It’s been a little while since I’ve done any hands-on work with Rubrik, but I recently had to jump on a cluster and power it down so it could be relocated. The process is simple (particularly if you have the correct credentials), but I’m noting it here more for my own reference than anything else. It’s important to note that if you’re running a version of CDM pre-5.1 and have the cluster shutdown for longer than 24 hours, it will be sad when it comes back online and you’ll need support’s help to get it back online. Note also that 5.1 introduced a new command line structure (support site registration required), so the command is slightly different. This page also has a bunch of useful, publicly visible information.

If you’re not in the DC with the cluster, ssh to one of the nodes to run the commands. For pre-5.1 environments, run

poweroff_cluster

For 5.1 and newer environments, run

cluster poweroff_cluster

Type yes to continue and you should be good to go.

Here’s a picture of one I prepared earlier.

Exciting? Not really. But useful to know when people are threatening to power off equipment regardless of the state it’s in.