I've been having one heck of a time for nearly the entire time I've been
running ovirt with disk IO performance.  I've tried a variety of things,
I've posted to this list for help several times, and it sounds like in most
cases the problems are due to design decisions and such.

My cluster has been devolving into nearly unusable performance, and I
believe its mostly disk IO related.  I'm currently using FreeNAS as my
primary VM storage (via NFS), but now it too is performing slowly (it
started out reasonable, but slowly devolved for unknown reasons).

I'm ready to switch back to gluster if I can get specific recommendations
as to what I need to do to make it work.  I feel like I've been trying
random things, and sinking money into this to try and make it work, but
nothing has really fixed the problem.

I have 3 Dell R610 servers with 750GB SSDs as their primary drive.  I had
used some Seagate SSHDs, but the internal Dell DRAC raid controller (which
had been configured to pass them through as a single disk volume, but still
wasn't really JBOD), but it started silently failing them, and causing
major issues for gluster.  I think the DRAC just doesn't like those HDDs.

I can put some real spinning disks in; perhaps a RAID-1 pair of 2TB?  These
servers only take 2.5" hdd's, so that greatly limits my options.

I'm sure others out there are using Dell R610 servers...what do  you use
for storage?  How does it perform?  What do I need to do to get this
cluster actually usable again?  Are PERC-6i storage controllers usable?
I'm not even sure where to go troubleshooting now...everything is so

BTW: I had a small data volume on the SSDs, and the gluster performance on
those was pretty poor.  performance of the hosted engine is pretty poor
still, and it is still on the SSDs.
Users mailing list -- users@ovirt.org
To unsubscribe send an email to users-le...@ovirt.org
Privacy Statement: https://www.ovirt.org/site/privacy-policy/
oVirt Code of Conduct: 
List Archives: 

Reply via email to