Stable kernels 2.6.27.22 and 2.6.28.10

[Posted May 3, 2009 by jake]

Stable kernels 2.6.27.22 (note that the announcement email's subject contains an off-by-one error) and 2.6.28.10 have been released. They contain fixes all over the tree (58 and 88 patches respectively); several have CVE numbers associated with them, so users are encouraged to upgrade. Also, "NOTE, this is the LAST update of the 2.6.28 kernel series, so all users are very strongly encouraged to upgrade to the 2.6.29 series at this point in time!" 2.6.27 will continue to be maintained by the stable folks for quite some time to come.

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 2:15 UTC (Mon) by jimmybgood (guest, #26142) [Link] (10 responses)

2.6.11 was the first stable kernel to have updates after release.
2.6.13 was the first stable kernel to have updates after the next stable kernel was released.
2.6.16 holds the records for most updates after release (62) and the longest period of support after release (2 yr 4 mo 1 day).
2.6.27 has had 22 updates over 6 months and 23 days so far.

Can anyone venture a guess as to how long "quite some time to come" is?

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 2:48 UTC (Mon) by dlang (guest, #313) [Link]

probably a couple of years, the people who were maintaining 2.6.16 are shifting their support to 2.6.27. at some point in the future they will see a case where a lot of distros are shipping the same kernel release, and decide that it makes sense to move their efforts to that release instead.

unfortunately plans are not precide enough to guarantee a particular time, but you should expect a couple of years from it.

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 3:21 UTC (Mon) by gregkh (subscriber, #8) [Link] (8 responses)

> Can anyone venture a guess as to how long "quite some time to come" is?

Does it really matter?

If so, why not just ask the people doing this maintenance, instead of asking on some random web site?

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 4:45 UTC (Mon) by jimmybgood (guest, #26142) [Link] (1 responses)

Well, I didn't expect such an existential discussion this evening. So to quote Queen, "nothing really matters to me." In particular, I doubt if I will ever use a kernel as old as the 2.6.27 series. Nevertheless, I continue doing things. If I ever felt I needed a good reason to do them, I would commence with self destruction. There appears to be a heroin dealer down the street. But, not to worry, I don't need a good reason, nor do I require things to "matter."

As to communicating directly with kernel developers:

A) I would estimate the odds of getting a response as being vanishingly small. Here, I guessed, I would get two or three reasonable responses. I've received one that seems reasonable already.

B) I don't think I've ever much appreciated the response any developer has given me with regards to any software project that (s)he was working on. Were I to receive any response from someone claiming to actually be working on stable updates for the 2.6.27 kernel, I would dismiss it as facetious, unrealistic wishful thinking, bureaucratic fantasy or otherwise unreliable.

C) I honestly don't want to waste their time. I appreciate their work on projects and encourage them to continue.

I'm under the understanding that guest comments can be filtered out, thus, I feel comfortable posting comments that reflect my idle curiosity and cynicism. Imagining myself as a subscriber, I would be too paralyzed with uncertainty to post responsibly so probably wouldn't.

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 14:47 UTC (Mon) by gregkh (subscriber, #8) [Link]

> As to communicating directly with kernel developers:
>
> A) I would estimate the odds of getting a response as being vanishingly
> small. Here, I guessed, I would get two or three reasonable responses.
> I've received one that seems reasonable already.

Heh, you do realize you just got one from the maintainer of the 2.6.27-stable
tree, right? :)

> B) I don't think I've ever much appreciated the response any developer has
> given me with regards to any software project that (s)he was working on. Were
> I to receive any response from someone claiming to actually be working on
> stable updates for the 2.6.27 kernel, I would dismiss it as facetious,
> unrealistic wishful thinking, bureaucratic fantasy or otherwise unreliable.

So if you did receive a response to your question, you would regard it as
"fantasy"? That's very insulting.

> C) I honestly don't want to waste their time. I appreciate their work on
> projects and encourage them to continue.

You should send emails with appreciation, otherwise all we get are complaints
which can be depressing over time...

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 9:42 UTC (Mon) by Alterego (guest, #55989) [Link] (3 responses)

> Does it really matter?
Yes it does when one have to chose a kernel for a compute farm, and need a rock solid debuged kernel, instead of a shiny brand new buggy one, or an too old kernel from one distro.

Why not asking here? Lwn is a good random choice (http://xkcd.com/221/ random number), people are well informed, kind, and this prevent annoying the maintainers with weird questions :-).

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 16:56 UTC (Mon) by ahoogerhuis (guest, #4041) [Link] (2 responses)

We have about a 100 VMs and have been running rock solid on 2.6.16.x (mostly .46) for a good while but had to upgrade to solve a few issues. After some testing it was found that 2.6.25ish untill 2.6.28.something is pretty much dead in the water on ESX 3, and have been running 2.6.28.9 happily.

The bugs were solved by Alok Kataria from VMWare around 2.6.28.6, I think. I can't find the specifics in the short logs for 2.6.28.x just right now, but if those have been backported to 2.6.27.x and it will be maintained for some time, then I think we'll look into staying on 2.6.27.x for an extended period.

-A

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 17:01 UTC (Mon) by gregkh (subscriber, #8) [Link] (1 responses)

> The bugs were solved by Alok Kataria from VMWare around 2.6.28.6, I think. I
> can't find the specifics in the short logs for 2.6.28.x just right now, but if
> those have been backported to 2.6.27.x and it will be maintained for some
> time, then I think we'll look into staying on 2.6.27.x for an extended period.

I think those changes are now all in the 2.6.27-stable series.

If not, please let me know (through the stable@kernel.org email address), and
we will work to resolve them.

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 17:27 UTC (Mon) by ahoogerhuis (guest, #4041) [Link]

I'll fire up a 2CPU VM with 2.6.27.latest and let it churn overnight. If it still lives in the mornig I'll take you word, otherwise I'll get on to it. :)

-A

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 4, 2009 22:34 UTC (Mon) by madscientist (subscriber, #16861) [Link] (1 responses)

David Miller just sent an email saying that Greg just sent an email saying that there won't be any more 2.6.27.x stable releases; is that so? I thought that 2.6.27.x was going to take over from 2.6.16.x as the next long-term supported stable release. Wasn't 2.6.27 only released about 7 months ago or so?

Or does this mean someone else (Adrian?) is taking over the 2.6.27.x stable branch from Greg?

I have confusion in me...!

http://marc.info/?l=linux-netdev&m=124146403006662&...

Stable kernels 2.6.27.22 and 2.6.28.10

Posted May 5, 2009 4:55 UTC (Tue) by gregkh (subscriber, #8) [Link]

See David's follow-up response:
http://marc.info/?l=linux-netdev&m=124147877430893&...