[LU-5338] sanity-lfsck test_11b: (9) expect lastid1 0x100000000:569

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.14.0
• Affects Version/s: Lustre 2.10.5, Lustre 2.10.7, Lustre 2.10.8
• Labels: lfsck
• Severity: 3
• Assignee: Andreas Dilger
• Reporter: Maloo

Description

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

This issue relates to the following test suite run: https://testing.hpdd.intel.com/test_sets/1d50d682-0990-11e4-a958-5254006e85c2.

This may be a dup of LU-5080 but the reported error message looks somewhat different. There have be occasional instances of this failure starting ~5/5, none before that I can find.

The sub-test test_11b failed with the following error:

(9) expect lastid1 0x100000000:569

Info required for matching: sanity-lfsck 11b

Issue Links

• is duplicated by: LU-5080 Test failure on test suite sanity-lfsck, subtest test_11b

Comments

[Peter Jones]

Landed for 2.14

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/34349/
Subject: LU-5338 tests: sanity-lfsck 11b allow larger last_id
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: a2e38010077df3d95c9942d4f9cde829f75e7267

[Gerrit Updater]

Andreas Dilger ([e-mail]) uploaded a new patch: https://review.whamcloud.com/34349
Subject: LU-5338 tests: sanity-lfsck 11b allow larger last_id
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: bdf000af472de7b605797dc88f42dcc8dcec6fbf

[Alex Zhuravlev]

https://testing.whamcloud.com/test_sets/aaebc59a-39b2-11e9-913f-52540065bddc

[Minh Diep]

+1 on b2_10: https://testing.whamcloud.com/test_sessions/7fd76655-6a66-4fe8-bd8e-d538bf73351d

[Jian Yu]

The failure occurred again on Lustre b2_10 branch:
https://testing.whamcloud.com/test_sets/2422d4e4-f4b7-11e8-bfe1-52540065bddc

[Andreas Dilger]

The test log shows:

Update not seen after 60s: wanted '569' got '577'

It looks like the test could be improved a bit:

        echo "the on-disk LAST_ID should have been rebuilt"
        wait_update_facet ost1 "$LCTL get_param -n \
                obdfilter.${ost1_svc}.last_id | grep 0x100000000 |
                awk -F: '{ print \\\$2 }'" "$lastid1" 60 || {
                $LCTL get_param -n obdfilter.${ost1_svc}.last_id
                error "(9) expect lastid1 0x100000000:$lastid1"
        }

The actual value should be printed in the error message, and it isn't clear that having a higher LAST_ID value is a failure?