Builders need to run a kernel >= 2.6.32

[Jarod Wilson]

On Sat, Jul 24, 2010 at 1:03 AM, Jarod Wilson <jarod at wilsonet.com> wrote:
> On Sat, Jul 24, 2010 at 12:50 AM, Jarod Wilson <jarod at wilsonet.com> wrote:
>> On Fri, Jul 23, 2010 at 6:45 PM, Jarod Wilson <jarod at wilsonet.com> wrote:
>>> On Fri, Jul 23, 2010 at 10:14 AM, Jarod Wilson <jarod at wilsonet.com> wrote:
>>>> On Fri, Jul 23, 2010 at 10:00 AM, Hans de Goede <j.w.r.degoede at gmail.com> wrote:
>>> ...
>>>>>>> Just a note to people hosting builders, due to glibc changes
>>>>>>> in rawhide all builders now need run kernels>= 2.6.32, otherwise
>>>>>>> rawhide / development builds will likely fail.
>>> ...
>>>>> AFAIK the current Fedora builders are running EL6 beta2
>>>>
>>>> Yep, Dennis mentioned that to me the other day. He didn't mention
>>>> exactly *why*, I just assumed we were eating our own dogfood... Guess
>>>> I'll start coming up with a plan for updating my builder.
>>>
>>> Shutting my builder vm down, copying the disk image, then starting a
>>> rhel6 upgrade attempt...
>>
>> Wasn't exactly smooth, but the vm is up and running rhel6 now, and its
>> reporting in to the build system... Flinging a build at it now... So
>> far, so good.
>
> Spoke too soon. Fail:
>
> ...
> DEBUG util.py:255:
>      updates-released 137 k
> DEBUG util.py:255:   xz-libs               x86_64
> 4.999.9-0.2.beta.20091007git.fc12
> DEBUG util.py:255:
>      updates-released  89 k
> DEBUG util.py:255:   xz-lzma-compat        x86_64
> 4.999.9-0.2.beta.20091007git.fc12
> DEBUG util.py:255:
>      updates-released  15 k
> DEBUG util.py:255:   zlib                  x86_64 1.2.3-23.fc12
>      fedora            69 k
> DEBUG util.py:255:  Transaction Summary
> DEBUG util.py:255:
> ================================================================================
> DEBUG util.py:255:  Install     102 Package(s)
> DEBUG util.py:255:  Upgrade       0 Package(s)
> DEBUG util.py:255:  Total download size: 81 M
> DEBUG util.py:255:  Installed size: 296 M
> DEBUG util.py:255:  warning: /etc/hosts created as /etc/hosts.rpmnew
> DEBUG util.py:255:  *** glibc detected *** /usr/bin/python: double
> free or corruption (!prev): 0x000000000934f740 ***
> DEBUG util.py:255:  ======= Backtrace: =========
> DEBUG util.py:328:  Child returncode was: -6
> DEBUG backend.py:593:  umount -n
> /var/lib/mock/fedora-12-x86_64-rpmfusion_free-5d5b7fcdebeb364e7600033c101fedcd3a63cd0c/root/proc
> DEBUG util.py:289:  Executing command: umount -n
> /var/lib/mock/fedora-12-x86_64-rpmfusion_free-5d5b7fcdebeb364e7600033c101fedcd3a63cd0c/root/proc
> DEBUG util.py:328:  Child returncode was: 0
> DEBUG backend.py:593:  umount -n
> /var/lib/mock/fedora-12-x86_64-rpmfusion_free-5d5b7fcdebeb364e7600033c101fedcd3a63cd0c/root/sys
> DEBUG util.py:289:  Executing command: umount -n
> /var/lib/mock/fedora-12-x86_64-rpmfusion_free-5d5b7fcdebeb364e7600033c101fedcd3a63cd0c/root/sys
> DEBUG util.py:328:  Child returncode was: 0
> DEBUG backend.py:593:  umount -n
> /var/lib/mock/fedora-12-x86_64-rpmfusion_free-5d5b7fcdebeb364e7600033c101fedcd3a63cd0c/root/dev/pts
> DEBUG util.py:289:  Executing command: umount -n
> /var/lib/mock/fedora-12-x86_64-rpmfusion_free-5d5b7fcdebeb364e7600033c101fedcd3a63cd0c/root/dev/pts
> DEBUG util.py:328:  Child returncode was: 0
> DEBUG backend.py:593:  umount -n
> /var/lib/mock/fedora-12-x86_64-rpmfusion_free-5d5b7fcdebeb364e7600033c101fedcd3a63cd0c/root/dev/shm
> DEBUG util.py:289:  Executing command: umount -n
> /var/lib/mock/fedora-12-x86_64-rpmfusion_free-5d5b7fcdebeb364e7600033c101fedcd3a63cd0c/root/dev/shm
> DEBUG util.py:328:  Child returncode was: 0
> DEBUG util.py:77:  remove tree:
> /var/lib/mock/fedora-12-x86_64-rpmfusion_free-5d5b7fcdebeb364e7600033c101fedcd3a63cd0c
> DEBUG util.py:97:  kill orphans
>
>
> I saw a similar glibc bomb running a yum remove while cleaning up some
> detritus earlier tonight too. Looks like something bugly somewhere in
> the rhel6 stack. Will look into it more tomorrow, its past my bed
> time. :\

Red Hat Bugzilla #608710 and friends. I've applied a patch attached to
that bug (patches an rpm lib) that is thought to possibly fix the
problem and installed the resulting build on my builder, and just
threw a previously failed build at it... Quite a few rhel6 kvm guest
installs are apparently hitting this one, so I expect if this patch
isn't the fix, someone is going to get to the bottom of this
post-haste. :) (Fingers crossed the one-line patch is indeed the fix
though).

-- 
Jarod Wilson
jarod at wilsonet.com