I had an 2.0.37 oops. The box in question has been very reliable in the
past (with 2.0.36 and 2.0.37). The only hitch was that I had to reboot
once because an oracle process went into state in which it couldn't be
killed even with kill -9.
Other than that, no problems; uptimes even more than a year.
A couple a days ago the box crashed, however. Before I noticed, it had
already booted, and the only thing I found in the logs was this:
general protection: 0000
CPU: 0
EIP: 0010:[<075fb000>]
EFLAGS: 00257706
eax: 00000000 ebx: 04f21fb4 ecx: 00000000 edx: 00000000
esi: 00000005 edi: bfffe3c8 ebp: 00000203 esp: 04f21f78
ds: 0018 es: 0018 fs: 002b gs: 002b ss: 0018
Process smbd (pid: 18364, process nr: 44, stackpage=04f21000)
Stack: 00000000 0012d03c 075fb000 075fb000 02c6d400 00000001 04f21fb4
bfffe388
080fee74 075fb000 0012b179 080fee74 04f21fb4 053ffc0c bffff09c
00000000
0010bbe9 080fee74 bfffe388 400fb6cc bffff09c 00000005 bfffe3c8
ffffffda
Call Trace: [namei+60/72] [sys_newstat+41/88] [system_call+85/124]
Code: 6c 69 62 72 61 72 69 65 73 2f 4f 62 6a 65 63 74 41 52 58 32
Of course, ksymoops wont get much out of that:
./ksymoops ../System.map < oops.13072000
Using `../System.map' to map addresses to symbols.
Code:
Code: 6c insb (%dx),%es:(%edi)
Code: 69 62 72 61 72 imull $0x65697261,0x72(%edx),%esp
Code: 69 65
Code: 73 2f jae 39 <_EIP+0x39>
Code: 4f decl %edi
Code: 62 6a 65 boundl 0x65(%edx),%ebp
Code: 63 74 41 52 arpl %si,0x52(%ecx,%eax,2)
Code: 58 popl %eax
Code: 32 00 xorb (%eax),%al
Code: 90 nop
Code: 90 nop
Code: 90 nop
Vanilla 2.0.37, Ppro200, 3C905. Nothing fancy.
This doesn't look like a tcp stack bug nor faulty memory...? Could there
still be a bug in 2.0 new_stat/namei?
-- v --
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Jul 31 2000 - 21:00:18 EST