[CLOUDSTACK-5140] A stopped vm cant start after disable threshold has been reached on the storage pool

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 4.2.0, 4.3.0
• Fix Version/s: 4.2.1, 4.3.0
• Component/s: Management Server
• Security Level: Public (Anyone can view this level - this is the default.)
• Assignee: Prachi Damle
• Reporter: Prachi Damle
• Created: 12/Nov/13 02:08
• Updated: 06/Dec/13 21:21
• Resolved: 06/Dec/13 21:21

Description

Incorrect primary storage allocation behaviour:

If a datastore reaches ‘disable threshold’ a stopped VM that was already on that datastore will refuse to start? Pre 4.2 thresholds were only applicable at the allocation time.

I still see the code is present in 4.2 and master as well that can cause this issue.
Following highlighted code is what causes it:
DeploymentPlanningMgrImpl :: findPotentialDeploymentResources () {
….

if (hostCanAccessSPool(potentialHost, potentialSPool))

{ hostCanAccessPool = true; if (multipleVolume) { List<Volume> requestVolumes = null; if (volumeAllocationMap.containsKey(potentialSPool)) requestVolumes = volumeAllocationMap.get(potentialSPool); else requestVolumes = new ArrayList<Volume>(); requestVolumes.add(vol); if (!_storageMgr.storagePoolHasEnoughSpace(requestVolumes, potentialSPool)) continue; volumeAllocationMap.put(potentialSPool, requestVolumes); }

storage.put(vol, potentialSPool);
haveEnoughSpace = true;
break;
}
…
}

Issue is that this check should NOT be done for a Volume that is already created and exists on the pool.

Happens with a VM having root and data volumes.