Funny you should say that. We are using an external url using https via
a load balancer. I switched over to the internal endpoint and it has
made a very big improvement.
Grant
On 03/06/16 13:52, David Medberry wrote:
On Fri, Jun 3, 2016 at 6:22 AM, Grant Morley <gr...@absolutedevops.io
<mailto:gr...@absolutedevops.io>> wrote:
Turns out is in an issue with the keystone tokens that are timing
out when the snapshot is taking place.
Yes, this is exactly the problem we have seen--token timeout. I didn't
see the size of this image but we've had that issue on any larger than
40G with a two hour token time. In some (our) environments, you will
also see issues when the external url is using https via a load
balancer. Switching to the internal os-endpoint-type made a sizable
improvement/speedup.
--
Grant Morley
Cloud Lead
Absolute DevOps Ltd
Units H, J & K, Gateway 1000, Whittle Way, Stevenage, Herts, SG1 2FP
www.absolutedevops.io <http://www.absolutedevops.io/>
gr...@absolutedevops.io <mailto:grant@absolutedevops.i> 0845 874 0580
_______________________________________________
OpenStack-operators mailing list
OpenStack-operators@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
