[Gluster-users] Files not synced after peer reboot

Pranith Kumar. Karampuri pranithk at gluster.com
Tue Mar 22 06:01:21 UTC 2011

Please refer to http://europe.gluster.org/community/documentation/index.php/Gluster_3.1:_Triggering_Self-Heal_on_Replicate


----- Original Message -----
From: "Panos Kontopoulos" <panos at kontopoulos.org>
To: gluster-users at gluster.org
Sent: Monday, March 21, 2011 12:53:00 PM
Subject: [Gluster-users] Files not synced after peer reboot

I am testing GlusterFS 3.1 on a EC2 environment with 2 servers in a replica mode 

sudo gluster volume create gluster-volume replica 2 transport tcp ip-10-227-166-178:/export/brick01 ip-10-48-79-22:/export/brick02

I have been connected to the file system using a third client, created a test file which appears in both servers. 
I reboot my GlusterFS 2nd server and at the same time I change contents of my test file which appears ok in the 1st still running server. 
When the 2nd server comes live, I run sudo /etc/init.d/glusterd start and remount my drive - brick

However the contents of test file in 2nd server are still the old ones before reboot. 
When I open the test file from the client (which is the correct latest one, BTW) I update  the file and save it, both files are again synced. 
Is this how GlusterFS is working by default ?
Is there a way to re-sync files without updating the actual file ?

Thanx in advance, 
Panos Kontopoulos

Gluster-users mailing list
Gluster-users at gluster.org

More information about the Gluster-users mailing list