[LU-1692] Test failure on test suite sanityn test_36

• Status: Resolved
• Project: Lustre
• Affects Version/s: Lustre 2.1.3
• Type: Bug
• Priority: Major
• Reporter: Maloo
• Assignee: Yang Sheng
• Resolution: Cannot Reproduce
• Severity: 3

Description

This issue was created by maloo for jay <[e-mail]>

This issue relates to the following test suite run: https://maloo.whamcloud.com/test_sets/3acd91f2-db0a-11e1-9ebb-52540035b04c.

From the log on the client-2, it had lots of short read:

== sanityn test 36: handle ESTALE/open-unlink corectly =============================================== 21:29:40 (1343708980)
50+0 records in
50+0 records out
52428800 bytes (52 MB) copied, 1.72915 seconds, 30.3 MB/s
multiop /mnt/lustre2/d0.sanityn/d36/file000 vO_r52428800c
TMPPIPE=/tmp/multiop_open_wait_pipe.14782
*** cycle(0) *** before(45020112):after_dd(44976080):after(45020112)
50+0 records in
50+0 records out
52428800 bytes (52 MB) copied, 1.63639 seconds, 32.0 MB/s
multiop /mnt/lustre2/d0.sanityn/d36/file000 vO_r52428800c
TMPPIPE=/tmp/multiop_open_wait_pipe.14782
*** cycle(1) *** before(45020112):after_dd(44976080):after(45020112)
50+0 records in
50+0 records out
52428800 bytes (52 MB) copied, 1.65245 seconds, 31.7 MB/s
multiop /mnt/lustre2/d0.sanityn/d36/file000 vO_r52428800c
TMPPIPE=/tmp/multiop_open_wait_pipe.14782
*** cycle(2) *** before(45020112):after_dd(44976080):after(45020112)
50+0 records in
50+0 records out
52428800 bytes (52 MB) copied, 1.66976 seconds, 31.4 MB/s
multiop /mnt/lustre2/d0.sanityn/d36/file000 vO_r52428800c
TMPPIPE=/tmp/multiop_open_wait_pipe.14782
*** cycle(3) *** before(45020112):after_dd(44976080):after(45020112)
50+0 records in
50+0 records out
52428800 bytes (52 MB) copied, 1.60533 seconds, 32.7 MB/s
multiop /mnt/lustre2/d0.sanityn/d36/file000 vO_r52428800c
TMPPIPE=/tmp/multiop_open_wait_pipe.14782
*** cycle(4) *** before(45020112):after_dd(44976080):after(45020112)
50+0 records in
50+0 records out
52428800 bytes (52 MB) copied, 1.63088 seconds, 32.1 MB/s
multiop /mnt/lustre2/d0.sanityn/d36/file000 vO_r52428800c
TMPPIPE=/tmp/multiop_open_wait_pipe.14782
short read: 51380224/52428800
short read: 0/1048576
short read: 0/1048576
short read: 0/1048576
...

Comments

[Jodi Levi (Inactive)]

We think this is a blocker based on the failure rate

[Peter Jones]

Yangsheng will look into this one

[Peter Jones]

Only seen on b2_1 so far

[Yang Sheng]

Haven't hit a long time. So close first, Please feel free to reopen it.