[LU-7141] sanity-quota test_21 hangs in dd

• Type: Bug
• Resolution: Cannot Reproduce
• Priority: Minor
• Affects Version/s: Lustre 2.8.0
• Environment: autotest
• Severity: 3
• Assignee: WC Triage
• Reporter: James Nunez (Inactive)

Description

sanity-quota test 21 times out while writing repeatedly to a file using dd. Logs for this time out are at https://testing.hpdd.intel.com/test_sets/8d0246a2-57b7-11e5-a084-5254006e85c2.

From the client(2) console log, we see

05:04:24:Lustre: DEBUG MARKER: /usr/sbin/lctl mark Set quota for 14 times
05:04:24:Lustre: DEBUG MARKER: Set quota for 14 times
05:04:24:Lustre: DEBUG MARKER: /usr/sbin/lctl mark Files for user \(quota_usr\):
05:04:24:Lustre: DEBUG MARKER: Files for user (quota_usr):
05:04:24:INFO: task dd:60608 blocked for more than 120 seconds.
05:04:24:      Not tainted 2.6.32-573.3.1.el6.x86_64 #1
05:04:24:"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
05:04:24:dd            D 0000000000000019     0 60608  60600 0x00000080
05:04:24: ffff880c3492bd18 0000000000000086 0000000000000000 ffff880c3492be08
05:04:24: ffff880c24bb703d ffff8806121bd340 00000a21bef69988 ffffffff8123285f
05:04:24: ffff880c3492bd68 0000000100a54afd ffff880b6ebea5f8 ffff880c3492bfd8
05:04:24:Call Trace:
05:04:24: [<ffffffff8123285f>] ? security_inode_permission+0x1f/0x30
05:04:24: [<ffffffff81539cd6>] __mutex_lock_slowpath+0x96/0x210
05:04:24: [<ffffffff815397fb>] mutex_lock+0x2b/0x50
05:04:24: [<ffffffff811a48c6>] do_filp_open+0x2d6/0xd20
05:04:24: [<ffffffff81156a20>] ? __vma_link_rb+0x30/0x40
05:04:24: [<ffffffff81156acb>] ? vma_link+0x9b/0xf0
05:04:24: [<ffffffff8129d9da>] ? strncpy_from_user+0x4a/0x90
05:04:24: [<ffffffff811b1d92>] ? alloc_fd+0x92/0x160
05:04:24: [<ffffffff8118e867>] do_sys_open+0x67/0x130
05:04:24: [<ffffffff8118e970>] sys_open+0x20/0x30
05:04:24: [<ffffffff8100b0d2>] system_call_fastpath+0x16/0x1b

Currently, sanity-quota test 21 is on the SLOW list. So, for patches submitted to master, this test is not run.

We’ve seen this issue at least once before on 2015-08-03 23:36:55 with logs at https://testing.hpdd.intel.com/test_sets/2f65e204-3b0d-11e5-9384-5254006e85c2