Ask Your Question
0

2 Availability Zones on one compute node

asked 2016-07-25 08:41:43 -0500

bertly gravatar image

updated 2016-07-25 08:58:53 -0500

Hi,

is it possible to have two AZ on one compute node? The problem is that I want to have a local storage and an own nova availability zone.

That means that I will have on compute-1 node ceph (globally available on all compute nodes) and lvm-1 (only on compute-1 available), but I want to have two AZ for that. One called ceph and another one that is called lvm-1 which can only start on compute-1 (to reduce network). I tried host aggregations, but i could not make sure that the virtual machine was started on the node where the local storage is. Any idea?

I am running mitaka, but should be the same as liberty.

Many Thanks

bertly

UPDATE: from what i see that means that nova-compute on compute-1 is available in two AZs. Is that even possible?

edit retag flag offensive close merge delete
add a comment

1 answer

Sort by ยป oldest newest most voted
0

answered 2016-07-29 08:11:08 -0500

Frank_SU_HH gravatar image

Hi, you can actually add 1 compute node to different aggregates. Your limitations are: 1 aggregate can only be in 1 availability zone. 1 availability zone can host multiple aggregations. From my perspective availability zones are morely a kind of meta tag you can use for seperation.

So the question is, does it make any sense to do what you asked? The aggregations are most helpful for scheduling tasks together with the filters nova-scheduler provides. By adding extra_specs to either images or flavors, or/and adding the Multitenancy Filter, you can fine grain the usge of your compute(s).

So, what are you up for?

cheers

Frank

edit flag offensive delete link more

Comments

Hi Frank, you are right. I think I dont need multiple AZs in this use case. What I want to try to do is to have an local storage (additional to the ceph) and make sure that machines that use the local storage are started on that machine. I did open the following question: http://bit.ly/2atp6Bu

bertly gravatar imagebertly ( 2016-08-01 06:15:59 -0500 )edit
add a comment

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

subscribe to rss feed

Stats

Asked: 2016-07-25 08:41:43 -0500

Seen: 562 times

Last updated: Jul 29 '16

Related questions

clarifications about storage and backends [closed]

Cinder Volume driver DellStorageCenterFCDriver not initialized

What happens if a host with LVM thin-provisioning runs out of space?

Issues creating volumes with storage node [closed]

ObjectNotFound: [errno 2] error connecting to the cluster (Cinder+Ceph))

Why Glance image upload is really slow using ceph storage integration?

issue with cinder with ceph kolla-ansible pike

custom roles for swift and cinder

Cinder volume usage by for each tenant

Swift and Cinder high availability