Does anyone know why sanboot redhat5 U7 OS kernel panic after running IO about 2 hours?

[dandandongdong]

Hi,

I sanboot installed redhat 5u7 x86_64 os on UCS, and then mapped 15 volumes to run IO,  os will become kernel panic less then 2 hours.

then I have to force power cycle the server.

The card used is M81KR.

Does anyone what is the problem? Thanks.

========================

Feb 26 19:43:59 arcucsb200095e0 kernel: INFO: task kjournald:6427 blocked for more than 120 seconds.
Feb 26 19:43:59 arcucsb200095e0 kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Feb 26 19:43:59 arcucsb200095e0 kernel: kjournald     D ffffffff801546d1     0  6427    465          6430  6424 (L-TLB)
Feb 26 19:43:59 arcucsb200095e0 kernel:  ffff81066b6f1cf0 0000000000000046 000000006e420048 ffff8106853ea9d0
Feb 26 19:43:59 arcucsb200095e0 kernel:  00000001ffffffff 000000000000000a ffff810c6c31b080 ffff8106853f07a0
Feb 26 19:43:59 arcucsb200095e0 kernel:  00000184ac3ecc0b 0000000000000edf ffff810c6c31b268 0000000b1693b738
Feb 26 19:43:59 arcucsb200095e0 kernel: Call Trace:
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff8006ec8f>] do_gettimeofday+0x40/0x90
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff800155d3>] sync_buffer+0x0/0x3f
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff800637ce>] io_schedule+0x3f/0x67
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff8001560e>] sync_buffer+0x3b/0x3f
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff800639fa>] __wait_on_bit+0x40/0x6e
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff800155d3>] sync_buffer+0x0/0x3f
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff80063a94>] out_of_line_wait_on_bit+0x6c/0x78
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff800a2e8b>] wake_bit_function+0x0/0x23
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff88033a41>] :jbd:journal_commit_transaction+0x553/0x10aa
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff8003d85b>] lock_timer_base+0x1b/0x3c
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff8004ad98>] try_to_del_timer_sync+0x7f/0x88
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff88037662>] :jbd:kjournald+0xc1/0x213
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff800a2e5d>] autoremove_wake_function+0x0/0x2e
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff880375a1>] :jbd:kjournald+0x0/0x213
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff800a2c45>] keventd_create_kthread+0x0/0xc4
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff80032722>] kthread+0xfe/0x132
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff8005dfb1>] child_rip+0xa/0x11
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff800a2c45>] keventd_create_kthread+0x0/0xc4
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff80032624>] kthread+0x0/0x132
Feb 26 19:43:59 arcucsb200095e0 kernel:  [<ffffffff8005dfa7>] child_rip+0x0/0x11

====================

Many thanks for your help.

[padramas]

Hello,

Please provide following information

UCSM version

Blade model

CIMC, BIOS and M81KR adapter firmware version

ENIC and FNIC driver version

What type of storage are you trying to mount ?

Are you able to consistently reproduce the issue ?

Padma

[dandandongdong]

Hi

Thanks for your help. 

UCSM version   : 2.0(1s)

Blade model  :  Cisco UCS B2000 M1

CIMC: 2.0( 1q)

BIOS: S5500.2.0.1a.0.081120111606

and M81KR adapter firmware version :  2.0(1q)

ENIC and FNIC driver version: 5.0(3)N2(2.1s)

What type of storage are you trying to mount : The storage is IBM SVC.

Thanks

[padramas]

Hello

You can get the driver version by executing following command in RHEL

modinfo enic

modinfo fnic

Can you consistently reproduce the crash ?

I am not aware of any known issues.

Please open a TAC service request with above information and also upload Redhat logs ( which can be collected by executing "  sosreport -k rpm.rpmva=off " command )

Padma