Hello,<br />
<br />
We are running Centos 6.2 on an intels5520hc with an Areca 1880i raid controller in RAID 6. We just this as a backup server, and every week during heavy load, the server crashes so bad that we have to hard reboot. Here are the errors we see. Can you help figure out where the problem is? Here are the errors. We also use XFS, so this could also be the issue?<br />
<br />
Jul 1 03:17:17 sjback01 kernel: INFO: task nfsd:2304 blocked for more than 120 seconds.<br />
Jul 1 03:17:17 sjback01 kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.<br />
Jul 1 03:17:17 sjback01 kernel: nfsd D 0000000000000001 0 2304 2 0x00000080<br />
Jul 1 03:17:17 sjback01 kernel: ffff8801e86d5cc0 0000000000000046 0000000000000000 ffff8801d69b29c0<br />
Jul 1 03:17:17 sjback01 kernel: 0000008500000007 0000000000000003 ffff880199179190 ffff880199179190<br />
Jul 1 03:17:17 sjback01 kernel: ffff8801e6cdc678 ffff8801e86d5fd8 000000000000f4e8 ffff8801e6cdc678<br />
Jul 1 03:17:17 sjback01 kernel: Call Trace:<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffff8118318f>] ? inode_permission+0xaf/0xd0<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffff814ee8ae>] __mutex_lock_slowpath+0x13e/0x180<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffff814ee74b>] mutex_lock+0x2b/0x50<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffffa03fff75>] nfsd_unlink+0xa5/0x290 [nfsd]<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffffa0408073>] nfsd3_proc_remove+0x83/0x120 [nfsd]<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffffa03f943e>] nfsd_dispatch+0xfe/0x240 [nfsd]<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffffa036e534>] svc_process_common+0x344/0x640 [sunrpc]<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffff8105ea30>] ? default_wake_function+0x0/0x20<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffffa036eb70>] svc_process+0x110/0x160 [sunrpc]<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffffa03f9b62>] nfsd+0xc2/0x160 [nfsd]<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffffa03f9aa0>] ? nfsd+0x0/0x160 [nfsd]<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffff810909c6>] kthread+0x96/0xa0<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffff8100c14a>] child_rip+0xa/0x20<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffff81090930>] ? kthread+0x0/0xa0<br />
Jul 1 03:17:17 sjback01 kernel: [<ffffffff8100c140>] ? child_rip+0x0/0x20<br />
<br />
Thank you for helping figure this bug out.<br />
<br />
-Steve
↧