Opses

Opses

[Markus Schaber]

Hello,

[I subscribed to the list, so no Cc:s are necessary.]

After some days of uptime, I always get an Oops in the kswapd. After that, there are some other Oopses (mostly "cannot handle Kernel page request", but also some other oopses). But as far as I can see, the kswapd is always the first Oops after the reboot.

This is what ksymoops tells me:

***
multimedia:/var/log# ksymoops <messages 
ksymoops 2.4.3 on i686 2.4.17.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.17/ (default)
     -m /boot/System.map-2.4.17 (default)

Warning: You did not tell me where to find symbol information.  I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc.  ksymoops -h explains the options.

Feb 14 09:08:04 multimedia kernel: c0140a17
Feb 14 09:08:04 multimedia kernel: Oops: 0002
Feb 14 09:08:04 multimedia kernel: CPU:    0
Feb 14 09:08:04 multimedia kernel: EIP:    0010:[prune_icache+103/208]    Not tainted
Feb 14 09:08:04 multimedia kernel: EFLAGS: 00010246
Feb 14 09:08:04 multimedia kernel: eax: fd22c080   ebx: c3129648   ecx: 00000000   edx: cd0207cf
Feb 14 09:08:04 multimedia kernel: esi: c3129640   edi: c3129848   ebp: c1435f64   esp: c1435f4c
Feb 14 09:08:04 multimedia kernel: ds: 0018   es: 0018   ss: 0018
Feb 14 09:08:04 multimedia kernel: Process kswapd (pid: 5, stackpage=c1435000)
Feb 14 09:08:04 multimedia kernel: Stack: 0000000f 000001d0 00000020 00001757 c3129448 c78dfcc8 00000006 c0140a9b 
Feb 14 09:08:04 multimedia kernel:        00001559 c012968d 00000006 000001d0 00000006 000001d0 00000006 000001d0 
Feb 14 09:08:04 multimedia kernel:        c021b5c8 00000000 c021b5c8 c01296dc 00000020 c021b5c8 00000001 c1434000 
Feb 14 09:08:04 multimedia kernel: Call Trace: [shrink_icache_memory+27/64] [shrink_caches+109/128] [try_to_free_pages+60/96] [kswapd_balance_pgdat+67/144] [kswapd_balance+22/48] 
Feb 14 09:08:04 multimedia kernel: Code: 89 50 04 89 02 89 73 f8 89 73 fc 8b 45 f8 89 58 04 89 03 8d 
Using defaults from ksymoops -t elf32-i386 -a i386

Code;  00000000 Before first symbol
00000000 <_EIP>:
Code;  00000000 Before first symbol
   0:   89 50 04                  mov    %edx,0x4(%eax)
Code;  00000002 Before first symbol
   3:   89 02                     mov    %eax,(%edx)
Code;  00000004 Before first symbol
   5:   89 73 f8                  mov    %esi,0xfffffff8(%ebx)
Code;  00000008 Before first symbol
   8:   89 73 fc                  mov    %esi,0xfffffffc(%ebx)
Code;  0000000a Before first symbol
   b:   8b 45 f8                  mov    0xfffffff8(%ebp),%eax
Code;  0000000e Before first symbol
   e:   89 58 04                  mov    %ebx,0x4(%eax)
Code;  00000010 Before first symbol
  11:   89 03                     mov    %eax,(%ebx)
Code;  00000012 Before first symbol
  13:   8d 00                     lea    (%eax),%eax


1 warning issued.  Results may not be reliable.
***

It happened using 2.4.14, 2.4.27 and one or two 2.4 versions I don't remember now. 

Does anybody know what this means?

If you need more diagnostics, tell me, and I'll try my best.

Thanks a lot,
Markus Schaber
-- 
"GPL software is not free - the cost is cooperation"

Re: Opses

[Manuel McLure]

Marcus Schaber said:
> After some days of uptime, I always get an Oops in the kswapd. After that,
there are some other Oopses (mostly "cannot handle Kernel page request", but
also some other oopses). But as far as I can see, the kswapd is always the
first Oops after the reboot.
>
> This is what ksymoops tells me:
>
> ***
> multimedia:/var/log# ksymoops <messages
> ksymoops 2.4.3 on i686 2.4.17.  Options used
>      -V (default)
>      -k /proc/ksyms (default)
>      -l /proc/modules (default)
>      -o /lib/modules/2.4.17/ (default)
>      -m /boot/System.map-2.4.17 (default)
>
> Warning: You did not tell me where to find symbol information.  I will
> assume that the log matches the kernel and modules that are running
> right now and I'll use the default options above for symbol resolution.
> If the current kernel and/or modules do not match the log, you can get
> more accurate output by telling me the kernel version and where to find
> map, modules, ksyms etc.  ksymoops -h explains the options.
>
> Feb 14 09:08:04 multimedia kernel: c0140a17
> Feb 14 09:08:04 multimedia kernel: Oops: 0002
> Feb 14 09:08:04 multimedia kernel: CPU:    0
> Feb 14 09:08:04 multimedia kernel: EIP:    0010:[prune_icache+103/208]
Not tainted
> Feb 14 09:08:04 multimedia kernel: EFLAGS: 00010246
> Feb 14 09:08:04 multimedia kernel: eax: fd22c080   ebx: c3129648   ecx:
00000000   edx: cd0207cf
> Feb 14 09:08:04 multimedia kernel: esi: c3129640   edi: c3129848   ebp:
c1435f64   esp: c1435f4c
> Feb 14 09:08:04 multimedia kernel: ds: 0018   es: 0018   ss: 0018
> Feb 14 09:08:04 multimedia kernel: Process kswapd (pid: 5,
stackpage=c1435000)
> Feb 14 09:08:04 multimedia kernel: Stack: 0000000f 000001d0 00000020
00001757 c3129448 c78dfcc8 00000006 c0140a9b
> Feb 14 09:08:04 multimedia kernel:        00001559 c012968d 00000006
000001d0 00000006 000001d0 00000006 000001d0
> Feb 14 09:08:04 multimedia kernel:        c021b5c8 00000000 c021b5c8
c01296dc 00000020 c021b5c8 00000001 c1434000
> Feb 14 09:08:04 multimedia kernel: Call Trace:
[shrink_icache_memory+27/64] [shrink_caches+109/128]
[try_to_free_pages+60/96] [kswapd_balance_pgdat+67/144]
[kswapd_balance+22/48]
> Feb 14 09:08:04 multimedia kernel: Code: 89 50 04 89 02 89 73 f8 89 73 fc
8b 45 f8 89 58 04 89 03 8d
> Using defaults from ksymoops -t elf32-i386 -a i386
>
> Code;  00000000 Before first symbol
> 00000000 <_EIP>:
> Code;  00000000 Before first symbol
>    0:   89 50 04                  mov    %edx,0x4(%eax)
> Code;  00000002 Before first symbol
>    3:   89 02                     mov    %eax,(%edx)
> Code;  00000004 Before first symbol
>    5:   89 73 f8                  mov    %esi,0xfffffff8(%ebx)
> Code;  00000008 Before first symbol
>    8:   89 73 fc                  mov    %esi,0xfffffffc(%ebx)
> Code;  0000000a Before first symbol
>    b:   8b 45 f8                  mov    0xfffffff8(%ebp),%eax
> Code;  0000000e Before first symbol
>    e:   89 58 04                  mov    %ebx,0x4(%eax)
> Code;  00000010 Before first symbol
>   11:   89 03                     mov    %eax,(%ebx)
> Code;  00000012 Before first symbol
>   13:   8d 00                     lea    (%eax),%eax
>
>
> 1 warning issued.  Results may not be reliable.
> ***
>
> It happened using 2.4.14, 2.4.27 and one or two 2.4 versions I don't
remember now.
>
> Does anybody know what this means?
>
> If you need more diagnostics, tell me, and I'll try my best.


Very interesting - last night I got a very similar Oops on 2.4.17:

Feb 14 04:03:06 leng kernel: Unable to handle kernel NULL pointer
dereference at virtual address 00000004
Feb 14 04:03:06 leng kernel:  printing eip:
Feb 14 04:03:06 leng kernel: c01434e2
Feb 14 04:03:06 leng kernel: *pde = 00000000
Feb 14 04:03:06 leng kernel: Oops: 0000
Feb 14 04:03:06 leng kernel: CPU:    0
Feb 14 04:03:07 leng kernel: EIP:    0010:[prune_icache+50/240]    Not
tainted
Feb 14 04:03:07 leng kernel: EIP:    0010:[<c01434e2>]    Not tainted
Feb 14 04:03:07 leng kernel: EFLAGS: 00010207
Feb 14 04:03:07 leng kernel: eax: 6e657272   ebx: 00000000   ecx: 00000000
edx: c2a03280
Feb 14 04:03:07 leng kernel: esi: c72cee00   edi: 00000000   ebp: c1257f54
esp: c1257f3c
Feb 14 04:03:07 leng kernel: ds: 0018   es: 0018   ss: 0018
Feb 14 04:03:07 leng kernel: Process kswapd (pid: 5, stackpage=c1257000)
Feb 14 04:03:07 leng kernel: Stack: 00000d3b c72cea28 c57a07e8 00000003
000001d0 00000005 00000005 c01435c5
Feb 14 04:03:07 leng kernel:        000003c8 c012b04e 00000005 000001d0
00000005 000001d0 c023ab68 00000005
Feb 14 04:03:08 leng kernel:        000001d0 c023ab68 00000000 c012b0a0
00000003 c023ab68 00000001 c1256000
Feb 14 04:03:08 leng kernel: Call Trace: [shrink_icache_memory+37/64]
[shrink_caches+110/144] [try_to_free_pages+48/80]
[kswapd_balance_pgdat+68/144] [kswapd_balance+22/48]
Feb 14 04:03:08 leng kernel: Call Trace: [<c01435c5>] [<c012b04e>]
[<c012b0a0>] [<c012b134>] [<c012b196>]
Feb 14 04:03:08 leng kernel:    [kswapd+161/192] [kswapd+0/192]
[rest_init+0/48] [kernel_thread+38/48] [kswapd+0/192]
Feb 14 04:03:08 leng kernel:    [<c012b2b1>] [<c012b210>] [<c0105000>]
[<c0105726>] [<c012b210>]
Feb 14 04:03:08 leng kernel:
Feb 14 04:03:08 leng kernel: Code: 8b 7f 04 8d 73 f8 8b 86 08 01 00 00 a9 38
00 00 00 75 61 0b

This was the first of many Oops in my log - after the Oops happened multiple
processes started segfaulting.

--
Manuel A. McLure KE6TAW | ...for in Ulthar, according to an ancient
<manuel@mclure.org>     | and significant law, no man may kill a cat.
<http://www.mclure.org> |             -- H.P. Lovecraft

Re: Opses

[Markus Schaber]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1613 bytes --]

Hello,

On Thu, 14 Feb 2002, Markus Schaber wrote:

> It happened using 2.4.14, 2.4.27 and one or two 2.4 versions I don't
> remember now.

Sorry, it should read 2.4.17, not 2.4.17 :-(

Some of you asked for additional information, so here it is:

I don't remember which older kernels also produced this behaviour, but it
is very likely that 2.4.10 is one of them.

The machine is used as an internet server, running apache, zope, php,
postgresql, ftp-proxy and nfc-chat. It also runs a tivoli storage manager
client (dsmc) to take part in the university backup system, and has a
local X server and sshd for maintainance purpses. There are some other
standard daemons (ntp, cron etc.) running, but the machine does no routing
/ firewalling etc.

cpuinfo tells the following: (Yes, I know it is no good to run a server on
a Duron, but this decision was not mine.)

processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 6
model           : 3
model name      : AMD Duron(tm) Processor
stepping        : 1
cpu MHz         : 700.054
cache size      : 64 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 1
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov
pat pse36 mmx fxsr syscall mmxext 3dnowext 3dnow
bogomips        : 1395.91


I also attached a .bzip2'ed variant of my 2.4.17 .config file.


[I'll be away 'till monday. I have access to the machine and my mail via
ISDN dialin over the weekend, but have to pay time-based.]

Thank you,
Markus Schaber

[-- Attachment #2: bzip2 compressed .config --]
[-- Type: APPLICATION/octet-stream, Size: 4244 bytes --]