I'm myself troubleshooting the same issue on 4.11 (for a customer) - even
with "Random" set for the 'vm.allocation.algorithm' - the orred of pools
(when you check the logs) is always alphabeticall/ID-ascending

Which ACS version have you tested/are running?

Best,

On Fri, 11 Jun 2021 at 10:49, Daan Hoogland <[email protected]> wrote:

> Sean,
>
> look at `storage.pool.allocators.order` and `vm.allocation.algorithm`.
> together these should enable you to do what you want. If not, you have a
> feature request. One thing to note is that the allocators default and
> description don't mention all Allocater-classes available.
>
> GarbageCollectingStoragePoolAllocator and RandomStoragePoolAllocator
> are not mentioned.
>
> "LocalStorage,ClusterScopeStoragePoolAllocator,ZoneWideStoragePoolAllocator"
> is the default allocator order, but you can insert the other two
> where/if you want.
>
> regards,
>
> On Wed, Jun 9, 2021 at 10:06 PM Sean Lair <[email protected]> wrote:
>
> > We have two primary storage pools setup, both are NFS.  Both are
> > functioning, but CloudStack is provisioning all our VMs on one of the two
> > pools.  Is there a way to have CloudStack provision new VMs on the pool
> > with the most available storage?  Or how does CloudStack pick between the
> > two pools?
> >
> > Thanks
> > Sean
> >
>
>
> --
> Daan
>


-- 

Andrija Panić

Reply via email to