I would also look at seeing how its doing it. IN the past what it did was drop packets over a specific threshold which is really really really terrible. We do some traffic policing on some of our vm's – but we do it outside of openstack via a qemu hook and setting up our own qdisc and ifb device for each tap device that we want to police.
https://github.com/godaddy/openstack-traffic-shaping ___________________________________________________________________ Kris Lindgren Senior Linux Systems Engineer GoDaddy From: Joseph Bajin <[email protected]<mailto:[email protected]>> Date: Wednesday, June 29, 2016 at 10:43 AM To: Daniel Levy <[email protected]<mailto:[email protected]>> Cc: OpenStack Operators <[email protected]<mailto:[email protected]>> Subject: Re: [Openstack-operators] Bandwidth limitations Hi there, It looks like QOS is already available within the Mitaka release. Maybe it doesn't have all the features you need, but looks to be a good start. http://docs.openstack.org/mitaka/networking-guide/adv-config-qos.html I haven't used it yet, but maybe someone else will pipe up with some expierence. --Joe On Wed, Jun 29, 2016 at 12:36 PM, Daniel Levy <[email protected]<mailto:[email protected]>> wrote: Hi all, I'd like to learn about potential solutions anyone out there is using for bandwidth limitations on VMs. Potentially applying QOS (quality of service) rules on the VM ports in an automated fashion. If there are no current solutions, I might submit a blue print to tackle this issue _______________________________________________ OpenStack-operators mailing list [email protected]<mailto:[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
