I had this question a couple of days back and searching around found this nice explanation:
http://blog.russellbryant.net/2013/05/21/availability-zones-and-host-aggregates-in-openstack-compute-nova/ So yes, if you create a new host aggregate, and pass an AZ name to it, users can use that to boot VMs with the --availability-zoe option. By default, default_schedule_zone points to nova in /etc/nova/nova.conf. So when no AZ is explicitly called out, the VM will boot in one of the default_schedule_zone. hth -balaji On 5 November 2015 at 10:51, Ken D'Ambrosio <k...@jots.org> wrote: > Trying to get a good idea of how they work. As I get it, host aggregates > are collections of (not necessarily mutually exclusive) hosts that share a > metadata value. And they can be part of availability zones which can be > (say) specific to networking hardware, or CPU, or something. What I'm > trying to accomplish is something where generic VMs being fired use the > default zone, "nova", but *only* ones where I specify a different zone -- > say, superfastcpu -- will make use of that zone. Do I have it right? Or > will VMs that I fire up where I don't specify a zone eventually make use of > the superfastcpu zone? > > Thanks! > > -Ken > > _______________________________________________ > Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack > Post to : openstack@lists.openstack.org > Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack -- http://balajin.net/blog http://flic.kr/balajijegan _______________________________________________ Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack Post to : openstack@lists.openstack.org Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
