Dell PowerStore 9000X EMC PowerStore Release Notes for PowerStore OS Version 1 - Page 12

VMware Knowledge Base article 1005757. In, SLN320477: Best practices for VMFS

Page 12 highlights

Issue ID MDT-156507 MDT-146763 MDT-151804 MDT-152993 Functional Area Virtualization Virtualization Virtualization Virtualization Description Workaround/Resolution 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. In vCenter server, increase the heartbeat timeout interval to 120 seconds. See VMware Knowledge Base article 1005757. In case some VMs appear as invalid or orphaned as a result of the ESXi host disconnect, see the VMware Knowledge Base article 1003742 for instructions on recovering the VMs. 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. Delete the orphaned vVols using PowerStore CLI or REST API. 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. Refer to the Dell knowledge base article SLN320477: "Best practices for VMFS datastores - for Bootstorm or Failover with VMware SRM" for additional details. 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 precreated by Trident 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. 12 Known Issues

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16

12
Known Issues
Issue ID
Functional
Area
Description
Workaround/Resolution
MDT-156507
Virtualization
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.
In vCenter server, increase the heartbeat
timeout interval to 120 seconds. See
VMware Knowledge Base article 1005757. In
case some VMs appear as invalid or
orphaned as a result of the ESXi host
disconnect, see the VMware Knowledge
Base article 1003742 for instructions on
recovering the VMs.
MDT-146763
Virtualization
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.
Delete the orphaned vVols using PowerStore
CLI or REST API.
MDT-151804
Virtualization
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.
Refer to the Dell knowledge base article
SLN320477: “Best practices for VMFS
datastores – for Bootstorm or Failover with
VMware SRM” for additional details.
MDT-152993
Virtualization
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 Trident 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 <id
where to create the folder> create_directory
-size 100G -name ISOs
By default PowerStore creates a 500 GB
folder, but you can specify another size if
necessary.