[OmniOS-discuss] [NEWSENDER] - Re: The Future of OmniOS

Sat Apr 29 09:05:47 UTC 2017

To whom it may concern

I am new in this forum. Jakob Blond-Sorensen, CTO of Lean-On A/S Denmark

Lean-On is a many year provider of an entire technology stack for optimized Citrix XenApp and Xendesktop use cases including high end Engineering remote desktop.
As part of the Lean-On Application Accelerator Platform ™ - we use Illumos as storage provider to the solution. 40G,extreme ZIL,  512+ Gbyte RAM and 40-60 VDEVS of Enterprise SAS SSD
The combination of Xendesktop and Illumos has brought us many good use cases – and happy customers. We are using OmniOS for High End Enterprise customers. We dispatch with just one single customer more data on a daily basis than a large Danish bank.
Please see https://www.citrixatleanon.dk/

We have customers still using Nexenta and we have several customers using OmniOs. And we have a rather large Cloud platform in our datacenter based on OmniOS as storage provider.

As such - We have a great setup working with HA, OmniOs and Napp-it.

We are very sad to see Robert pulling the plug on OmniOs – and I had hoped that he instead had called for better backup rather than dropping the support

OmniOs is a great concept – and a great platform.. and it’s an anchor used by a lot of the other distris…
Some of these has been kind of  hanging on the back of OmniOS…. Nice but not fair !
Dan/Robert  has been doing a super job over the years – so the real reason must be around funding for Dan & Co and an irritation of others just using rather than participate financially or code wise.

But I am not in Roberts shoes.

Yesterday I was approached by GEA from Napp-it.

He wanted to be sure that I was aware of the discussion… so he forwarded among others the link to the excel sheet for future funding of Illumos in some form – OmniOs or a new fork.
It made me happy to see that several folks out there is seeking as we a way to continue the life of OmniOS. All of you – get your funding heart going !

For me – the obvious way is to have Dan continuing the great work – that has been  going on for the last years. But I am not in his shoes either…and If Dan refuses to do so…who else will take on the job…. Suggestions ?
I do not expect this to be a non paid for service. We need to create funding for the right people to secure the releases.

Have no doubt – we would like to see someone to take over the things that Peter Tribble refers to… so that this great OS can grow and become a continued solid alternative to the proprietary  stuff out there that we do not like.
The faster this process can be executed upon – the better.

Please feel free to contact me – I believe that if just Robert and a few others would throw some $ into the spreadsheet as well… we will see a funding fit for a solid community where some folks are paid for taking care of QA and Packaging. It might even be that Robert would change his mind ☺ - and keep the baby.

Anyway.  I have included a abstract of my mail to Gea of last night

To GEA April 27 - 2017

I have just added the amount that we see doable for continued support – but would like to emphasize that if a strong support can be enabled – then we are willing to increase our commitment.

We are backing up any initiative that will handle a continued package and license like the Current setup including  handling and support  like Dan/Dale  has been doing it for the last several years.
We paid last year 10K for licensed support/Illumos releases +  150-200 $ /hour for support in case we had trouble.
Our budget for the coming year is 2-3K$ / month   + support hours of which we do not use many of. I think we prepaid 10 hours last year – and used maybe only 5 – the remaining unused 5 hours is lost.

I have suggested Dan to continue the OmniOs distri like today or in any other form that you support as well.  For me it is of importance that we have you GEA backing us as well.

We expect that 2018- 2019 will raise the monthly amount to around 4-5K$ /month – maybe more if the setup looks promising.

We only use OmniOs for storage – nothing else.

Our wishes are indeed limited !
Some upgrade on ISCSI and configuration support for MPIO NFS for XenServer.

Well – only using OmniOS for Storage is somewhat a shame but that is where we are at the moment. We are truly XenServer people due to us being CITRIX people…  Windows App people with great skills in the CITRIX XenApp and XenDesktop world and Microsoft in general. !!

So for all sorts of purposes – yes we will back up a continued OmniOS/OI/GEA/Dan/iLLUMOS or what ever people wants to do… as long as you support it… and that a ready to deploy package, QA is in place

I really like what Dan has done – and I would like to stress that I would like to see him in the game going forward. But at the end of the day -  any relative secure setup is fine for us.

We are not specialist in OmniOS – so the “only” thing we can bring into the discussion is funding… limited but steady funding. Since we have no deep skills in ZFS – we rather pay the community for that service as long as our customers can get their ISCSI and NFS with the speed of ZIL and ARC/L2ARC.

And for anyone who wishes to involve us – we are:

Lean-On A/S
Industriparken 4
2750 Ballerup
Denmark
Jakob Blond-Sorensen, CTO
jbs at lean-on.com<mailto:jbs at lean-on.com>
Mobile +45 22106554

Fra: OmniOS-discuss [mailto:omnios-discuss-bounces at lists.omniti.com] På vegne af Peter Tribble
Sendt: 27. april 2017 22:07
Til: Robert Treat
Cc: omnios-discuss
Emne: [NEWSENDER] - Re: [OmniOS-discuss] The Future of OmniOS

On Thu, Apr 27, 2017 at 12:21 AM, Robert Treat <robert at omniti.com<mailto:robert at omniti.com>> wrote:
I hope the answer to the best safest place is OmniOS. To that end, we have no plans to stop running OmniOS at this time, and presuming the project can make the jump to community maintainership, we won't need to develop any.

So, assuming such a jump were to happen, what are we talking about?
(My comments here, btw, are from me as an illumos community member
and maintainer of another distro, rather than as an OmniOS customer.)

Some items:

The OmniOS name/brand
The website/wiki
The illumos-omnios fork (the code)
The github repos (the infrastructure)
The pkg repo servers
What happens to those? The real question here is whether they stay,
with community maintainership, or whether the "new" regime is an
independent fork? I would guess the latter to be the safest course.
That's assuming a world in which OmniOS-whatever remains as a separate
entity, rather than taking the view that you throw it away and replace it with
a stable branch of some other distro. I'm not entirely convinced that would
(a) actually work, and (b) not compromise the other distro.

So that gives you a newly forked project that maintains a distro based
on the current OmniOS code (preserving the effort that the likes of
Dan have put in to get LX running), pulling in updates from illumos
and the other upstreams, with the same sort of release cadence
and update approach (so you know you're going to get security updates)
we have at the moment.
Ignoring the who and how for now, would that be what people actually want?
Or are people looking for a different future?

My personal (illumos/Tribblix) selfishness here is twofold; first, I want
to ensure we keep the current codebase as a going concern (as an
unmaintained museum it's not interesting), and secondly an active
project shows the health of the illumos ecosystem in a better light.

--
-Peter Tribble
http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://omniosce.org/ml-archive/attachments/20170429/09781400/attachment-0001.html>