Open Stack

> To fix this, i must run "service neutron-plugin-openvswitch-agent restart" on compute-nodes that these
> instances is hosted.
> After restart neutron-agent, i can ping these instances normally.

I cannot help you with a "fix" but this is what I would do:

1) Check if there is a proper bug report which describes your issue
(ways to reproduce and what EXACTLY goes wrong, without this it is hard for devs to fix it)
2 ) If there is no bug report make sure it is not something wrong with your own setup.
My first guess would be that the flows to allow traffic are not created. (check with "ovs-ofctl dump-flows $bridge)
Now find out why:
I would enable (debug) logging on the neutron-agent.
Trigger the issue and look through the neutron logs to see if there is anything useful in there.
3) Depending on outcome of step two: Fix a issue in the setup / ask for help in the maillinglists / create a bug report with all the info you can gather.


Cheers,
Robert van Leeuwen

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20140331/4eb6abb4/attachment.html>