Gluster users - Oct 2015 - Please help in tuning Gluster FS in 2 node application system

Hi,

I have created Volume for Gluster FS in distributed replicated mode-

 gluster volume info gvol0

Volume Name: gvol0
Type: Replicate
Volume ID: 550c1759-f708-4199-8273-4aa2e6b1ce56
Status: Started
Number of Bricks: 1 x 2 = 2
Transport-type: tcp
Bricks:
Brick1: glust2:/var/lib/gvol0/mediation
Brick2: glust3:/var/lib/gvol0/mediation
Options Reconfigured:
diagnostics.latency-measurement: off
performance.write-behind: off
nfs.rpc-auth-allow: 10.61.*
nfs.export-volumes: off
nfs.addr-namelookup: off
nfs.disable: on
auth.allow: 10.61.*
performance.readdir-ahead: on


But Sync time is too much between 2 nodes thats why my processing
performance is very bad.

My Second node is getting same data from first node very late thats why
second node is not able to process and remain in idle mode.

Please provide tuning/syncing parameters.

Please provide good values for these parameters for small and big size
files processing.

Please provide response ASAP. (need urgently)

Thank you

Regards
Tarkesh
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://www.gluster.org/pipermail/gluster-users/attachments/20151008/d6029b47/attachment.html>

More details needed.
- Hardware on each Node, CPU, RAM, Disks
- Op. Sys. & Version
- Details of Disk setup, Filing System type, Raid Levels, Cache
- Network Setuip, speed, Bonding 
- What benchmarks have you run? What read/write and IOPS are yo seeing?



Sent from Mail for Windows 10



From: M.Tarkeshwar Rao
Sent: Friday, 9 October 2015 3:43 AM
To: gluster-users at gluster.org
Cc: Priyanka Agrawal
Subject: [Gluster-users] Please help in tuning Gluster FS in 2 nodeapplication
system


Hi,

I have created Volume for Gluster FS in distributed replicated mode-

?gluster volume info gvol0

Volume Name: gvol0
Type: Replicate
Volume ID: 550c1759-f708-4199-8273-4aa2e6b1ce56
Status: Started
Number of Bricks: 1 x 2 = 2
Transport-type: tcp
Bricks:
Brick1: glust2:/var/lib/gvol0/mediation
Brick2: glust3:/var/lib/gvol0/mediation
Options Reconfigured:
diagnostics.latency-measurement: off
performance.write-behind: off
nfs.rpc-auth-allow: 10.61.*
nfs.export-volumes: off
nfs.addr-namelookup: off
nfs.disable: on
auth.allow: 10.61.*
performance.readdir-ahead: on


But Sync time is too much between 2 nodes thats why my processing performance is
very bad.

My Second node is getting same data from first node very late thats why second
node is not able to process and remain in idle mode.

Please provide tuning/syncing parameters.

Please provide good values for these parameters for small and big size files
processing.

Please provide response ASAP. (need urgently)

Thank you

Regards
Tarkesh


-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://www.gluster.org/pipermail/gluster-users/attachments/20151009/c8284844/attachment.html>