[Bugs] [Bug 1300978] New: I/O failure during a graph change followed by an option change.

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1300978

            Bug ID: 1300978
           Summary: I/O failure during a graph change followed by an
                    option change.
           Product: GlusterFS
           Version: 3.7.7
         Component: protocol
          Assignee: bugs at gluster.org
          Reporter: rkavunga at redhat.com
                CC: bugs at gluster.org, rgowdapp at redhat.com
        Depends On: 1300564
            Blocks: 1283957



+++ This bug was initially created as a clone of Bug #1300564 +++

Description of problem:

I/O failed for fuse mount when a graph changed followed by an option changed

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

mainline

How reproducible:

very inconsistent. It is a race

Steps to Reproduce:

1. Start i/o on fuse mount
2. Disable write-behind (just to have a graph change)
3. immediately enable trash (just to have an option change in server side)
4. I/o will fail with transport end point not connected

Actual results:

I/O failure

Expected results:

I/O should not fail

Additional info:

--- Additional comment from Vijay Bellur on 2016-01-21 02:27:12 EST ---

REVIEW: http://review.gluster.org/13271 (protocol/server: Race between
server_reconfigure and server_setvolume) posted (#1) for review on master by
mohammed rafi  kc (rkavunga at redhat.com)

--- Additional comment from Vijay Bellur on 2016-01-21 02:38:55 EST ---

REVIEW: http://review.gluster.org/13271 (protocol/server: Race between
server_reconfigure and server_setvolume) posted (#2) for review on master by
mohammed rafi  kc (rkavunga at redhat.com)

--- Additional comment from Vijay Bellur on 2016-01-21 03:27:03 EST ---

REVIEW: http://review.gluster.org/13271 (protocol/server: Race between
server_reconfigure and server_setvolume) posted (#3) for review on master by
mohammed rafi  kc (rkavunga at redhat.com)

--- Additional comment from Vijay Bellur on 2016-01-22 03:52:23 EST ---

COMMIT: http://review.gluster.org/13271 committed in master by Raghavendra G
(rgowdapp at redhat.com) 
------
commit 16f6579cb3e1214b1386fb530b8e16c8cbfdef33
Author: Mohammed Rafi KC <rkavunga at redhat.com>
Date:   Thu Jan 21 12:41:55 2016 +0530

    protocol/server: Race between server_reconfigure and server_setvolume

    During server_reconfigure we authenticate each connected clients
    against the current options. To do this authentication we store
    previous values in a dictionary during the connection establishment
    phase (server_setvolume). If the authentication fails during
    reconfigure then we will disconnect the transport.

    Here it introduce a race between server_setvolume and reconfugure.
    If a reconfigure called before doing a setvolume, the transport
    will be disconnected

    Change-Id: Icce2c28a171481327a06efd3901f8a5ee67b05ab
    BUG: 1300564
    Signed-off-by: Mohammed Rafi KC <rkavunga at redhat.com>
    Reviewed-on: http://review.gluster.org/13271
    Smoke: Gluster Build System <jenkins at build.gluster.com>
    Reviewed-by: Raghavendra G <rgowdapp at redhat.com>
    CentOS-regression: Gluster Build System <jenkins at build.gluster.com>
    Reviewed-by: Raghavendra Talur <rtalur at redhat.com>
    NetBSD-regression: NetBSD Build System <jenkins at build.gluster.org>


Referenced Bugs:

https://bugzilla.redhat.com/show_bug.cgi?id=1283957
[Bug 1283957] Data Tiering:tier volume status shows as in-progress on all
nodes of a cluster even if the node is not part of volume
https://bugzilla.redhat.com/show_bug.cgi?id=1300564
[Bug 1300564] I/O failure during a graph change followed by an option
change.
-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.