> -----Original Message-----
> From: Joey Hess [mailto:[EMAIL PROTECTED]
> Sent: Saturday, March 24, 2007 8:29 AM
> To: Erik Cummings
> Cc: debian-user@lists.debian.org
> Subject: Re: Etch Installer RC1 and RC2 hang "Retrieving file 82 of
82"
>
> Erik Cummings wrote:
> > Finally managed to get the install to hang while logging was
enabled.
> > Don't know if this will be valuable or not...but zipped and
attached.
>
> Sorry, after all the work to get the debconf debug log, I don't need
it;
> everything looks ok there and it's clear from your log that your
kernel
> is very unhappy around when dpkg-preconfigure hangs:
>
> Mar 23 15:05:37 kernel: Call Trace:
> Mar 23 15:05:37 kernel: [<c011c321>] do_sigaction+0xf7/0x11f
> Mar 23 15:05:37 kernel: [<c0120c0d>] worker_thread+0xc1/0x118
> Mar 23 15:05:37 kernel: [<c0111cba>] default_wake_function+0x0/0xc
> Mar 23 15:05:37 kernel: [<c0120b4c>] worker_thread+0x0/0x118
> Mar 23 15:05:37 kernel: [<c0122b64>] kthread+0xaf/0xdb
> Mar 23 15:05:37 kernel: [<c0122ab5>] kthread+0x0/0xdb
> Mar 23 15:05:37 kernel: [<c0101005>] kernel_thread_helper+0x5/0xb
> Mar 23 15:05:37 kernel: xfsdatad/0 S 00000009 0 10690 5
> 5587 10688 (L-TLB)
> Mar 23 15:05:37 kernel: f42d3f8c 00000046 00000001 00000009
> f4112030 2a7037c3 00000051 00000269
> Mar 23 15:05:37 kernel: f4112140 2a702850 00000051 000005d0
> 00000000 f4151cd0 f4151cc0 f4151cc8
> Mar 23 15:05:37 kernel: 00000000 c0120c0d 00000001 00000000
> f4a0eab0 00010000 00000000 00000000
>
> This and the pages of this following suggest a problem with the kernel
and
> your hardware. The mention of xfs is interesting -- are you using xfs?
It
> also seems to dump a call trace for every process that is running at
this
> point. I can't tell why it's doing this, although the kernel output
seems
> to
> have something to do with the disk (and dpkg-preconfigure is one of
the
> highest disk stress points during the install), but the log would make
for
> a
> good bug report against the linux-2.6 package.
>
> --
> see shy jo
- Definitely NOT using XFS.
- Can these traces be part of the output of the PS commands I'm
running to get the ps to kill? These entries don't show in the log
until AFTER I do the kill...
- Is there a set of other switches and/or files to gather to
file the kernel bug (and shortest path to HOW to file that kernel bug?)?
Erik
