Messages in this thread | | | Subject | 2.1.117 NFS Oops | From | Achim Oppelt <> | Date | 20 Aug 1998 17:49:07 +0200 |
| |
Hello,
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
| |