Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

If i set default_availability_zone in cinder.conf on a controller node, will this mean that the volumes will be created only in that az or, if it is needed will create volumes on nodes from a different az?

I have a multi node openstack with havana on CentOS 6.5 . 2 blockstorage nodes. So i set each blockstorage node in different az. But it doesn't work if i don't set default_availability_zone in the controller nodes. You can set only one default_availability_zone per controller. This means, to my understanding, that the controller will create volumes only in the az set in default_availability_zone. But does this mean that the scheduler will not create any volumes if there isn't any space left in the az he is set as default? Will the scheduler go to the next az if it doesn't have enough dpace on the one set as default?

If i set default_availability_zone in cinder.conf on a controller node, will this mean that the How are volumes will be created only in that az or, if it is needed will create volumes on nodes from a within different az?availability zones?

I have a multi node openstack with havana on CentOS 6.5 . 2 blockstorage nodes. So i set each blockstorage node in different az. But it doesn't work if i don't set default_availability_zone in the controller nodes. You can set only one default_availability_zone per controller.

If i set default_availability_zone in cinder.conf on a controller node, will this mean that the volumes will be created only in that az or, if it is needed will create volumes on nodes from a different az?

This means, to my understanding, that the controller will create volumes only in the az set in default_availability_zone. But does this mean that the scheduler will not create any volumes if there isn't any space left in the az he is set as default? Will the scheduler go to the next az if it doesn't have enough dpace on the one set as default?