[LU-2918] Test failure on test suite sanity Created: 06/Mar/13  Updated: 01/Oct/14  Resolved: 13/Mar/13

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.4.0
Fix Version/s: Lustre 2.4.0

Type: Bug Priority: Blocker
Reporter: Maloo Assignee: Jinshan Xiong (Inactive)
Resolution: Fixed Votes: 0
Labels: HB

Issue Links:
Related
is related to LU-2805 Test failure on test suite sanity, su... Resolved
Severity: 3
Rank (Obsolete): 7015

 Description   

This issue was created by maloo for jay <jay@whamcloud.com>

This issue relates to the following test suite run: https://maloo.whamcloud.com/test_sets/fcce89f4-85a6-11e2-926e-52540035b04c.

== sanity test 184c: Concurrent write and layout swap == 01:39:55 (1362476395)
50+0 records in
50+0 records out
52428800 bytes (52 MB) copied, 10.6071 s, 4.9 MB/s
48+0 records in
48+0 records out
50331648 bytes (50 MB) copied, 10.3514 s, 4.9 MB/s
ref file size: ref1(52428800), ref2(50331648)
error: cannot open for write /mnt/lustre/d0.sanity/d184/184c/file1: No such file or directory (2)
3200+0 records in
3200+0 records out
52428800 bytes (52 MB) copied, 3.44912 s, 15.2 MB/s
sanity test_184c: @@@@@@ FAIL: swap of /mnt/lustre/d0.sanity/d184/184c/file1 and /mnt/lustre/d0.sanity/d184/184c/file2 failed
Trace dump:
= /usr/lib64/lustre/tests/test-framework.sh:3973:error_noexit()
= /usr/lib64/lustre/tests/test-framework.sh:3996:error()
= /usr/lib64/lustre/tests/sanity.sh:9433:test_184c()
= /usr/lib64/lustre/tests/test-framework.sh:4251:run_one()
= /usr/lib64/lustre/tests/test-framework.sh:4284:run_one_logged()
= /usr/lib64/lustre/tests/test-framework.sh:4139:run_test()
= /usr/lib64/lustre/tests/sanity.sh:9451:main()
Dumping lctl log to /logdir/test_logs/2013-03-05/lustre-reviews-el6-x86_64-review-1_2_1_13755_-70011916101860-001958/sanity.test_184c.*.1362476424.log
CMD: client-13vm1.lab.whamcloud.com,client-13vm2,client-13vm3,client-13vm4 /usr/sbin/lctl dk > /logdir/test_logs/2013-03-05/lustre-reviews-el6-x86_64-review-1_2_1_13755_-70011916101860-001958/sanity.test_184c.debug_log.\$(hostname -s).1362476424.log;
dmesg > /logdir/test_logs/2013-03-05/lustre-reviews-el6-x86_64-review-1_2_1_13755_-70011916101860-001958/sanity.test_184c.dmesg.\$(hostname -s).1362476424.log



 Comments   
Comment by Jinshan Xiong (Inactive) [ 06/Mar/13 ]

Apparently when it tried to swap layout, the dd process wasn't even started to copy file. Usually the script sleeps a little while to yield CPU but it's broken sometimes on a busy VM.

I think we need a reliable way to yield CPU and make sure the execution order of processes. It will be implemented in framework.

Comment by Jinshan Xiong (Inactive) [ 06/Mar/13 ]

patch is at: http://review.whamcloud.com/5617

Generated at Sat Feb 10 01:29:21 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.