Open Stack

Hi Darragh,

Thanks for the input! I'm using Ubuntu 12.04 with the latest 3.5.0 kernel
from stable with Grizzly.

I followed a suggestion to upgrade OVS from stable to the latest Havana
release under the Cloud Archive. This moved me from 1.4.0 -> 1.10.2. The
migration was relatively seamless, and has fixed the issue.

I'm not sure if this was a bug in 1.4.0 (which maybe the veth setting would
have fixed), however the upgrade has completely fixed my latency issues.


On Wed, Oct 16, 2013 at 5:47 AM, Darragh O'Reilly <
dara2002-openstack at yahoo.com> wrote:

> Jacob,
>
> what operating and kernel are you using?
> what ovs datapath (dpkg -l openv\*) ?
> what version of openstack?
>
> Maybe it is an issue getting in and/or out of the router namespace. Try
> running tcpdump on the qr-xxxxxx-xx (from namespace) and the physical
> interface in br-ex (from root namespace), and do a ping from namespace to
> outside. Then compare the timestamps on the two tcpdumps. Then try a ping
> from outside to in.
>
>
> Jame Page reported a router performance issue recently, and setting
> ovs_use_veth=true somehow fixed it.
>
> http://lists.openstack.org/pipermail/openstack/2013-October/001811.html
>
>
> Re, Darragh.
>
>
> _______________________________________________
> 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/20131016/a76bc013/attachment.html>