Re: Test results and Performance Tuning efforts ...

[Paul Cuzner <pcuzner@xxxxxxxxxx>]

Hi,

*IF* your seeing crashes in glusterd, Atin sent out an workaround that needs to be applied to 3.7.x to avoid the issue (introduced with epoll)

add
# for epoll issue glusterd crash fix
    option ping-timeout 0
    option event-threads  1

to your glusterd.vol files (/etc/glusterfs/glusterd.vol)

On Tue, Oct 13, 2015 at 2:09 PM, Lindsay Mathieson <lindsay.mathieson@xxxxxxxxx> wrote:

On 13 October 2015 at 07:54, Ben Turner <bturner@xxxxxxxxxx> wrote:

Random IO has vastly improved with MT epoll introduced in 3.7, try a test on 3.7 with server and client event threads set to 4.  If you want to confirm this before you upgrade run top -H during your testing and look for a hot thread(single thread / CPU pegged at 100%).  If you see this during your runs on 3.6 then the MT epoll implementation in 3.7 will def help you out.

Not 100%, but it gets up to 90%

Oddly, the feature list seems to show it as being done in 3.6 (http://www.gluster.org/community/documentation/index.php/Features/Feature_Smallfile_Perf#multi-thread-epoll)

Seems to be causing 3.7.4 to crash regularly at the moment to :(

--

Lindsay

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://www.gluster.org/mailman/listinfo/gluster-users

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://www.gluster.org/mailman/listinfo/gluster-users