HyperFlex Offline Upgrade Workflow

Offline Upgrade Guidelines


Restriction

Offline Upgrade to HyperFlex Release 6.0(1a) is not supported. To upgrade to HXDP 6.0(1a) use the upgrade process found in Cisco HXDP Upgrade using HX Installer.


Important

  • Offline upgrade can be performed from HX Connect UI for either combined or split upgrade. Before you proceed, consider following guidelines:

  • Cisco recommends performing online upgrade from HX Connect UI for hitless upgrade experience without any impact operations.

  • Offline upgrade requires the cluster to be shutdown.

  • Offline cluster upgrades require the node(s) to reboot. Applications running on the cluster during an offline upgrade are impacted because they do not have access to the HyperFlex storage during the cluster reboot.

  • Nodes are upgraded with the new version of the Cisco HX Data Platform software and rebooted one at a time.

  • Remote-to-Remote Upgrades in HXDP Release 6.0(x)

    • Protected VMs: Once Site A is upgraded to HXDP Release 6.0(x), there should not be any protected VMs under Other DRO. VMs under the Other DRO category on Site A, are moved to the Standalone tab in Site A.

    • Datastores: Once Site A is upgraded to HXDP Release 6.0(x), there should not be any datastores listed for pairing under Other DRO. Datastores listed for pairing under the Other DRO category on Site A, are relocated to the Native Protection tab. At Site B, this reassignment only occurs during the upgrade process to HXDP Release 6.0(x).

  • Unmapping DRO datastore pairs simultaneously with any other Native datastore mapping or unmapping actions is not supported after upgrading the Site A and Site B clusters to HXDP Release 6.0(x). Thus, unmapping of DRO datastores must be carried out individually.

  • Offline cluster upgrades with nested vCenter is not supported.

Offline Upgrade Process Workflow

Step

Description

Reference

1.

If UCSM (A-bundle) or UCS Server Firmware (C-bundle) upgrade is required, download Cisco UCS Infrastructure A, blade bundle B, and rack bundle C.

Software Download

2.

Upgrade Cisco UCS Infrastructure bundle as required.

Upgrade UCS Infrastructure Firmware

3.

Launch the vSphere Web Client and power down all user VMs (HyperFlex Controller VMs should stay powered on) residing on HX servers and all user VMs running on HX datastores. This includes VMs running on Compute-only nodes. After the VMs are shut down, verify the health state of the cluster and perform a graceful shutdown.

Important

 

HyperFlex controller VMs (stCtlVMs) must remain powered on.

For more details, see Shut Down and Power Off the Cisco HX Storage Cluster.

4.

(Optional) SSH to cluster management IP as admin user and disable the snapshot schedule.

Run the command stcli snapshot-schedule --disable.

5.

Log into HX Connect as admin user and perform either a combined upgrade or an individual component upgrade.

For a combined upgrade see Upgrading HyperFlex Data Platform Software, VMware ESXi, and Cisco UCS Server Firmware using HX Connect.

For an individual component upgrade see Upgrade UCS Firmware, HX Data Platform and VMware vSphere - Individual Component Upgrade.

6.

Confirm the upgrade is complete and perform the post upgrade tasks .

Post Upgrade Tasks

7.

Start the cluster.

Preparing for HX Storage Cluster Maintenance

8.

(Optional) Enable snapshot schedule if it was disabled previously in step 4.

Run the command stcli snapshot-schedule --enable.