Hi,
I got oops recently; they are in the following.
bdg:~# ksymoops -V ksym.out
ksymoops 2.4.5 on i686 2.5.8. Options used
-V (specified)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.5.8/ (default)
-m /boot/System.map-2.5.8 (default)
Unable to handle kernel paging request at virtual address 19fbcf80
c0145e67
*pde = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<c0145e67>] Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010286
eax: 00000000 ebx: 00000000 ecx: 00004300 edx: c67ee320
esi: 00000000 edi: bfffdcfc ebp: c5a87fbc esp: c5a87f7c
ds: 0018 es: 0018 ss: 0018
Stack: c5a86000 0000000a bfffdcfc c010ca0a 00018003 bfffdbbc 00000000
00000000 00000800 c5a87fb4 c67ee320 fffffff2 00004300 00000000 bfffdbfc
c470f000 bfffdbfc c010708b bfffdcfc 00000004 00000002 0000000a bfffdcfc
bfffdbfc Call Trace: [<c010ca0a>] [<c010708b>] Code: 8b 04 91 31 d2 e8 37 9d
fe ff 31 db 3b 5d e8 7d 13 8b 75 f0
EIP; c0145e67 <sys_poll+29b/2ea> <=====
ecx; 00004300 Before first symbol
edx; c67ee320 <_end+64e77d4/85d04b4>
edi; bfffdcfc Before first symbol
ebp; c5a87fbc <_end+5781470/85d04b4>
esp; c5a87f7c <_end+5781430/85d04b4>
Trace; c010ca0a <sys_ipc+14e/278>
Trace; c010708b <syscall_call+7/b>
Code; c0145e67 <sys_poll+29b/2ea>
00000000 <_EIP>:
Code; c0145e67 <sys_poll+29b/2ea> <=====
0: 8b 04 91 mov (%ecx,%edx,4),%eax <=====
Code; c0145e6a <sys_poll+29e/2ea>
3: 31 d2 xor %edx,%edx
Code; c0145e6c <sys_poll+2a0/2ea>
5: e8 37 9d fe ff call fffe9d41 <_EIP+0xfffe9d41> c012fba8
<free_pages+0/24>
Code; c0145e71 <sys_poll+2a5/2ea>
a: 31 db xor %ebx,%ebx
Code; c0145e73 <sys_poll+2a7/2ea>
c: 3b 5d e8 cmp 0xffffffe8(%ebp),%ebx
Code; c0145e76 <sys_poll+2aa/2ea>
f: 7d 13 jge 24 <_EIP+0x24> c0145e8b
<sys_poll+2bf/2ea>
Code; c0145e78 <sys_poll+2ac/2ea>
11: 8b 75 f0 mov 0xfffffff0(%ebp),%esi
bdg:~#
From man:
"DESCRIPTION
ksymoops extracts kernel Oops reports from the Oops.file
and uses various sources of symbol information to convert
the addresses and code to meaningful text."
I guess the developers were a bit over confident...
Anyway, the original oops file (ksym.out) said that it was squid that caused
it. Would somebody please give a hand in deciphering the output...? I could
only guess, it was the sys_poll routine that had problems.
Actually, the machine didn't crash (could still be pinged); but problem is,
when it has oops, I could no longe ssh'ing into it. And at times (well,
usually), typing something on the login prompt giving me having more oops
printed on the screen. Because I can't access the machine in any way, pulling
out the power plug is usually the proper way to solve the problem. Could it be
due to a heavy DoS attack...? (I haven't tried pulling out the UTP cables;
yes, it makes me wonder too...)
BTW, what is the proper way to subscribe to the list in [EMAIL PROTECTED]
(http://www.gnu.org/directory/ksymoops.html).
Thanks in advance,
Oki
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
