Tux still unstable

[Kees Hoekzema]

Hello,

After a bit of testing the latest tux (2.6.18.1) with the tux-3.2.21 I came
again to the conclusion that we cannot run tux on our production servers. I
tried running tux from time to time before, starting when it was still
running on 2.4 etc.
The first testing went fine, it gave me my images, it send everything else
to apache. One user browsing the website is going very well, and it doesn't
bug onder two clients trying to get 1.000.000 images as fast as possible.

So I put the server back in the webpool and had three kernel panics:

1 - I forgot to turn of logging (debugging was off already) and tus crashed
withing two seconds after getting requests 'from the net'

So I rebooted, turned off logging, restarted tux

2 - I decided to see for one second what tux was doing, so I did a "echo 1 >
/proc/sys/net/tux/logging && sleep 1 && echo 0 > /proc/sys/net/tux/logging".
Two seconds after this tux crashed again with a kernel panic and an
unreacable site.

3 - Well, so I rebooted and stayed far away from the logging facility. After
around 3 minutes of serving nonstop images tux decided to crash again with a
kernel panic.

So tux is exit for me, but I probably try it again after some development
has been done.

-kees

Ps, the third kernel panic:
 argus kernel: SMP
 argus kernel: CPU:    0
 argus kernel: EIP is at do_send_abuf+0x343/0x3cd [tux]
 argus kernel: eax: 00000000   ebx: f6f40800   ecx: 00000000   edx: 013f9b00
 argus kernel: esi: f6f40800   edi: ffffff98   ebp: 00000000   esp: f6ff5ac4
 argus kernel: ds: 007b   es: 007b   ss: 0068
 argus kernel: Process tux (pid: 1272, ti=f6ff4000 task=f7cc9ab0
task.ti=f6ff4000)
 argus kernel: Stack: f6b92380 c146e2a0 00000559 00000070 00004040 00000286
f89ca2cc f89ca140
 argus kernel: f6f40800 f6f40800 f6f40828 f89ca140 f8999156 f6f40800
00000000 0000002d
 argus kernel: 0000b8b1 00000000 00000001 f899cb75 f6f40800 00000000
f89ca2f4 c18070dc
 argus kernel: Call Trace:
 argus kernel: Code: 9b f8 e8 89 0f 76 c7 e8 de 95 74 c7 e9 2e ff ff ff a1
a8 ef 43 c0 85 c0 75 57 c6 86 ec 03 00 00 03 c7 46 04 00 00 00 00 8b 46 6c
<c7> 40 20 00 00 00 00 c7 40 24 00 00 00 00 8b 86 34 01 00 00 31
 argus kernel: EIP: [<f89baea6>] do_send_abuf+0x343/0x3cd [tux] SS:ESP
0068:f6ff5ac4
 argus kernel: Oops: 0002 [#2]
 argus kernel: SMP
 argus kernel: CPU:    0
 argus kernel: EIP is at tux_flush_workqueue+0xb0/0x132 [tux]
 argus kernel: eax: 00000000   ebx: ea335000   ecx: e35c1028   edx: ea335028
 argus kernel: esi: f89ca140   edi: 00000006   ebp: f89ca2d0   esp: f6ff550c
 argus kernel: ds: 007b   es: 007b   ss: 0068
 argus kernel: Process tux (pid: 1272, ti=f6ff4000 task=f7cc9ab0
task.ti=f6ff4000)
 argus kernel: Stack: f6bde800 00000000 00000000 00000000 f89ca2cc f89ca2b4
f89ca2b4 f89ca140
 argus kernel: 00000001 f89b5804 f89ca140 c02ba1a7 c18eb4c0 f6c9a380
000001ec 00000010
 argus kernel: 00000000 00000001 f89ca140 f89b80e4 f89c14e0 00000000
00000001 00000000
 argus kernel: Call Trace:
 argus kernel: Code: 00 8b 0a 39 ca 75 02 0f 0b 8b 42 04 89 41 04 89 08 89
12 89 52 04 c6 86 8c 01 00 00 01 fb c7 83 dc 03 00 00 00 00 00 00 8b 43 6c
<c7> 40 20 00 00 00 00 c7 40 24 00 00 00 00 c7 43 04 00 00 00 00
 argus kernel: EIP: [<f899cc9d>] tux_flush_workqueue+0xb0/0x132 [tux] SS:ESP
0068:f6ff550c