Gluster users - Dec 2010 - Two bricks in one replicated volume, different sizes, what to do?

Heya,

I have one replicate gluster volume made of two bricks. Both servers are live
and connected right now but to my surprise, they are reporting different
Used/Available statistics. I don't seem to be running into any issues on the
client ends but I'd like to know what's the safest way to sync all the
files so that both volumes have the same content, without causing downtime to my
file servers (preferably).

I tried doing "ls -lahR * .*" on one of the client but it didn't
get everything copied over.

Thanks!

Server 1 (where clients connect in fstab):

Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/md0             412847280   7590928 384284936   2% /ebs_raid

gluster> peer status
Number of Peers: 1

Hostname: 1-gfs
Uuid: b3fbe1b0-d2f4-4ae8-975a-921643e56894
State: Peer in Cluster (Connected)
gluster> volume info

Volume Name: 1-gfs-edgar
Type: Replicate
Status: Started
Number of Bricks: 2
Transport-type: tcp
Bricks:
Brick1: 1-gfs:/ebs_raid
Brick2: 2-gfs:/ebs_raid
Options Reconfigured:
performance.quick-read: on
performance.cache-max-file-size: 512KB
performance.cache-size: 512MB
performance.stat-prefetch: on
network.frame-timeout: 60


Server 2:

Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/md0             412847280   4080260 387795604   2% /ebs_raid

gluster> peer status
Number of Peers: 1

Hostname: 2-gfs
Uuid: 61915cd1-2735-4c0e-ba2f-829aa6bc4a9a
State: Peer in Cluster (Connected)
gluster> volume info

Volume Name: 1-gfs-edgar
Type: Replicate
Status: Started
Number of Bricks: 2
Transport-type: tcp
Bricks:
Brick1: 1-gfs:/ebs_raid
Brick2: 2-gfs:/ebs_raid
Options Reconfigured:
performance.quick-read: on
performance.cache-max-file-size: 512KB
performance.cache-size: 512MB
performance.stat-prefetch: on
network.frame-timeout: 60

--
Pierre-Luc Brunet

On Tuesday 21 December 2010 09:27 PM, Pierre-Luc Brunet
wrote:
> Heya,
>
> I have one replicate gluster volume made of two bricks. Both servers are
live and connected right now but to my surprise, they are reporting different
Used/Available statistics. I don't seem to be running into any issues on the
client ends but I'd like to know what's the safest way to sync all the
files so that both volumes have the same content, without causing downtime to my
file servers (preferably).
>
> I tried doing "ls -lahR * .*" on one of the client but it
didn't get everything copied over.
>
>    
Can you please trigger self-heal by following instructions at:

http://www.gluster.com/community/documentation/index.php/Gluster_3.1:_Triggering_Self-Heal_on_Replicate

Regards,
Vijay