[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

NFS+RAID5 oops

To: linux-xfs@oss.sgi.com
Subject: NFS+RAID5 oops
From: Matteo Centonza <matteo@sif.it>
Date: Wed, 4 Apr 2001 10:34:50 +0200 (CEST)
Sender: owner-linux-xfs@oss.sgi.com

Dear All,

congratulations for the great work. This oops refers to the PR 10 as
of 2001-03-21. My test system (RH 7.0) has three ide disks:

WDC AC36400L (6.4 Gb)
QUANTUM FIREBALL540A (540 Mb)
QUANTUM LP120A GM120A01X (120 Mb)

on a dual channel controller (ASUS P2B motherboard, Celeron 400 Mhz)

in raid 5 configuration (XFS, no mount or mkfs options used). I've run the
bonnie++ test (v. 1.01) without problems (except for the poor rating ;-)).

I've NFS exported and mounted (locally through net loopback) my RAID5
partition and while i was copying a file bigger than the remaining
filesystem size, the oops (repeated twice, output through ksymoops
attached) arose instead of the "no space left on device" error.

hope this helps,

M.C.

Apr  4 01:49:31 speedster kernel: Bad write on page 0xc1029db4
Apr  4 01:49:31 speedster kernel: Bad write on page 0xc1029d70
Apr  4 01:49:34 speedster kernel: kernel BUG at page_buf_io.c:906!
Apr  4 01:49:34 speedster kernel: invalid operand: 0000
Apr  4 01:49:34 speedster kernel: CPU:    0
Apr  4 01:49:34 speedster kernel: EIP:    0010:[hook_buffers_to_page_delay+36/100]
Apr  4 01:49:34 speedster kernel: EFLAGS: 00010282
Apr  4 01:49:34 speedster kernel: eax: 00000021   ebx: c1029db4   ecx: 00000000   edx: f$
Apr  4 01:49:34 speedster kernel: esi: c21479c0   edi: c09da000   ebp: c1ba55c0   esp: c$
Apr  4 01:49:34 speedster kernel: ds: 0018   es: 0018   ss: 0018
Apr  4 01:49:34 speedster kernel: Process nfsd (pid: 450, stackpage=c2d4f000)
Apr  4 01:49:34 speedster kernel: Stack: c030f985 c030fac7 0000038a c21479c0 00000000 c0$
Apr  4 01:49:34 speedster kernel:        00001000 c01b76d4 c21479c0 c1029db4 00000000 00$
Apr  4 01:49:34 speedster kernel:        ffffffff 00001000 c2d4fce8 c2d4fcf4 00001000 00$
Apr  4 01:49:34 speedster kernel: Call Trace: [__pb_block_commit_write_async+69/76] [pag$
Apr  4 01:49:34 speedster kernel:        [linvfs_write+276/320] [nfsd_write+320/680] [nf$
Apr  4 01:49:34 speedster kernel:
Apr  4 01:49:34 speedster kernel: Code: 0f 0b 83 c4 0c 68 00 10 00 00 0f b7 46 28 50 53 $

ksymoops 2.3.4 on i686 2.4.2-SGI_XFS_0.10-1.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.2-SGI_XFS_0.10-1/ (default)
     -m /boot/System.map-2.4.2-SGI_XFS_0.10-1 (specified)

Warning (compare_maps): mismatch on symbol partition_name  , ksyms_base says c029b120, System.map says c014cb60.  Ignoring ksyms_base entry
Apr  4 01:49:34 speedster kernel: invalid operand: 0000
Apr  4 01:49:34 speedster kernel: CPU:    0
Apr  4 01:49:34 speedster kernel: EIP:    0010:[hook_buffers_to_page_delay+36/100]
Apr  4 01:49:34 speedster kernel: EFLAGS: 00010282
Apr  4 01:49:34 speedster kernel: eax: 00000021   ebx: c1029db4   ecx: 00000000   edx: f$
Apr  4 01:49:34 speedster kernel: esi: c21479c0   edi: c09da000   ebp: c1ba55c0   esp: c$
Apr  4 01:49:34 speedster kernel: ds: 0018   es: 0018   ss: 0018
Apr  4 01:49:34 speedster kernel: Process nfsd (pid: 450, stackpage=c2d4f000)
Apr  4 01:49:34 speedster kernel: Stack: c030f985 c030fac7 0000038a c21479c0 00000000 c0$
Apr  4 01:49:34 speedster kernel:        00001000 c01b76d4 c21479c0 c1029db4 00000000 00$
Apr  4 01:49:34 speedster kernel:        ffffffff 00001000 c2d4fce8 c2d4fcf4 00001000 00$
Apr  4 01:49:34 speedster kernel: Call Trace: [__pb_block_commit_write_async+69/76] [pag$
Apr  4 01:49:34 speedster kernel: Code: 0f 0b 83 c4 0c 68 00 10 00 00 0f b7 46 28 50 53 $
Warning (Oops_code): trailing garbage ignored on Code: line
  Text: 'Code: 0f 0b 83 c4 0c 68 00 10 00 00 0f b7 46 28 50 53 $'
  Garbage: ' $'
Using defaults from ksymoops -t elf32-i386 -a i386

Code;  00000000 Before first symbol
00000000 <_EIP>:
Code;  00000000 Before first symbol
   0:   0f 0b                     ud2a   
Code;  00000002 Before first symbol
   2:   83 c4 0c                  add    $0xc,%esp
Code;  00000005 Before first symbol
   5:   68 00 10 00 00            push   $0x1000
Code;  0000000a Before first symbol
   a:   0f b7 46 28               movzwl 0x28(%esi),%eax
Code;  0000000e Before first symbol
   e:   50                        push   %eax
Code;  0000000f Before first symbol
   f:   53                        push   %ebx


2 warnings issued.  Results may not be reliable.

Bad write on page 0xc1029db4
Bad write on page 0xc1029d70
kernel BUG at page_buf_io.c:906!
invalid operand: 0000
CPU:    0
EIP:    0010:[<c01b8170>]
EFLAGS: 00010282
eax: 00000021   ebx: c1029db4   ecx: 00000000   edx: ffffffff
esi: c21479c0   edi: c09da000   ebp: c1ba55c0   esp: c2d4fc90
ds: 0018   es: 0018   ss: 0018
Process nfsd (pid: 450, stackpage=c2d4f000)
Stack: c030f985 c030fac7 0000038a c21479c0 00000000 c01b8685 c21479c0 c1029db4
       00001000 c01b76d4 c21479c0 c1029db4 00000000 00001000 00000000 c1ba55c0
       ffffffff 00001000 c2d4fce8 c2d4fcf4 00001000 00000000 c1029db4 00000000
Call Trace: [<c01b8685>] [<c01b76d4>] [<c02279fb>] [<c0202fa4>] [<c0203080>] [<c01defcc>] [<c0227fbe>]
       [<c0224098>] [<c0179c18>] [<c02ae9f2>] [<c0176dc7>] [<c0176473>] [<c02db8f8>] [<c0176241>] [<c010744f>]

Code: 0f 0b 83 c4 0c 68 00 10 00 00 0f b7 46 28 50 53 e8 4b 74 f7

ksymoops 2.3.4 on i686 2.4.2-SGI_XFS_0.10-1.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.2-SGI_XFS_0.10-1/ (default)
     -m /boot/System.map-2.4.2-SGI_XFS_0.10-1 (specified)

Warning (compare_maps): mismatch on symbol partition_name  , ksyms_base says c029b120, System.map says c014cb60.  Ignoring ksyms_base entry
invalid operand: 0000
CPU:    0
EIP:    0010:[<c01b8170>]
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010282
eax: 00000021   ebx: c1029db4   ecx: 00000000   edx: ffffffff
esi: c21479c0   edi: c09da000   ebp: c1ba55c0   esp: c2d4fc90
ds: 0018   es: 0018   ss: 0018
Process nfsd (pid: 450, stackpage=c2d4f000)
Stack: c030f985 c030fac7 0000038a c21479c0 00000000 c01b8685 c21479c0 c1029db4
       00001000 c01b76d4 c21479c0 c1029db4 00000000 00001000 00000000 c1ba55c0
       ffffffff 00001000 c2d4fce8 c2d4fcf4 00001000 00000000 c1029db4 00000000
Call Trace: [<c01b8685>] [<c01b76d4>] [<c02279fb>] [<c0202fa4>] [<c0203080>] [<c01defcc>] [<c0227fbe>]
       [<c0224098>] [<c0179c18>] [<c02ae9f2>] [<c0176dc7>] [<c0176473>] [<c02db8f8>] [<c0176241>] [<c010744f>]
Code: 0f 0b 83 c4 0c 68 00 10 00 00 0f b7 46 28 50 53 e8 4b 74 f7

>>EIP; c01b8170 <hook_buffers_to_page_delay+24/64>   <=====
Trace; c01b8685 <__pb_block_commit_write_async+45/4c>
Trace; c01b76d4 <pagebuf_iozero+f0/16c>
Trace; c02279fb <xfs_zero_eof+31b/5d4>
Trace; c0202fa4 <xfs_ilock+0/18>
Trace; c0203080 <xfs_iunlock+0/68>
Trace; c01defcc <xfs_bmapi+0/12d8>
Trace; c0227fbe <xfs_write+30a/660>
Trace; c0224098 <linvfs_write+114/140>
Trace; c0179c18 <nfsd_write+140/2a8>
Trace; c02ae9f2 <nf_hook_slow+ca/10c>
Trace; c0176dc7 <nfsd_proc_write+c7/d0>
Trace; c0176473 <nfsd_dispatch+cb/168>
Trace; c02db8f8 <svc_process+2ac/544>
Trace; c0176241 <nfsd+191/2f8>
Trace; c010744f <kernel_thread+23/30>
Code;  c01b8170 <hook_buffers_to_page_delay+24/64>
00000000 <_EIP>:
Code;  c01b8170 <hook_buffers_to_page_delay+24/64>   <=====
   0:   0f 0b                     ud2a      <=====
Code;  c01b8172 <hook_buffers_to_page_delay+26/64>
   2:   83 c4 0c                  add    $0xc,%esp
Code;  c01b8175 <hook_buffers_to_page_delay+29/64>
   5:   68 00 10 00 00            push   $0x1000
Code;  c01b817a <hook_buffers_to_page_delay+2e/64>
   a:   0f b7 46 28               movzwl 0x28(%esi),%eax
Code;  c01b817e <hook_buffers_to_page_delay+32/64>
   e:   50                        push   %eax
Code;  c01b817f <hook_buffers_to_page_delay+33/64>
   f:   53                        push   %ebx
Code;  c01b8180 <hook_buffers_to_page_delay+34/64>
  10:   e8 4b 74 f7 00            call   f77460 <_EIP+0xf77460> c112f5d0 <_end+d441fc/4438c2c>


1 warning issued.  Results may not be reliable.

Follow-Ups:
- Re: NFS+RAID5 oops
  - From: Steve Lord <lord@sgi.com>

Prev by Date: TAKE - xfs_fsr
Next by Date: Re: shutdown umount hangs
Previous by thread: TAKE - xfs_fsr
Next by thread: Re: NFS+RAID5 oops
Index(es):
- Date
- Thread