[LU-15981] cascading_rw fails with enospc when osts are small Created: 28/Jun/22  Updated: 28/Jun/22

Status: Open
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Minor
Reporter: Vladimir Saveliev Assignee: Vladimir Saveliev
Resolution: Unresolved Votes: 0
Labels: None

Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

On unlink ost object destroys get delayed until osp_sync_thread decides to cancel unlink records. This makes cascading_rw test to fail due to enospc if OSTs are not big enough as the test consumes disk space faster that osp_sync_thread generates OST_DESTROYs.

OSTSIZE=200000 SLOW=yes sh parallel-scale.sh should be enough to see the problem.



 Comments   
Comment by Gerrit Updater [ 28/Jun/22 ]

"Vladimir Saveliev <vladimir.saveliev@hpe.com>" uploaded a new patch: https://review.whamcloud.com/47823
Subject: LU-15981 tests: truncate before unlink in cascading_rw
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 627ba9d74ca147f40afb06f3bd0fc11d74e4e964

Generated at Sat Feb 10 03:22:57 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.