[Bugs] [Bug 1434296] New: [Disperse] Metadata version is not healing when a brick is down

bugzilla at redhat.com bugzilla at redhat.com
Tue Mar 21 08:27:03 UTC 2017


https://bugzilla.redhat.com/show_bug.cgi?id=1434296

            Bug ID: 1434296
           Summary: [Disperse] Metadata version is not healing when a
                    brick is down
           Product: GlusterFS
           Version: 3.10
         Component: disperse
          Assignee: bugs at gluster.org
          Reporter: sheggodu at redhat.com
                CC: aspandey at redhat.com, bugs at gluster.org,
                    pkarampu at redhat.com, xhernandez at datalab.es
        Depends On: 1425703



+++ This bug was initially created as a clone of Bug #1425703 +++

Description of problem:

When a brick-1 is down and we create a file on mount point, index entries will
be created. Now when this brick is UP and the other brick goes down, heal
starts on the bric-1 and it heal all the data and also its version. However on
the metadata version on the brick-1 is not getting healed and remains 0




Version-Release number of selected component (if applicable):

[root at apandey glusterfs]# gluster --version
glusterfs 3.11dev
Repository revision: git://git.gluster.org/glusterfs.git
Copyright (c) 2006-2016 Red Hat, Inc. <https://www.gluster.org/>
GlusterFS comes with ABSOLUTELY NO WARRANTY.
It is licensed to you under your choice of the GNU Lesser
General Public License, version 3 or any later version (LGPLv3
or later), or the GNU General Public License, version 2 (GPLv2),
in all cases as published by the Free Software Foundation

How reproducible:
100%

Steps to Reproduce:
1. Create a (4+2) volume and mount it.
2. Kill a brick, brick-1, and create 10 files on mount point.
3. Start the volume using force and kill other brick, brick-2, immediately.     
4. start index heal and give enough time to heal.
5. At this point all the files on brick-1 should have correct version and size
as with other 4 UP bricks and all the files should be healed.

6. Check trusted.ec.{version,size}, it should be same on all the 5 up bricks

Actual results:
Step 6 shows metadata version on brick-1 has not been healed.

Expected results:
Step 6 should show metadata version on brick-1 has been healed and similar to
other 4 good UP bricks.

Additional info:

--- Additional comment from Worker Ant on 2017-02-27 05:27:27 EST ---

REVIEW: https://review.gluster.org/16772 (cluster/ec: Metadata healing fails to
update the version) posted (#1) for review on master by Sunil Kumar Acharya
(sheggodu at redhat.com)

--- Additional comment from Worker Ant on 2017-03-07 07:57:29 EST ---

REVIEW: https://review.gluster.org/16772 (cluster/ec: Metadata healing fails to
update the version) posted (#2) for review on master by Sunil Kumar Acharya
(sheggodu at redhat.com)

--- Additional comment from Worker Ant on 2017-03-15 10:29:05 EDT ---

REVIEW: https://review.gluster.org/16772 (cluster/ec: Metadata healing fails to
update the version) posted (#3) for review on master by Sunil Kumar Acharya
(sheggodu at redhat.com)

--- Additional comment from Worker Ant on 2017-03-16 06:16:41 EDT ---

REVIEW: https://review.gluster.org/16772 (cluster/ec: Metadata healing fails to
update the version) posted (#4) for review on master by Sunil Kumar Acharya
(sheggodu at redhat.com)

--- Additional comment from Worker Ant on 2017-03-16 07:34:28 EDT ---

REVIEW: https://review.gluster.org/16772 (cluster/ec: Metadata healing fails to
update the version) posted (#5) for review on master by Sunil Kumar Acharya
(sheggodu at redhat.com)

--- Additional comment from Worker Ant on 2017-03-17 00:18:27 EDT ---

REVIEW: https://review.gluster.org/16772 (cluster/ec: Metadata healing fails to
update the version) posted (#6) for review on master by Sunil Kumar Acharya
(sheggodu at redhat.com)

--- Additional comment from Worker Ant on 2017-03-17 02:03:35 EDT ---

REVIEW: https://review.gluster.org/16772 (cluster/ec: Metadata healing fails to
update the version) posted (#7) for review on master by Sunil Kumar Acharya
(sheggodu at redhat.com)

--- Additional comment from Worker Ant on 2017-03-20 09:45:31 EDT ---

REVIEW: https://review.gluster.org/16772 (cluster/ec: Metadata healing fails to
update the version) posted (#8) for review on master by Sunil Kumar Acharya
(sheggodu at redhat.com)

--- Additional comment from Worker Ant on 2017-03-21 03:08:28 EDT ---

COMMIT: https://review.gluster.org/16772 committed in master by Xavier
Hernandez (xhernandez at datalab.es) 
------
commit 0c2253942dd0e6176918a7d530e56053a9f26e6d
Author: Sunil Kumar Acharya <sheggodu at redhat.com>
Date:   Mon Feb 27 15:35:17 2017 +0530

    cluster/ec: Metadata healing fails to update the version

    During meatadata heal, we were not updating the version
    though all the inode attributes were in sync.

    Updated the code to adjust version when all the inode
    attributes are in sync.

    BUG: 1425703
    Change-Id: I6723be3c5f748b286d4efdaf3c71e9d2087c7235
    Signed-off-by: Sunil Kumar Acharya <sheggodu at redhat.com>
    Reviewed-on: https://review.gluster.org/16772
    Smoke: Gluster Build System <jenkins at build.gluster.org>
    Reviewed-by: Xavier Hernandez <xhernandez at datalab.es>
    NetBSD-regression: NetBSD Build System <jenkins at build.gluster.org>
    Reviewed-by: Pranith Kumar Karampuri <pkarampu at redhat.com>
    CentOS-regression: Gluster Build System <jenkins at build.gluster.org>


Referenced Bugs:

https://bugzilla.redhat.com/show_bug.cgi?id=1425703
[Bug 1425703] [Disperse] Metadata version is not healing when a brick is
down
-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.


More information about the Bugs mailing list