Note: this article is intended to be used in the event that all VMs and hosts at a site must be shut down. If only specific hosts with VRAs need to be shut down, see How to Prepare to Power Off a Site with Zerto Virtual Replication Installed.
In order to prepare a site with Zerto Virtual Replication installed to be powered off, follow these steps:
- Perform a Guest OS Shut Down of the ZVM at the site that will be powered off.
- Once the ZVM is powered off, select each VRA in the vSphere Client, and perform a Guest OS Shutdown.
- Note: a proprietary third party version of VMware tools is installed on the VRAs and it is highly recommended to use the Guest OS Shutdown option.
- Following the above steps, alerts are registered at peer sites regarding disconnections to this site. These are expected as the resources are now unavailable.
- If VPGs are configured with the affected site as a Target Site, they will enter "Bitmap Sync" status until this site is once again available. See the article Explanation of VPG Statuses for further details.
- If VPGs are configured with the affected site as a Source Site, the VPG will be in a status of "Recovery Possible" at the target site. See the article Explanation of VPG Statuses for further details.
- Confirm that the vCenter VM is online and available.
- Confirm that all hosts are online and connected to vCenter.
- Confirm that all datastores are online and available to the hosts.
- Confirm that the network connection between sites is available, healthy, and that any relevant port rules required by Zerto Virtual Replication are preserved.
- Power on the ZVM VM.
- Power on the VRA VMs.
- A few minutes after ZVM VM has been powered on, verify that the the Zerto Virtual Replication GUI is available via the vSphere Client.
- Confirm that the connection between Zerto Virtual Replication site components is reestablished - this will be noted in the "Alerts" report area of the Zerto Virtual Replication GUI.
Note: be advised that it may take up to 15 minutes for all components to successfully reestablish their connections and for all VPGs to proceed with their appropriate syncs.