[LU-3786] performance-sanity test_6: mkdir hung (sys_mkdir)

• Type: Bug
• Resolution: Cannot Reproduce
• Priority: Major
• Affects Version/s: Lustre 2.4.0, Lustre 2.4.1, Lustre 2.6.0, Lustre 2.5.1, Lustre 2.5.3, Lustre 2.8.0, Lustre 2.10.0
• Labels: 22pl zfs
• Environment: Lustre Build: http://build.whamcloud.com/job/lustre-b2_4/33/; FSTYPE=zfs
• Severity: 3
• Assignee: Nathaniel Clark
• Reporter: Jian Yu

Description

performance-sanity test_6 hung as follows:

06:15:46:Lustre: DEBUG MARKER: ===== mdsrate-lookup-10dirs.sh Test preparation: creating 10 dirs with 6904 files.
06:15:46:INFO: task mkdir:3833 blocked for more than 120 seconds.
06:15:46:"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
06:15:46:mkdir         D 0000000000000000     0  3833   3826 0x00000080
06:15:46: ffff88000e523dd8 0000000000000086 ffffffffa10ccf38 0000000000000000
06:15:46: ffff88000e523e48 ffff88007afe7540 ffff88007afe7540 ffff88007afe7540
06:15:46: ffff88007afe7af8 ffff88000e523fd8 000000000000fb88 ffff88007afe7af8
06:15:46:Call Trace:
06:15:46: [<ffffffff811a1c00>] ? mntput_no_expire+0x30/0x110
06:15:46: [<ffffffff8150f3de>] __mutex_lock_slowpath+0x13e/0x180
06:15:46: [<ffffffff8150f27b>] mutex_lock+0x2b/0x50
06:15:46: [<ffffffff8118e930>] lookup_create+0x30/0xd0
06:17:49: [<ffffffff8119221c>] sys_mkdirat+0x7c/0x130
06:17:49: [<ffffffff811a1c00>] ? mntput_no_expire+0x30/0x110
06:17:49: [<ffffffff810dc867>] ? audit_syscall_entry+0x1d7/0x200
06:17:49: [<ffffffff811922e8>] sys_mkdir+0x18/0x20
06:17:49: [<ffffffff8100b072>] system_call_fastpath+0x16/0x1b
06:17:50:INFO: task mkdir:3835 blocked for more than 120 seconds.

Maloo report: https://maloo.whamcloud.com/test_sets/8358068c-0962-11e3-ad8a-52540035b04c

Issue Links

• is related to: LU-4427 performance-sanity test_6: mdsrate hung (sys_mknod)
• is related to: LU-16658 performance sanity test_6 mdsrate-lookup-10dirs - UCX ERROR
• is related to: LU-3753 MDS operation rate dropping sporadically