[Openstack-operators] Storage backend for glance

Robert Starmer robert at kumul.us
Wed Jan 27 18:42:58 UTC 2016

Glusterfs backend works great for shared glance, and can be configured for
a bit of redundancy at the disk level (rather than non distributed NFS,
which needs the NFS server to be present), much like the Ceph model Kevin
suggests.  Is your database also resiliant (e.g. some form of mysql
replication), then the glance-api services are effectively stateless, and
you've protected disk backend (gluster/ceph), and image catalog (mysql).


On Wed, Jan 27, 2016 at 8:28 AM, Fox, Kevin M <Kevin.Fox at pnnl.gov> wrote:

> ceph would work pretty well for that use case too. We've run a ceph with
> two ost's, with the replication set to 2, to back both cinder and glance
> for HA. Nothing complicated needed to get it working. Less complicated then
> drbd I think. You can then also easily scale it out as needed.
> Thanks,
> Kevin
> ________________________________________
> From: Hauke Bruno Wollentin [Hauke-Bruno.Wollentin at innovo-cloud.de]
> Sent: Wednesday, January 27, 2016 7:04 AM
> To: Sławek Kapłoński; openstack-operators at lists.openstack.org
> Subject: Re: [Openstack-operators] Storage backend for glance
> Hi Slawek,
> we use a shared NFS Export to save images to Glance. That enables HA in
> (imho) the simplest way.
> For your setting you could use something like a hourly/daily/whenever
> rsync job and set the 'second' Glance node to passive/standby in the load
> balancing. Also it will be possible to run some kind of cluster between the
> 2 nodes, like DRBD or GlusterFS (but that would be a bit overpowered maybe).
> cheers,
> hauke
> -----Ursprüngliche Nachricht-----
> Von: Sławek Kapłoński [mailto:slawek at kaplonski.pl]
> Gesendet: Mittwoch, 27. Januar 2016 15:23
> An: openstack-operators at lists.openstack.org
> Betreff: [Openstack-operators] Storage backend for glance
> Hello,
> I want to install Openstack with at least two glance nodes (to have HA)
> but with local filesystem as glance storage. Is it possible to use
> something like that in setup with two glance nodes? Maybe someone of You
> already have something like that?
> I'm asking because AFAIK is image will be stored on one glance server and
> nova-compute will ask other glance host to download image then image will
> not be available to download and instance will be in ERROR state.
> So maybe someone used it somehow in similar setup (maybe some NFS or
> something like that?). What are You experience with it?
> --
> Best regards / Pozdrawiam
> Sławek Kapłoński
> slawek at kaplonski.pl
> _______________________________________________
> OpenStack-operators mailing list
> OpenStack-operators at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
> _______________________________________________
> OpenStack-operators mailing list
> OpenStack-operators at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-operators/attachments/20160127/73178976/attachment.html>

More information about the OpenStack-operators mailing list