Skip to main content
  • Place orders quickly and easily
  • View orders and track your shipping status
  • Enjoy members-only rewards and discounts
  • Create and access a list of your products

Dell PowerStore X model Software Upgrade Guide

Update the internal ESXi hosts on the cluster (PowerStoreOS 1.0.x to 2.0.x)

Use the vSphere Client to update the ESXi hosts on the nodes of a PowerStore X model cluster. For more information about updating ESXi hosts, see the VMware vSphere documentation.

Prerequisites

Ensure that the ESXi update package has been imported into vSphere Update Manager or vSphere Lifecycle Manager. For instructions, see Upgrading PowerStore X model clusters (PowerStoreOS 1.0.x to 2.0.x).
CAUTION: The internal nodes of a PowerStore X model cluster can only use ESXi updates that are qualified by Dell. Do not use ESXi updates that are obtained from VMware or any other source.

About this task

Perform the following steps to update the ESXi hosts on the PowerStore X model cluster:
NOTE: Some management operations that involve vVols may be unavailable during the upgrade.

Steps

  1. Identify the primary appliance and identify the IP addresses of the internal ESXi hosts on the primary and secondary nodes of the appliance. For instructions, see Minimizing internal ESXi host failover occurrences (PowerStoreOS 1.0.x to 2.0.x).
    NOTE: The terms peer node and secondary node are used interchangeably in PowerStore documentation.
  2. If there are multiple appliances in the cluster, perform the following steps to collect information about the other appliances:
    1. Select Settings > Network IPs in PowerStore Manager.
    2. Record the Host Node IP address, Appliance name, and Node letter of each node in the cluster.
    3. For the other appliances in the cluster, select the Hardware > Appliances > appliance > Components card in PowerStore Manager.
    4. Select the Internal View tab.
      The text (primary) is displayed next to the primary node of the selected appliance, and the secondary node is opposite of the primary node. For example, if the text (primary) is displayed for Node A, the secondary node is Node B.
    5. Use the collected information to identify the IP addresses of the internal ESXi hosts on the primary and secondary nodes of the other appliances in the cluster.
  3. In PowerStore Manager, select Compute > vCenter Server Connection and select Launch vSphere to start the vSphere Client.
  4. In the vSphere Client, log in to the vCenter Server and open vSphere Update Manager or vSphere Lifecycle Manager.
  5. Create a baseline for the update using the Patch content type.
    1. On the Select Patches Automatically page, clear the Automatically update this baseline with patches that match the following criteria checkbox to prevent unqualified VIBs from being installed on the ESXi hosts.
    2. On the Select Patches Manually page, select the filter icon in the ID column to display the filter field.
    3. In the filter field, type the ESXi update package name (without the .zip extension) listed in the PowerStore Release Notes, and select the ESXi package to add it to the baseline.
      NOTE: For vSphere 7.0 and later, type the package ID (8-digit build number) rather than the name.
  6. Stage the baseline to all the internal ESXi hosts in the PowerStore cluster.
  7. Before placing the ESXi host on one node of an appliance into maintenance mode, ensure that system resources are available on the other ESXI host to run the migrated user VMs. If the CPU and memory utilization on each ESXi host is greater than 75%, migrate the VMs to another ESXi host in vCenter Server using vMotion or power off the user VMs.
  8. Place the ESXi host on the secondary node of the primary appliance into maintenance mode.
    User VMs automatically migrate to the other node in the appliance. After all user VMs are migrated, the controller VM automatically shuts down and the ESXi host enters maintenance mode.
    NOTE: Some user VMs may not migrate due to local dependencies or lack of resources. User intervention may be required to manually power off these user VMs to allow the ESXi host to enter maintenance mode. See ESXi and VM considerations (PowerStoreOS 1.0.x to 2.0.x) for more information.
    CAUTION: Do not manually power off the controller VMs, which are named PSTX-<service_tag>-<node_letter>. The controller VMs are automatically powered off when an ESXi host is placed into maintenance mode.
  9. On the Updates tab, select the baseline that is attached to the ESXi host on the secondary node of the primary appliance and select Remediate to update the ESXi host.
    The ESXi update takes approximately 30 minutes to one hour to complete per node. The amount of time that an ESXi update takes depends on whether preupgrade and node firmware packages are applied when the ESXi host reboots.
  10. Ensure that the remediation process has completed for the ESXi host on the secondary node, and then exit maintenance mode.
    The controller VM automatically starts when the ESXi host exits maintenance mode.
  11. Ensure that the ESXi host and controller VM are operating normally before continuing.
    • Monitor the Protocol Endpoint (PE) paths in vCenter to identify when the paths to the ESXi host are Active before continuing.
    • Monitor the alerts in PowerStore Manager to identify when the node is up. Then, wait 15 minutes after the alert is cleared before continuing.
  12. Repeat steps 7 through 10 for the ESXi host on the primary node of the appliance.
    NOTE: PowerStore Manager temporally stops displaying performance metrics when the primary node of the primary appliance reboots during the ESXi update.
  13. Migrate VMs back to the preferred node using vMotion.
  14. Repeat steps 6 to 12 to perform ESXI updates on the nodes of the other appliances in the cluster.

Results

The ESXi hosts on the nodes of the cluster are now powered on and running the updated version of ESXi. The version of ESXi that the ESXi hosts are running is visible on the Upgrades page in PowerStore Manager.
  • If the ESXi versions on the nodes of an appliance do not match, an alert is generated.
  • If the ESXi versions on the appliances in the cluster do not match, an alert is generated.

Rate this content

Accurate
Useful
Easy to understand
Was this article helpful?
0/3000 characters
  Please provide ratings (1-5 stars).
  Please provide ratings (1-5 stars).
  Please provide ratings (1-5 stars).
  Please select whether the article was helpful or not.
  Comments cannot contain these special characters: <>()\