[LU-13513] osp_precreate_reserve() should not take d->opd_pre_status == -EIO as fatal Created: 04/May/20  Updated: 12/Apr/23  Resolved: 26/Feb/21

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: Lustre 2.15.0

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

Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

When OST_CREATE (not resendable rpc) sent by precreate thread fails
with network error, osp_pre_update_status() sets d->opd_pre_status to
EIO. osp_precreate_reserve() considers EIO as fatal and does not wait
for another attempt from precreate thread. That may make
mdt_intent_open to return ENOSPC confusing a caller.
ENOSPC comes from lod_alloc_rr().

 Comments   
Comment by Gerrit Updater [ 04/May/20 ]

Vladimir Saveliev (c17830@cray.com) uploaded a new patch: https://review.whamcloud.com/38472
Subject: LU-13513 osp: let precreate thread repeat after network error
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: ee5cf0ad2e89a2749588066cf7d05eed457a1808

Comment by Gerrit Updater [ 26/Feb/21 ]

Oleg Drokin (green@whamcloud.com) merged in patch https://review.whamcloud.com/38472/
Subject: LU-13513 osp: make neterr not fatal for precreate_reserve
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 4bba67075aa3d8739d8ca99642ff2b2836774479

Comment by Peter Jones [ 26/Feb/21 ]

Landed for 2.15

Generated at Sat Feb 10 03:01:55 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.