lists.omniti.com/mailman/listinfo/omnios-discuss From jimklimov at cos.ru Wed Nov 30 19:47:34 2016 From: jimklimov at cos.ru (Jim Klimov) Date: Wed, 30 Nov 2016 20:47:34 +0100 Subject: [OmniOS-discuss] Host is not being rebooted if it uses ZFS over iSCSI In-Reply-To: <326FA912-A3C6-4CC0-8D1A-D62DE81395E0@jet.msk.su> References: <20161130150522.43BE25A0044@apps1.cs.toronto.edu> <326FA912-A3C6-4CC0-8D1A-D62DE81395E0@jet.msk.su> Message-ID: <8ADA4920-626E-4313-97A3-B933E96E70CD@cos.ru> 30 ?????? 2016??. 17:45:53 CET, "??????? ????????" ?????: >The 'reboot' works, but it kills all processes at once (including >iscsid), leaving no chances for ZFS to flush write cache. And I suspect >that with current iscsi/initiator SMF service implementation the same >happens (but for some reason sync(), which is being called by 'reboot' >after killing all processes, is not hangs). > >> 30 ????. 2016 ?., ? 18:05, Chris Siebenmann >???????(?): >> >>> Does anybody uses ZFS over iSCSI? >>> >>> There is problem with reboots as iscsi-initiator service does not >>> taking care of ZFS while shutting down. It leads to zpool goes into >>> UNAVAIL state and then first sync() issued gets blocked with >following >>> stack: >> >> We have a significant ZFS-over-iSCSI fileserver environment and >> haven't seen this, but we may be shutting our systems down in a >> different way than you are. >> >> We're on OmniOS r151014 and we haven't seen this behavior when we >> reboot or shut down our systems via 'reboot' or 'poweroff'. For us it >> seems to go the other way around; running 'reboot' won't completely >shut >> down the iSCSI initiator before the machine reboots, apparently >because >> it's being held busy by our active ZFS pools. >> >> It may be relevant that we're using OmniOS multipathing across two >> IPs for each target (configured statically); this might change the >> device stack inside the kernel. >> >> I don't think we've ever tried 'shutdown' itself on a live production >> system, so I don't know how that would work. >> >> - cks > >-- >Dmitry Glushenok >Jet Infosystems > > > >------------------------------------------------------------------------ > >_______________________________________________ >OmniOS-discuss mailing list >OmniOS-discuss at lists.omniti.com >http://lists.omniti.com/mailman/listinfo/omnios-discuss Shameless plug: did you try an approach like https://github.com/jimklimov/illumos-smf-zfspools to define an SMF dependency between a pool whose vdevs reside on an iscsi initiator, or an iscsi target residing on a pool, and perhaps other services like zones to depend on the non-rpool zfs pool (SMFized)? At least for the case of well ordered and synced graceful shutdowns and startups, this should help you. HTH, Jim Klimov -- Typos courtesy of K-9 Mail on my Samsung Android