[LU-16290] Long LNet ping unlink times Created: 02/Nov/22  Updated: 29/Jul/23  Resolved: 14/Nov/22

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: Lustre 2.16.0

Type: Bug Priority: Minor
Reporter: Chris Horn Assignee: Chris Horn
Resolution: Fixed Votes: 0
Labels: None

Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

If an lnet ping fails to send an LNET_EVENT_SEND is generated with a non-zero status. However, the ping event handler only calls signals the waiting thread when the unlink event occurs. This means that the thread which issued the ping must wait for full timeout to hit before it wakes up and calls LNetMDUnlink(). The ping event handler should signal completion if we get a non-zero RC on the send event.



 Comments   
Comment by Gerrit Updater [ 02/Nov/22 ]

"Chris Horn <chris.horn@hpe.com>" uploaded a new patch: https://review.whamcloud.com/c/fs/lustre-release/+/49020
Subject: LU-16290 lnet: Signal completion on ping send failure
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: d3c8f7f02b16d2b21d6ac09086eb340b704b7f1c

Comment by Gerrit Updater [ 14/Nov/22 ]

"Oleg Drokin <green@whamcloud.com>" merged in patch https://review.whamcloud.com/c/fs/lustre-release/+/49020/
Subject: LU-16290 lnet: Signal completion on ping send failure
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 48c34c71de65e8a251a218bc9ecb7c5ed522d786

Comment by Peter Jones [ 14/Nov/22 ]

Landed for 2.16

Generated at Sat Feb 10 03:25:41 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.