Since I've upgraded to OCFS2 1.2.4 on my two node RAC cluster, I've been
getting numerous kernel panics.
[mflusche@tul1tmdbd1 ~]$ uname -a
Linux tul1tmdbd1.tvguide.com 2.6.9-42.0.8.ELsmp #1 SMP Tue Jan 23
12:49:51 EST 2007 x86_64 x86_64 x86_64 GNU/Linux
[mflusche@tul1tmdbd1 ~]$ rpm -qa |grep ocfs2
ocfs2console-1.2.3-1
ocfs2-tools-1.2.3-1
ocfs2-tools-debuginfo-1.2.3-1
ocfs2-2.6.9-42.0.8.ELsmp-1.2.4-2
Any ideas? The same cluster ran fine with 1.2.3.
Thanks,
Matt
Kernel BUG at spinlock:74
invalid operand: 0000 [1] SMP
CPU 0
Modules linked in: 8021q hangcheck_timer mptctl mptbase cpqci(U)
netconsole netdump i2c_dev i2c_core ocfs2(U) debugfs(U) nfs lockd
nfs_acl sg ocfs2_dlmfs(U) ocfs2_dlm(U) oc
fs2_nodemanager(U) configfs(U) sunrpc lpfcdfc emcphr(U) emcpmpap(U)
emcpmpaa(U) emcpmpc(U) emcpmp(U) emcp(U) emcplib(U) button battery ac
ohci_hcd hw_random tg3(U) floppy d
m_snapshot dm_zero dm_mirror ext3 jbd dm_mod lpfc scsi_transport_fc
cciss sd_mod scsi_mod
Pid: 6647, comm: dlm_thread Tainted: P 2.6.9-42.0.8.ELsmp
RIP: 0010:[<ffffffff8030b1d8>] <ffffffff8030b1d8>{_spin_unlock+9}
RSP: 0000:00000107f84bde30 EFLAGS: 00010213
RAX: 0000000000000000 RBX: 0000010654000088 RCX: 0000010600000000
RDX: 0000000000000206 RSI: ffffffffa03a2bad RDI: 0000010654000088
RBP: 0000010654000000 R08: 00000107f84bc000 R09: 0000000000000000
R10: 0000000300000000 R11: ffffffffa03a2bad R12: 00000107f8dddc00
R13: 0000000000000000 R14: 00000000000002dd R15: 0000000000000000
FS: 0000000005eb2ae0(0000) GS:ffffffff804e5880(0000)
knlGS:00000000080568c0
CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b
CR2: 00000000939a000e CR3: 0000000000101000 CR4: 00000000000006e0
Process dlm_thread (pid: 6647, threadinfo 00000107f84bc000, task
00000100f54107f0)
Stack: ffffffffa03b90ae 0000000000000030 0000000000000001
0000010654000000
0000010654000088 00000107f8dddc00 ffffffffa03ba049
00000001b7eadfb2
0000006400000001 0000000000000000
Call Trace:<ffffffffa03b90ae>{:ocfs2_dlm:dlm_purge_lockres+805}
<ffffffffa03ba049>{:ocfs2_dlm:dlm_thread+472}
<ffffffff80135756>{autoremove_wake_function+0}
<ffffffff80135756>{autoremove_wake_function+0}
<ffffffff8014b4f4>{keventd_create_kthread+0}
<ffffffffa03b9e71>{:ocfs2_dlm:dlm_thread+0}
<ffffffff8014b4f4>{keventd_create_kthread+0}
<ffffffff8014b4cb>{kthread+200}
<ffffffff80110f47>{child_rip+8}
<ffffffff8014b4f4>{keventd_create_kthread+0}
<ffffffff8014b403>{kthread+0}
<ffffffff80110f3f>{child_rip+0}
Code: 0f 0b 19 4f 32 80 ff ff ff ff 4a 00 8b 07 85 c0 7e 0c 0f 0b
RIP <ffffffff8030b1d8>{_spin_unlock+9} RSP <00000107f84bde30>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://oss.oracle.com/pipermail/ocfs2-users/attachments/20070323/9a0071e7/attachment.html
On Fri, Mar 23, 2007 at 12:15:55PM -0500, Matthew Flusche wrote:> Since I've upgraded to OCFS2 1.2.4 on my two node RAC cluster, I've been > getting numerous kernel panics.The stack trace below isn't as complete as I'd like, but this looks like a bug which has been recently fixed in 1.2.5 which will be out soon. --Mark -- Mark Fasheh Senior Software Developer, Oracle mark.fasheh@oracle.com