On 13/05/15 18:06 +0000, Fox, Kevin M wrote:
Sahara also has the same problem, but worse, since they currently only support ssh with their agent, so its either assign floating ip's to all nodes, or your sahara controller must be on your one and only network node so it can tunnel. :/Should we have a chat with them too?
We've scheduled a discussion with Trove's team on Thursday at 5pm. It'd be great to have this discussion once and together to know what the common issues are and what things need to be done. I'll ping folks from both teams to invite them to this session. If they can't make it, I'm happy to use another working session slot. Cheers, Flavio http://libertydesignsummit.sched.org/event/59dc6ec910a732cdbf5970b6792e1cef#.VVRL0PYU9hE
Thanks, Kevin ________________________________________ From: Zane Bitter [[email protected]] Sent: Wednesday, May 13, 2015 10:26 AM To: [email protected] Subject: Re: [openstack-dev] [trove][zaqar] Trove and Zaqar integration. Summit working session On 11/05/15 05:49, Flavio Percoco wrote:On 08/05/15 00:45 -0700, Nikhil Manchanda wrote:3) The trove-guest-agent is in vm. it is connected by taskmanager by rabbitmq. We designed it. But is there some prectise to do this? how to make the vm be connected in vm-network and management network?Most deployments of Trove that I am familiar with set up a separate RabbitMQ server in cloud that is used by Trove. It is not recommended to use the same infrastructure RabbitMQ server for Trove for security reasons. Also most deployments of Trove set up a private (neutron) network that the RabbitMQ server and guests are connected to, and all RPC messages are sent over this network.We've discussed trove+zaqar in the past and I believe some folks from the Trove team have been in contact with Fei Long lately about this. Since one of the projects goal's for this cycle is to provide support to other projects and contribute to the adoption, I'm wondering if any of the members of the trove team would be willing to participate in a Zaqar working session completely dedicated to this integration?+1 I learned from a concurrent thread ("[Murano] [Mistral] SSH workflow action") that Murano are doing exactly the same thing with a separate RabbitMQ server to communicate with guest agents. It's a real waste of energy when multiple OpenStack projects all have to solve the same problem from scratch, so a single answer to this would be great. In that thread I suggested (and Murano developers agreed with) making the transport pluggable so that operators could choose Zaqar instead. I would strongly support doing the same here.
+1 :) Flavio
cheers, Zane.It'd be a great opportunity to figure out what's really needed, edge cases and get some work done on this specific case. Thanks, Flavio __________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: [email protected]?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev__________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: [email protected]?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev __________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: [email protected]?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
-- @flaper87 Flavio Percoco
pgpjR8SbiJejp.pgp
Description: PGP signature
__________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: [email protected]?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
