Dell PowerStore 1200T EMC PowerStore Release Notes for PowerStore OS Version 3 - Page 15

Issue ID MDT-109970 MDT-111956 MDT-156507 MDT-146763 MDT-151804 MDT-152993 Functional Area Virtualization Virtualization Virtualization Virtualization Virtualization Virtualization Description Workaround/Resolution Unable to create a vVol storage container in PowerStore, after having just deleted a vVol datastore with the same name from vSphere. In rare case, the automated mounting of a user-created PowerStore vVol storage container on an ESXi host could fail with the following error message displayed in vCenter "Timedout waiting to get datastore information from host". In high scale VMware vSphere environments, where the ESXi hosts and vCenter server are highly loaded (for example, simultaneous powering on a large number of VMs), heartbeat messages sent by the ESXi hosts to the vCenter server may not be delivered in time. When the heartbeat is not received in the 60-second timeout interval, the host is considered as not responding, and the host is marked as disconnected. This may result in the vSphere High Availability (HA) to trigger VM migrations to other hosts in the cluster. During a vMotion process, some vVols are deleted. If the VASA provider becomes unavailable during a storage vMotion operation, the deletion of the vVols fails. These vVols are left orphaned and may appear as duplicates. In a PowerStore cluster with multiple VMFS datastores that are mapped to multiple hosts, a performance impact might occur if several VMs are powered off or powered on simultaneously. Either create the vVol storage container in PowerStore using a different name or allow some more time before creating the new vVol storage container in PowerStore with the same name that was used in vSphere. Manually mount the vVol storage container from the VMware vSphere GUI or CLI. In vCenter server, increase the heartbeat timeout interval to 120 seconds. See VMware Knowledge Base article 1005757: "ESXi host disconnects intermittently from vCenter Server." In case some VMs appear as invalid or orphaned as a result of the ESXi host disconnect, see the VMware Knowledge Base article 1003742: "Virtual machines appear as invalid or orphaned in vCenter Server." for instructions on recovering the VMs. Delete the orphaned vVols using PowerStore CLI or REST API. Refer to the Dell knowledge base article 000320477: Best practices for VMFS datastores - for Bootstorm or Failover with VMware SRM. A periodic, automatic cleanup process on vSphere that is performed randomly (every few days) deletes empty config vVols (directories). However, this process also deletes the ISOs folder that is pre-created by PowerStore in case you have not placed any ISO images into the folder. Recreate the ISOs folder using the following PowerStore CLI command: pstcli -service storage_container -id create_directory -size 100G -name ISOs By default PowerStore creates a 500 GB folder, but you can specify another size if necessary. Known Issues 15