Mike Dorman wrote:
We do have a backlog story to investigate this more deeply, we just have not
had the time to do it yet. For us, it’s been easier/faster to add more
hardware to conductor to get over the hump temporarily.
We kind of have that work earmarked for after the Liberty upgrade, in hopes
that maybe it’ll be fixed there.
If anybody else has done even some trivial troubleshooting already, it’d be
great to get that info as a starting point. I.e. which specific calls to
conductor are causing the load, etc.
Mike
+1 I think we in the #openstack-performance channel really need to
investigate this, because it really worries me personally from hearing
many many rumors about how the remote conductor falls over. Please join
there and we can try to work through a plan to figure out what to do
about this situation. It would be great if the nova people also joined
there (because in the end, likely something in nova will need to be
fixed/changed/something else to resolve what appears to be a problem for
many operators).
On 11/12/15, 9:36 AM, "Matt Riedemann"<[email protected]> wrote:
On 11/12/2015 8:15 AM, Andrew Laski wrote:
On 11/12/15 at 09:31pm, Yaguang Tang wrote:
There are still performance issue in large deployment to use
nova-conductor, how do we make this decision to deprecate local mode?
I see
the patch is merged only three days after submit.
There is no timeline for removal of local mode at this time for exactly
the reason you mention. It will be removed no earlier than the N cycle
though. The deprecation is intended to signal that deployers should
start thinking about how to move to remote conductor and not expect
local mode to be around forever. And it was always intended to be
temporary while deployments transitioned.
It would be helpful if you could report the performance issues that you
are seeing so that remote conductor can be improved to accommodate
deployments of all sizes.
Yes, also because this came up in the operators list:
http://lists.openstack.org/pipermail/openstack-operators/2015-October/008577.html
Conductor was called out but I'm not sure if people have dug into the
main issues. I had replied to that other thread if we wanted to pick
this up there.
On Thu, Nov 12, 2015 at 10:51 AM, Matt Riedemann
<[email protected]
wrote:
Details are in the change:
https://review.openstack.org/#/c/242168/
--
Thanks,
Matt Riedemann
_______________________________________________
OpenStack-operators mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
--
Yaguang Tang
Technical Support, Mirantis China
*Phone*: +86 15210946968
_______________________________________________
OpenStack-operators mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
_______________________________________________
OpenStack-operators mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
--
Thanks,
Matt Riedemann
_______________________________________________
OpenStack-operators mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
_______________________________________________
OpenStack-operators mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
_______________________________________________
OpenStack-operators mailing list
[email protected]
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
