[Bugs] [Bug 1147420] New: geo-rep: file with same name has diffrent gfid in master and slave

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1147420

            Bug ID: 1147420
           Summary: geo-rep: file with same name has diffrent gfid in
                    master and slave
           Product: GlusterFS
           Version: 3.6.0
         Component: geo-replication
          Severity: high
          Priority: high
          Assignee: gluster-bugs at redhat.com
          Reporter: avishwan at redhat.com
                CC: bugs at gluster.org
        Depends On: 1060683, 1143853, 1146369



+++ This bug was initially created as a clone of Bug #1143853 +++

Description of problem:
Editors like vi, RENAMES original file to temp name and then create the new
file for data modification. Geo-rep fails to sync when temp files RENAME
involved.

How reproducible:
Always.

Steps to Reproduce:
1. Create a file in master mount
2. Open the file in vi editor, modify and save it.

Actual results:
Geo-rep fails to sync the change, and file GFID in master will not match with
GFID of slave.

Expected results:
GFID should match in both master and slave, and geo-rep should sync the
changes.

--- Additional comment from Anand Avati on 2014-09-18 04:22:20 EDT ---

REVIEW: http://review.gluster.org/8761 (geo-rep: fix same file different gfid
in master and slave) posted (#1) for review on master by Aravinda VK
(avishwan at redhat.com)

--- Additional comment from Anand Avati on 2014-09-26 03:19:02 EDT ---

COMMIT: http://review.gluster.org/8761 committed in master by Venky Shankar
(vshankar at redhat.com) 
------
commit 3c29c50cf60135245290133fbfed11aa3cf01e39
Author: Aravinda VK <avishwan at redhat.com>
Date:   Wed Sep 17 12:59:52 2014 +0530

    geo-rep: fix same file different gfid in master and slave

    While processing RENAME in changelog, if the file is unlinked
    in master, then geo-rep was sending UNLINK to slave instead of
    RENAME.

    If rsync job fails if one of the file failed to sync in the job.
    This patch adds logic to remove GFID from data list if the same
    changelog has UNLINK entry for it after the DATA. Or it removes
    those GFIDs during retry of changelogs processing.

    BUG: 1143853
    Change-Id: I982dc976397cd0ab676bb912583f66a28f821926
    Signed-off-by: Aravinda VK <avishwan at redhat.com>
    Reviewed-on: http://review.gluster.org/8761
    Tested-by: Gluster Build System <jenkins at build.gluster.com>
    Reviewed-by: Jeff Darcy <jdarcy at redhat.com>
    Reviewed-by: Kotresh HR <khiremat at redhat.com>
    Reviewed-by: Venky Shankar <vshankar at redhat.com>
    Tested-by: Venky Shankar <vshankar at redhat.com>

--- Additional comment from Anand Avati on 2014-09-29 04:30:00 EDT ---

REVIEW: http://review.gluster.org/8879 (geo-rep: fix same file different gfid
in master and slave) posted (#1) for review on release-3.6 by Aravinda VK
(avishwan at redhat.com)


Referenced Bugs:

https://bugzilla.redhat.com/show_bug.cgi?id=1143853
[Bug 1143853] geo-rep: file with same name has diffrent gfid in master and
slave
-- 
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=SAcHXbS9x5&a=cc_unsubscribe