[Gluster-users] GlusterFS + Virtual Machine
driver at megahappy.net
Mon Apr 9 20:52:04 UTC 2012
The VM's might lockup for 1-2min while they get copied over -
depending on how big the qcow2 files are. Typically our qcow2 files
are around 4Gig. Otherwise they are still useable (processes not doing
IO will continue to run fine).
>From the VM, there will be a kernel traceback/oops into logs that IO
timed out. Depending on how your applications handle that timeout
really depends on your use case.
As for performance I've posted some stats in the past. Check the archives.
On Sun, Apr 8, 2012 at 5:30 AM, fc lists <fclists at pr-z.info> wrote:
> may i ask what kind of setup you have for the gluster and what kind of
> performances you get?
> I just started testing glusterfs out, my performances are not really inline
> for something like VM storage backend , most probably my fault.
> Just curious to know from someone currently using it.
> Also, how bad is the "block" issue on 3.2.x?
> I went straight for 3.3beta2 being just a test.
> On Tue, Apr 3, 2012 at 4:35 PM, Bryan Whitehead <driver at megahappy.net>
>> Yes. I currently use glusterfs to replicate qcow2 files with running
>> Annoying issue with the 3.2.x series is the entire qcow2 file can get
>> blocked if a rebuild is needed. the 3.3 series (still in beta) fixes
>> On Tue, Apr 3, 2012 at 2:48 AM, <swb3 at gmx.de> wrote:
>> > Hey Guys,
>> > I want to ask if its possible to use my GlusterFS Share for an Virtual
>> > Machine Storage.
>> > Thanks for your help
>> > --
>> > NEU: FreePhone 3-fach-Flat mit kostenlosem Smartphone!
>> > Jetzt informieren: http://mobile.1und1.de/?ac=OM.PW.PW003K20328T7073a
>> > _______________________________________________
>> > Gluster-users mailing list
>> > Gluster-users at gluster.org
>> > http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
>> Gluster-users mailing list
>> Gluster-users at gluster.org
More information about the Gluster-users