we can reproducde a kernel Oops on a dual PPro running 2.1.117 SMP. The
program in question dies with a segmentation fault when running in a
directory mounted over NFS, it works great when run locally. The relevant
Oops is:
Aug 20 14:53:36 cognac kernel: Oops: 0000
Aug 20 14:53:36 cognac kernel: CPU: 0
Aug 20 14:53:36 cognac kernel: EIP: 0010:[<c88818cd>]
Aug 20 14:53:36 cognac kernel: EFLAGS: 00010292
Aug 20 14:53:36 cognac kernel: eax: 14000000 ebx: c4f41f4c ecx: c021cd8c e
dx: c021cd70
Aug 20 14:53:36 cognac kernel: esi: c021cd70 edi: c4f41f4c ebp: c7afab40 e
sp: c4f41f3c
Aug 20 14:53:36 cognac kernel: ds: 0018 es: 0018 ss: 0018
Aug 20 14:53:36 cognac kernel: Process DROBER.exe (pid: 4665, process nr: 101, s
tackpage=c4f41000)
Aug 20 14:53:36 cognac kernel: Stack: c42c4e00 00000000 c7d9e3a0 00000246 c4f400
00 c021cd8c 00000000 00000000
Aug 20 14:53:36 cognac kernel: c8880169 c42c4e00 00000000 00000000 000000
00 c1e8a9c0 c42c4e00 c012840a
Aug 20 14:53:36 cognac kernel: c42c4e00 c1e8a9c0 c1e8a9c0 00000000 ffffff
f7 bffe8bd0 c0128487 c1e8a9c0
Aug 20 14:53:36 cognac kernel: Call Trace: [<c8880169>] [<c012840a>] [<c0128487>
] [<c012851c>] [<c010ae98>]
Aug 20 14:53:36 cognac kernel: Code: 8b 40 5c 8b 40 44 50 e8 93 7a fe ff 83 c4 0
8 e9 af fe ff ff
Looking in the system map and the output of ksyms, I would interpret the
call trace like this (unfortunately neither ksymoops nor syslogd/klogd
seems capable of decoding this):
c8878954 lockd_up_Rsmp_f6933c48 [lockd]
c01283e8 T __fput
c0128438 T close_fp
c01284bc T sys_close
c010ae80 t tracesys
The sequence of system calls related to the file that is being closed seems
to be:
open("fort.3", O_RDONLY) = 9
read(9, "(\0\0\0\3\0\0\0-\0\0\0\0\0\0\0\0"..., 360) = 360
close(9 <unfinished ...>
+++ killed by SIGSEGV +++
The directory is mounted via NFS (using autofs) from a machine running
Linux 2.0.34.
Is this a known problem? Is there anything further I can do to help debug
this?
Achim
-
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.altern.org/andrebalsa/doc/lkml-faq.html