[LU-4404] sanity-quota test_0: FAIL: SLOW IO for quota_usr (user): 50 KB/sec

• Status: Closed
• Project: Lustre
• Affects Version/s: Lustre 2.4.2
• Type: Bug
• Priority: Minor
• Reporter: Jian Yu
• Assignee: WC Triage
• Resolution: Cannot Reproduce
• Labels: yuc2

Environment

Lustre Tag: 2.4.2 RC1
Lustre Client: CentOS 6.5/x86_64 (kernel version: 2.6.32-431.1.2.0.1.el6.x86_64)
Lustre Server: CentOS 6.4/x86_64 (kernel version: 2.6.32-358.23.2.el6_lustre.x86_64)

• Severity: 3

Description

sanity-quota test 0 failed as follows:

running as uid/gid/euid/egid 60000/60000/60000/60000, groups:
 [dd] [if=/dev/zero] [bs=1M] [of=/mnt/lustre/d0.sanity-quota/d0/f.sanity-quota.0-0] [count=100] [conv=fsync]
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 2046.85 s, 51.2 kB/s
 sanity-quota test_0: @@@@@@ FAIL: SLOW IO for quota_usr (user): 50 KB/sec 

Dmesg on OSS showed that:

LustreError: 11-0: lustre-MDT0000-lwp-OST0001: Communicating with 10.10.18.154@tcp, operation ldlm_enqueue failed with -3.
LustreError: 42466:0:(qsd_handler.c:344:qsd_req_completion()) $$$ DQACQ failed with -3, flags:0x1 qsd:lustre-OST0001 qtype:grp id:60000 enforced:1 granted:0 pending:0 waiting:2064 req:1 usage:0 qunit:0 qtune:0 edquot:0
LustreError: 42466:0:(qsd_handler.c:768:qsd_op_begin0()) $$$ ID isn't enforced on master, it probably due to a legeal race, if this message is showing up constantly, there could be some inconsistence between master & slave, and quota reintegration needs be re-triggered. qsd:lustre-OST0001 qtype:grp id:60000 enforced:1 granted:0 pending:0 waiting:1032 req:0 usage:0 qunit:0 qtune:0 edquot:0

Maloo report:
https://maloo.whamcloud.com/test_sets/a3c4691c-6941-11e3-9036-52540035b04c

Issue Links

• is related to: LU-6382 quota : inconsistence between master ...

Comments

[Jian Yu]

Re-running the test passed on the same build and configuration:
https://maloo.whamcloud.com/test_sets/2fecef54-6a2d-11e3-81c0-52540035b04c

[Jodi Levi (Inactive)]

Yu Jian,
Do you think we should close this ticket or do you want to try reproducing again and then determine?

[Jian Yu]

Let's close this ticket. If I hit the failure again, I'll reopen this ticket.