Pre- and post-backup scripts for Veeam designed to find and protect only the active node in a vCenter High Availability configuration.
This is a set of pre- and post-backup Powershell scripts that get called on a job that backs up the vCenter Server Appliance 6.5 (vCSA) when it is in a vCenter High Availability (vCHA) configuration. VMware supports VADP-based backups of only the active node in a vCHA configuration. The script uses the vCenter API to detect which node has the active role, then excludes from the job the node with the passive role. At the conclusion of the job, the script removes this exclusion essentially "resetting" the job until its next run when it reruns the process. By using these scripts, one can ensure only the correct node is backed up each time, which has the added benefit of reducing the amount of backup storage data required to protect a vCSA and reducing backup times.
These scripts will require vCenter 6.5 and PowerCLI 6.5.1. PowerCLI 6.5.1 should be installed on the VBR server. They have been tested with vCenter 6.5d, PowerCLI 6.5.1, and Veeam Backup & Replication 9.5.
Create a Veeam backup job and add a vCenter container object where your vCHA nodes are listed. For example, if you have them all in a folder called "Management" in vCenter's inventory, add that folder to the job. Assuming the Management folder contains all three of the vCHA nodes (active, passive, and witness), manually exclude from your job the witness node. These scripts assume the witness has already been excluded. Download and store these scripts somewhere local to the VBR server, and edit the variables section at the top of each to customize appropriately to your environment. Edit your job and add these scripts to the appropriate pre- and post-job sections. When the job runs next, it should call out to vCenter to determine the state of the cluster, find the name of the VM that holds the passive role, and edit the job to exclude that passive vCHA node thereby only protecting the active node. When the job is complete, the post-job script runs and removes this passive node from the job configuration. Since nodes can fail over at any time and assume an active role, this ensures the next run of the job rediscovers the vCHA topology and only protects the VM that holds the active role at that time.