[LU-14938] fail_abort() in t-f should take care of MDTs

• Status: Resolved
• Project: Lustre
• Affects Version/s: Upstream
• Fix Version/s: Lustre 2.15.0
• Type: Bug
• Priority: Minor
• Reporter: Alex Zhuravlev
• Assignee: Alex Zhuravlev
• Resolution: Fixed
• Severity: 3

Description

currently fail_abort() in t-f ensures that all the clients are back, using lfs df: the first df ensures that the clients recognize its eviction, initiates reconnect and the second one verifies the clients are really back.
this doesn't work with MDT and the consequence is that such an evicted state is deteted by another MDT(s) at random point, results in EIO and breaks testing in constructions like mkdir $DIR/$tdir || error ...
IMO, fail_abort() should do something similar for MDTs.

Comments

[Gerrit Updater]

"Alex Zhuravlev <[e-mail]>" uploaded a new patch: https://review.whamcloud.com/44671
Subject: LU-14938 tests: fail_abort() in t-f to take care of MDTs
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 5e8b1989dc40319c3793e0e7f9f7261041cb00b4

[Andreas Dilger]

Shouldn't this be handled automatically for the MDS connections, rather than depending on the test framework to fix the connections? Otherwise there will be issues for systems on real recovery?

[Alex Zhuravlev]

I'm not sure - this is an eviction case which means "IO error is possible" ? and this is why we do have clients_up() in fail_abort() ?

[Gerrit Updater]

"Oleg Drokin <[e-mail]>" merged in patch https://review.whamcloud.com/44671/
Subject: LU-14938 tests: fail_abort() in t-f to take care of MDTs
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 436cd4fd21ffee5830c9b4e75055db80c47547d5

[Peter Jones]

Landed for 2.15