Hi,everyone:
I use cluster-2.03.00 in CentOS-4.6-Final.
Kernel comiled by myself is 2.6.24.4 .
I test it with two nodes.
GFS2 runs fine. But GFS1 would cause kernel panic (40 dd write). the panic
log is:
BUG: unable to handle kernel NULL pointer dereference at virtual address 00000040
printing eip: c0120cf1 *pdpt = 0000000036015001 *pde = 0000000000000000
Oops: 0000 [#1] SMP
Modules linked in: gfs lock_dlm dlm configfs gfs2 binfmt_misc dm_mirror dm_round_robin dm_multipath thermal processor fan containerd
Pid: 4031, comm: bash Not tainted (2.6.24.4.OWLsmp #1)
EIP: 0060:[<c0120cf1>] EFLAGS: 00010046 CPU: 3
EIP is at pick_next_task_fair+0x12/0x1f
EAX: 00000000 EBX: f783e030 ECX: c039c740 EDX: 00000000
ESI: 00000003 EDI: f6da2ac0 EBP: f6da2ac0 ESP: f7119e98
DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
Process bash (pid: 4031, ti=f7118000 task=f5f898f0 task.ti=f7118000)
Stack: c0391bb2 c202bc80 f5f89a54 7fffffff f5c53000 f7119f1c bff5b72f c03921ae
f60c5c00 00000011 f554f000 c0239c5b 00000000 f73f7c00 f5c53000 00000246
c013917e f5c5300c f5c5300c f5c53000 c0238a7b f554f011 7fffffff 00000000
Call Trace:
[<c0391bb2>] __sched_text_start+0x11a/0x350
[<c03921ae>] schedule_timeout+0x13/0x8d
[<c0239c5b>] pty_write+0x2f/0x35
[<c013917e>] add_wait_queue+0x12/0x30
[<c0238a7b>] read_chan+0x28e/0x496
[<c012341a>] default_wake_function+0x0/0x8
[<c012341a>] default_wake_function+0x0/0x8
[<c0234ab4>] tty_read+0x75/0xac
[<c0175a12>] vfs_read+0x89/0x12e
[<c0175d24>] sys_read+0x41/0x67
[<c0104e7a>] sysenter_past_esp+0x6b/0xa1
[<c0390000>] p9_create_tauth+0x1f/0xb2
=======================
Code: da 77 11 72 04 39 c8 73 0b 5b 89 f8 5e 5f 5d e9 74 f7 ff ff 5b 5e 5f 5d c3 83 c0 34 31 d2 83 78 08 00 74 11 e8 4a fe ff ff 89
EIP: [<c0120cf1>] pick_next_task_fair+0x12/0x1f SS:ESP 0068:f7119e98
---[ end trace 07f7603309b1b483 ]--- Could anyone help ?
Thanks !
2008-04-23
dhongqian
|
-- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster