[Gluster-users] Restoring glusterfs cluster.

[Tao Lin]

Hello, deer glustefs experts:
I've found that my cluster of only two nodes had ran into a big trouble.
Here are some information about my glusterfs.

[root at yq35 ~]# gluster --version
glusterfs 3.2.6 built on Mar 22 2012 10:44:28
Repository revision: git://git.gluster.com/glusterfs.git
Copyright (c) 2006-2011 Gluster Inc. <http://www.gluster.com>
GlusterFS comes with ABSOLUTELY NO WARRANTY.
You may redistribute copies of GlusterFS under the terms of the GNU General
Public License.


[root at yq35 ~]# gluster volume info

Volume Name: volume1
Type: Replicate
Status: Started
Number of Bricks: 2
Transport-type: tcp
Bricks:
Brick1: 10.67.15.33:/media/data1/brick1
Brick2: 10.67.15.35:/media/data1/brick1
Options Reconfigured:
features.quota: off



When i use 'peer status' command, one of the nodes shows strange infomation
as following:

[root at yq33 ~]# gluster peer status
Number of Peers: 2

Hostname: 10.67.15.33
Uuid: c213f7ab-18c1-40e1-85c8-dd7ae97fad03
State: Peer in Cluster (Connected)

Hostname: 10.67.15.35
Uuid: 98949cd6-2b61-4ba8-8b67-76d8b58d4ce8
State: Peer in Cluster (Connected)

tip:yq33 has the address of 10.67.15.33


The same command, however, shows diffrent information on another node:

[root at yq35 ~]# gluster peer status
Number of Peers: 1

Hostname: 10.67.15.33
Uuid: c213f7ab-18c1-40e1-85c8-dd7ae97fad03
State: Accepted peer request (Connected)

tip:yq35 has the address of 10.67.15.35

I also find that there are two files in glusterfs' work dir on yq33
[root at yq33 ~]# ll /etc/glusterd/peers/
total 8
-rw-r--r-- 1 root root 72 Oct 25 00:11 98949cd6-2b61-4ba8-8b67-76d8b58d4ce8
-rw-r--r-- 1 root root 72 Aug 22 17:38 c213f7ab-18c1-40e1-85c8-dd7ae97fad03

[root at yq33 ~]# cat /etc/glusterd/peers/c213f7ab-18c1-40e1-85c8-dd7ae97fad03
uuid=c213f7ab-18c1-40e1-85c8-dd7ae97fad03
state=3
hostname1=10.67.15.33

[root at yq33 ~]# cat /etc/glusterd/peers/98949cd6-2b61-4ba8-8b67-76d8b58d4ce8
uuid=98949cd6-2b61-4ba8-8b67-76d8b58d4ce8
state=3
hostname1=10.67.15.35


but there is only one file on yq35
[root at yq35 ~]# ll /etc/glusterd/peers/c213f7ab-18c1-40e1-85c8-dd7ae97fad03
-rw-r--r-- 1 root root 72 Oct 25 00:11
/etc/glusterd/peers/c213f7ab-18c1-40e1-85c8-dd7ae97fad03

[root at yq35 ~]# cat /etc/glusterd/peers/c213f7ab-18c1-40e1-85c8-dd7ae97fad03
uuid=c213f7ab-18c1-40e1-85c8-dd7ae97fad03
state=4
hostname1=10.67.15.33

My question is:
Can i just modify these files (modify 'state' to 3) and delete the file
c213f7ab-18c1-40e1-85c8-dd7ae97fad03 on yq33 so that my cluster can be
restored, and how?
Thanks a lot.

Regards.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20121025/baf38e74/attachment.html>