[Gluster-users] brick become to offline

[M S Vishwanath Bhat]

On 15 June 2015 at 07:38, 何亦军 <heyijun at greatwall.com.cn> wrote:

>  Hi,
>
>
>
>          I have done gluster upgrade to 3.7.1, then after downgrade back
> 3.6.2.    I don’t know something wrong, the server brick become to
> offline.  How to fix that problem?  Thanks so much.
>
>
>
>     [root at gwgfs01 ~]# gluster volume status
>
> Status of volume: vol01
>
> Gluster process                                         Port    Online  Pid
>
>
> ------------------------------------------------------------------------------
>
> Brick gwgfs01:/data/brick1/vol01                        N/A     N       N/A
>
> Brick gwgfs03:/data/brick2/vol01                        49152   Y
> 17566
>
> Brick gwgfs01:/data/brick2/vol01                        N/A     N       N/A
>
> Brick gwgfs02:/data/brick2/vol01                        49152   Y
> 4109
>
> Brick gwgfs02:/data/brick1/vol01                        49153   Y
> 4121
>
> Brick gwgfs03:/data/brick1/vol01                        49153   Y
> 17623
>
> Self-heal Daemon on localhost                           N/A     Y
> 12720
>
> Quota Daemon on localhost                               N/A     Y
> 12727
>
> Self-heal Daemon on gwgfs02                             N/A     Y
> 4412
>
> Quota Daemon on gwgfs02                                 N/A     Y
> 4422
>
> Self-heal Daemon on gwgfs03                             N/A     Y
> 17642
>
> Quota Daemon on gwgfs03                                 N/A     Y
> 17652
>
>
>
> Task Status of Volume vol01
>
>
> ------------------------------------------------------------------------------
>
> Task                 : Rebalance
>
> ID                   : 0bb30902-e7d3-4b2f-9c83-b708ebbad592
>
> Status               : failed
>
>
>
>
>
> some log in data-brick1-vol01.log :
>
>
>
> [2015-06-15 02:06:54.757968] I [MSGID: 100030] [glusterfsd.c:2018:main]
> 0-/usr/sbin/glusterfsd: Started running /usr/sbin/glusterfsd version 3.6.2
> (args: /usr/sbin/glusterfsd -s gwgfs01 --volfile-id
> vol01.gwgfs01.data-brick1-vol01 -p
> /var/lib/glusterd/vols/vol01/run/gwgfs01-data-brick1-vol01.pid -S
> /var/run/ecf2e5c591c01357cf33cbaf3b700bc6.socket --brick-name
> /data/brick1/vol01 -l /var/log/glusterfs/bricks/data-brick1-vol01.log
> --xlator-option *-posix.glusterd-uuid=b80f71d0-6944-4236-af96-e272a1f7e739
> --brick-port 49152 --xlator-option vol01-server.listen-port=49152)
>
> [2015-06-15 02:06:56.808733] W [xlator.c:191:xlator_dynload] 0-xlator:
> /usr/lib64/glusterfs/3.6.2/xlator/features/trash.so: cannot open shared
> object file: No such file or directory
>
> [2015-06-15 02:06:56.808793] E [graph.y:212:volume_type] 0-parser: Volume
> 'vol01-trash', line 9: type 'features/trash' is not valid or not found on
> this machine
>
> [2015-06-15 02:06:56.808896] E [graph.y:321:volume_end] 0-parser: "type"
> not specified for volume vol01-trash
>
> [2015-06-15 02:06:56.809044] E [MSGID: 100026]
> [glusterfsd.c:1892:glusterfs_process_volfp] 0-: failed to construct the
> graph
>
> [2015-06-15 02:06:56.809369] W [glusterfsd.c:1194:cleanup_and_exit] (-->
> 0-: received signum (0), shutting down
>

I'm not an expert here in analysing the logs, But looks like trash xlator
(which is present only in 3.7.x) is causing problems. Because the feature
is unavailable in 3.6.x versions. It's picking up the wrong volfile maybe?

Also did you downgrade the whole setup or just the part of the cluster?

Cheers,
Vishwanath




> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20150615/0dcebc27/attachment.html>