Stable kernel tree status, January 18, 2010

Stable kernel tree status, January 18, 2010

[Greg KH]

Here's the state of the -stable kernel trees, as of January 18, 2010.

2.6.27-stable

The 2.6.27-stable kernel tree is still living on, as a "long-term"
stable release.  But, I do have to warn users of this tree, the older it
gets, the less viable it becomes.  Not all bugfixes are being backported
to this kernel version due to massive code changes in the over 2 years
since this kernel has been released.  I am doing my best to backport
fixes that I become aware of, and I encourage anyone who does fix any
types of bugs in the main kernel tree to let me know if the change
should be applied to this older kernel version.

I'll probably keep maintaining it for at least 6-8 more months, but
after that, I can not guarantee it's viability.  Note, one other
developer has volunteered to pick up the tree after I am finished with
it, but I can not speak for him at this time.


2.6.31-stable

Today the last 2.6.31-stable kernel was released, all users of this
kernel series are strongly encouraged to switch to the 2.6.32 kernel
series, as there will not be any more updates for this branch in the
future.


2.6.32-stable

I'd like to announce that the 2.6.32-stable tree is also going to be
maintained as a "long-term" stable release, living for 2-3 years, like
the 2.6.27 kernel is.  This is because a number (i.e. more than 2) Linux
distributions are basing their "enterprise" releases on this kernel
version, and it will make their lives easier if I keep it alive.

Note, the viability of me keeping this tree alive for such a length of
time relies on the developers working for those distros to keep me
informed of patches that need to be backported and applied to it.
Without their help, I will have no problem in stopping the maintenance
of the tree.


Submitting patches for stable trees

Again, the easiest way to get your patch into a -stable tree is to
merely add the line:
	Cc: stable <stable@kernel.org>
to the Signed-off-by: area of your patch.  When the patch goes into
Linus's tree, it will be automatically sent to the stable address, and I
will know to apply it to the trees.  If I have any problem applying it
at that time, I will email the author and reviewers of the patch about
it.

If you forgot to add this line to the patch, or you have found a patch
written by someone else that you wish to have applied to the stable
trees, email the git commit id of the patch as it shows up in Linus's
tree to the stable@kernel.org email address.  Any stable correspondence
sent to my personal accounts has the chance of being lost in the
shuffle, so please try to not do that.

If a patch needs to be backported to one of the stable trees because it
does not apply directly, please send the backported patch, along with
the git commit id of the original patch, to the stable@kernel.org
address, with a description of which kernel tree it should be applied
to.


If anyone has any other questions about stable releases, please let me
know.

thanks,

greg k-h

Re: Stable kernel tree status, January 18, 2010

[Willy Tarreau]

On Mon, Jan 18, 2010 at 11:09:39AM -0800, Greg KH wrote:
> Here's the state of the -stable kernel trees, as of January 18, 2010.
> 
> 2.6.27-stable
> 
> The 2.6.27-stable kernel tree is still living on, as a "long-term"
> stable release.  But, I do have to warn users of this tree, the older it
> gets, the less viable it becomes.  Not all bugfixes are being backported
> to this kernel version due to massive code changes in the over 2 years
> since this kernel has been released.  I am doing my best to backport
> fixes that I become aware of, and I encourage anyone who does fix any
> types of bugs in the main kernel tree to let me know if the change
> should be applied to this older kernel version.
> 
> I'll probably keep maintaining it for at least 6-8 more months, but
> after that, I can not guarantee it's viability.  Note, one other
> developer has volunteered to pick up the tree after I am finished with
> it, but I can not speak for him at this time.

I can confirm that I proposed to Greg to take it up to give time to
users to switch to newer releases. BUT it is important to understand
that it will then considerably slow down, basically just like 2.4,
with one release every few months when a major issue has to be fixed.
This is simply because I cannot spend as much time as Greg on that
task. Right now I pick most of the 2.4 fixes from Greg's patches that
I backport, so that should not change much with another kernel. However
Greg's rule above still applies : if a fix cannot easily be backported,
better drop it than break everything else.

For this reason, the kernel will eventually become obsolete, and
everyone should consider migrating to 2.6.32-stable.

BTW, thanks Greg for doing that again with 2.6.32. I hope that your
early announcement will incite more people to adopt it so that many
regressions are quickly spotted and fixed.

Regards,
Willy