mike-tutkowski commented on a change in pull request #2253: [Managed Storage]
When starting a VM, make sure it has the correct volume access group
URL: https://github.com/apache/cloudstack/pull/2253#discussion_r136167911
##########
File path:
engine/orchestration/src/org/apache/cloudstack/engine/orchestration/VolumeOrchestrator.java
##########
@@ -1369,6 +1369,13 @@ public void prepare(VirtualMachineProfile vm,
DeployDestination dest) throws Sto
if (task.type == VolumeTaskType.NOP) {
pool =
(StoragePool)dataStoreMgr.getDataStore(task.pool.getId(),
DataStoreRole.Primary);
vol = task.volume;
+ if (pool.isManaged()) {
Review comment:
@syed I am wondering if we should instead call revokeAccess on the volume
for the old cluster and grantAccess on the volume for the new cluster. That way
the storage plug-in doesn't need to have any new, special logic around
automatically removing the volume from an existing VAG when adding it to a new
VAG. It also provides a level of symmetry in CloudStack's core code around
grant/revokeAccess (the core code never gets into a grantAccess + grantAccess
scenario - it's always grantAccess + revokeAccess + grantAccess + revokeAccess,
etc.).
What do you think?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
