lkml.org 
[lkml]   [1998]   [Jun]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectFreeze Up...

Hi - hope this is the right place to post this [and I hope no-one on
linux-kernel minds if I post this there too]: I just had our webserver,
an x86 box:

processor : 0
cpu : 586
model : Pentium 75+
vendor_id : GenuineIntel
stepping : 6
fdiv_bug : no
hlt_bug : no
f00f_bug : yes
fpu : yes
fpu_exception : yes
cpuid : yes
wp : yes
flags : fpu vme de pse tsc msr mce cx8
bogomips : 40.04

with the following memory:

MemTotal: 47216 kB
MemFree: 2180 kB
MemShared: 35416 kB
Buffers: 12848 kB
Cached: 17824 kB
SwapTotal: 66524 kB
SwapFree: 66412 kB

freeze up on me. I could get it to flip between virtual consoles with
alt-f<n>, but that was about it, not even the vulcan nerve pinch worked.

inspection of the /var/adm/kernel showed the following:

Jun 18 12:20:18 tanya linux: Unable to handle kernel paging request at virtual address c8109b18
Jun 18 12:20:18 tanya linux: current->tss.cr3 = 02dcb000, %cr3 = 02dcb000
Jun 18 12:20:18 tanya linux: *pde = 00000000
Jun 18 12:20:18 tanya linux: Oops: 0000
Jun 18 12:20:18 tanya linux: CPU: 0
Jun 18 12:20:18 tanya linux: EIP: 0010:[<00000004>]
Jun 18 12:20:18 tanya linux: EFLAGS: 00010a07
Jun 18 12:20:18 tanya linux: eax: 002cabb0 ebx: fffffff7 ecx: 0810a058 edx: 00000003
Jun 18 12:20:18 tanya linux: esi: 08109b18 edi: 02145c0c ebp: 01b70600 esp: 02e2dfb0
Jun 18 12:20:18 tanya linux: ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018
Jun 18 12:20:18 tanya linux: Process xload (pid: 11643, process nr: 44, stackpage=02e2d000)
Jun 18 12:20:18 tanya linux: Stack: 00000000 bffff744 0010a642 00000003 0810a058 00000040 08109b18 00000040
Jun 18 12:20:18 tanya linux: bffff768 ffffffda 0000002b 0000002b 0000002b 0000002b 00000004 4001cbc8
Jun 18 12:20:18 tanya linux: 00000023 00000202 bffff764 0000002b
Jun 18 12:20:18 tanya linux: Call Trace: [<0010a642>]
Jun 18 12:20:19 tanya linux: Code: 6f ef 00 f0 c3 e2 00 f0 6f ef 00 f0 6f ef 00 f0 54 ff 00 f0
Jun 18 12:20:19 tanya linux: general protection: 0000

followed by several entries like this:

Jun 18 12:20:19 tanya linux: general protection: 0000
Jun 18 12:20:19 tanya linux: CPU: 0
Jun 18 12:20:19 tanya linux: EIP: 0010:[<001232e4>]
Jun 18 12:20:19 tanya linux: EFLAGS: 00010206
Jun 18 12:20:19 tanya linux: eax: 00c956a4 ebx: fffffff7 ecx: 40007000 edx: 00000001
Jun 18 12:20:19 tanya linux: esi: 000004a9 edi: 029ef810 ebp: 01b70f00 esp: 01903fac
Jun 18 12:20:19 tanya linux: ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018
Jun 18 12:20:19 tanya linux: Process tail (pid: 10085, process nr: 59, stackpage=01903000)
Jun 18 12:20:19 tanya linux: Stack: bfffec4c 00000003 bfffec0c 0010a642 00000001 40007000 000004a9 000004a9
Jun 18 12:20:19 tanya linux: 40007000 bfffeb58 ffffffda 0000002b 0000002b 0000002b 0000002b 00000004
Jun 18 12:20:19 tanya linux: 4001cbc8 00000023 00000206 bfffeb54 0000002b
Jun 18 12:20:19 tanya linux: Call Trace: [<0010a642>]
Jun 18 12:20:19 tanya linux: Code: c3 8d 76 00 8b 80 9c 00 00 00 8b 50 40 85 d2 74 23 6a 00 66

There were more entries of both kinds after this. I can send the full
section of the log to anyone who wants to have a look - basically
processes kept dying like flies until the machine froze up, about 80
seconds after the first error.

I've seen one error like this before, last week, when I was running
2.0.33, [patched against the ip attacks with the patch from Alan Cox ],
which killed a few processes, and made a zombie of kswapd, but left the
machine standing [but with the load apparently stuck at at least 1,
although no process seemed to be using much cpu].

Anybody have any idea what's going on? Is this likely to be hardwarre
related? [eg will replacing the SIMMs help? Is it time to phase out the
server? ]

--
"whip me, beat me, make me administer aix!" -andrew


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

\
 
 \ /
  Last update: 2005-03-22 13:43    [W:0.286 / U:0.516 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site