[LU-8067] sanityn test_31a: read dd: reading `/mnt/lustre2/d31a.sanityn/f31a.sanityn': Input/output error 0 blocks, must be

• Status: Resolved
• Project: Lustre
• Type: Bug
• Priority: Minor
• Reporter: Maloo
• Assignee: WC Triage
• Resolution: Duplicate
• Severity: 3

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/d0ef7830-0b2b-11e6-b5f1-5254006e85c2.

The sub-test test_31a failed with the following error:

read dd: reading `/mnt/lustre2/d31a.sanityn/f31a.sanityn': Input/output error 0 blocks, must be 1

This error just started happening a lot ~4/21.
Suspect it's related to the recent landing of "LU-7791 ldlm: signal vs CP callback race"
That's just a guess. The timeframe is right, but no strong evidence that's the cause.

Info required for matching: sanityn 31a

Issue Links

• is related to: LU-8062 recovery-small test_115b: @@@@@@ FAI...
• is related to: LU-7434 lost bulk leads to a hang
• is related to: LU-7791 signal vs CP callback race

Comments

[Andreas Dilger]

Looks like this failure is also a recent regression from landing either LU-7791 or LU-7434, since it also started failing on 2016-04-21.

[Vitaly Fertman]

related to LU-7791

[Andreas Dilger]

Closing as a duplicate of LU-7791 so that the updated patch can land under the original ticket.