You can't re-join them to the cluster if they're on different AsyncOS versions or builds numbers.
As a good practice, save the configuration file with the passwords unmasked before upgrading.
1. Do an administrative disconnect on both machines.
2. Suspend the listener on the first machine and then upgrade that one, while the second appliance is still processing mail.
3. Once the first appliance is complete in the upgrade, you can reboot it. Once it comes back up, resume the listener and verify mail flow
(tail mail_logs)
4. Then suspend the listener on the second appliance and repeat the upgrade and reboot process. Once the 2nd appliance is back up, resume the listener and verify mail flow
5. Once both machines are verified to be receiving mail, you can re-join both appliances to the cluster.
Iv patched them separately before. So Iv been through that process. But now that they are clustered, I am not sure how to proceed. I think i should do this:
1) Patch 1 unit. ( warning says all will be disconnected)
2) when it reboots, re-join it to the cluster.
3) patch 2nd unit (should already be un-clustered??)
4) re-join when the 2nd unit reboots.
am I even close?
