Project

General

Profile

Actions
Copy link

Bug #3738

closed

kclient fsx truncate/write multi-client race

Added by Sam Lang over 11 years ago. Updated over 10 years ago.

Status:
Resolved
Priority:
Normal
Assignee:
Sam Lang
Category:
-
Target version:
-
% Done:

0%

Source:
Development
Tags:
Backport:
Regression:
Severity:
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
Labels (FS):
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

This bug is similar to #3681, but occurs only in the non-exclusive case (multiple clients), where a truncate doesn't update i_size in the kclient, sends a setattr request to the mds, which races with a cap update triggered by a write completion.

I'm not 100% sure this bug exists, but it seems likely. Needs to first be verified with fsx-mpi.

A possible fix is to delay the truncate request until the caps are no longer dirty.

Related issues 1 (0 open — 1 closed)

Follows CephFS - Bug #3681: kclient fsx fails nightlyResolved12/26/2012

Actions
Actions
Copy link
 #1

Updated by Sage Weil over 11 years ago

  • Project changed from Linux kernel client to CephFS
  • Category deleted (fs/ceph)
Actions
Copy link
 #2

Updated by Zheng Yan over 10 years ago

  • Status changed from New to Resolved

now mds revokes Fw before doing truncate

Actions
Copy link
 #3

Updated by Greg Farnum over 10 years ago

Is this a side effect of 3c3b2ceb03e7294704f5bf3e1e420012a0166585, or some other patch? Please reference them when closing tickets. :)

Actions
Copy link
 #4

Updated by Zheng Yan over 10 years ago

by commit 755581977c2bc9eb81c9d9d955024cbedded2161

Actions
Copy link

Also available in: Atom PDF