[LU-5642] getxattr failing with EIO Created: 19/Sep/14  Updated: 14/Oct/14  Resolved: 14/Oct/14

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.5.3
Fix Version/s: Lustre 2.7.0

Type: Bug Priority: Major
Reporter: Henri Doreau (Inactive) Assignee: James Nunez (Inactive)
Resolution: Fixed Votes: 0
Labels: None

Severity: 3
Rank (Obsolete): 15806

 Description   

getxattr() calls occasionally fail with EIO. We believe that this is due to a race between concurrent calls on the same file (see ll_xattr_cache_refill):

/* Matched but no cache? Cancelled on error by a parallel refill. */
if (unlikely(req == NULL)) {
CDEBUG(D_CACHE, "cancelled by a parallel getxattr\n");
GOTO(out_maybe_drop, rc = -EIO);
}

This should be adjusted so that both calls end consistenty.



 Comments   
Comment by Peter Jones [ 20/Sep/14 ]

James

Could you please look into this one?

Thanks

Peter

Comment by Sebastien Buisson (Inactive) [ 02/Oct/14 ]

Hi there,

I think this issue could be a duplicate of LU-3544.

Comment by Bruno Travouillon (Inactive) [ 02/Oct/14 ]

Not a dup, but patch http://review.whamcloud.com/#/c/10693/ can solve the issue. We are testing it right now.

Comment by James Nunez (Inactive) [ 09/Oct/14 ]

Bruno,
Any results on if http://review.whamcloud.com/#/c/10693/ solves the getxattr failing with EIO problem?

Thanks.

Comment by Bruno Travouillon (Inactive) [ 10/Oct/14 ]

James,

Yes, problem solved with the patch.

Comment by James Nunez (Inactive) [ 14/Oct/14 ]

Thank you for the feedback, Bruno.

I'm resolving this ticket because patch http://review.whamcloud.com/#/c/10693/ fixes the issue.

Generated at Sat Feb 10 01:53:16 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.