Hi Suresh, thanks for your answer!.

About the tag on the compute offering like you said, we use it to force a deploy in a specific storage.

Now, we are working in the new version with automatic deploy thorugh the API, like terraform, and we want to leave cloudstack select the storage with his algorithm, but we don't know how it really works.

We are inspecting the global setting "storage.pool.allocators.order" and it has the value "LocalStorage,ClusterScopeStoragePoolAllocator,ZoneWideStoragePoolAllocator"

We can't understand how it works, can you said if there is any documentation to know more about it or what values we can set to custom this selection?. For example, select the storage with more capacity, the storage with less load, or something like that.

Thanks!

Mauro

El 18/1/2022 a las 03:30, Suresh Anaparti escribió:
Hi Mauro,

Happy new year!

Cloudstack chooses the storage pool using the disk/storage specs (size, pool 
type: local/shared, tags, etc) mentioned in compute offering, when deploying a 
VM. For shared pool, it picks cluster-wide or zone-wide pool first based on the 
order mentioned in the global setting 'storage.pool.allocators.order'.

In order to select a particular storage pool while deploying a VM, you can add 
a tag to the storage pool and specify that storage tag in the compute offering, 
and then use that compute offering to pick the storage pool with the tag. 
Similarly, use storage tags in the disk offering for data disks.
btw, what is the cloudstack release used for deployment?


Regards,
Suresh

On 17/01/22, 9:49 PM, "Mauro Ferraro - G2K Hosting"<[email protected]>  
wrote:

     Hi guys, how are you?. Happy new year!

     We are deploying a new cloudstack integration with 2 primary storages
     and we are wondering how cloudstack choose the storage to deploy de VM.

     Somebody knows the criteria to select?, can we configure cloudstack to
     decide automatically with some criteria?.

     Thanks

     Mauro


Reply via email to