[LU-8826] recovery hard time should not be shrunk for IR Created: 11/Nov/16  Updated: 12/Apr/17  Resolved: 17/Dec/16

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: Lustre 2.10.0

Type: Bug Priority: Minor
Reporter: Niu Yawei (Inactive) Assignee: Niu Yawei (Inactive)
Resolution: Fixed Votes: 0
Labels: None

Issue Links:
Duplicate
Related
Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

Otherwise, soft limit will be very close to hard limit, and on sluggish network, extended soft limit can reach hard limit easily, and the whole recovery will be aborted easier. Actually, IR is used to shorten the client reconnect time during recovery, so shrink soft limit is good enough. It doesn't make sense to shrink the hard limit.



 Comments   
Comment by Niu Yawei (Inactive) [ 11/Nov/16 ]

Another problem is that even when exp_need_sync is set, we missed to turn transactions on MDT/OST into synchronous, I'll fix it together.

Comment by Gerrit Updater [ 11/Nov/16 ]

Niu Yawei (yawei.niu@intel.com) uploaded a new patch: http://review.whamcloud.com/23716
Subject: LU-8826 recovery: sync client data deletion
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 13bf499bdad68d8bda84bb603cc7545d925f842e

Comment by Gerrit Updater [ 17/Dec/16 ]

Oleg Drokin (oleg.drokin@intel.com) merged in patch https://review.whamcloud.com/23716/
Subject: LU-8826 recovery: don't shrink recovery hard time
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: f0d5693584bc70913541201c54c3666a473715b2

Comment by Peter Jones [ 17/Dec/16 ]

Landed for 2.10

Generated at Sat Feb 10 02:20:53 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.