1 of 1 people found this helpful
(1) This is correct, for High Availability (HA) you basically have redundant services running on both workers. That way if the services on one worker go down for some reason, you still have services running on the other worker. To complete the HA picture you also want to be able to recover if something happens to the Primary. So if for some reason your Primary goes down (eg hard drive crash) you can run the command to switch over to your Backup Primary. If you don't have a Backup Primary and your Primary goes down and you're unable to bring it up, the workers nodes will eventually fail their internal licensing check (since they can no longer reach the Primary) and they will become unlicensed.
(2) The Backup Primary is mostly in regards to having another node that can take over licensing as well as coordinate the workers. When you setup the Backup Primary you're basically telling the workers to accept commands from either the Primary or the Backup Primary. But you wont' be running the Primary and Backup Primary at the same time, only one should be active. If your Primary fails you'll need to run the tabadmin failoverprimary command to switchover to your Backup Primary.
(3) The Primary would not be hosting the repository in a HA environment. More than likely your team has already been reviewing the links in Help but for the benefit of folks who might be viewing this thread including this link below:
You can see in the diagram that the repository is not hosted on the Primary. If the main repository fails one one worker it should automatically cutover to using the backup repository which is kept in sync. The Backup Primary only comes into play if your Primary goes down and cannot be brought up for whatever reason.
Thanks for confirming and answering the questions Mark. Appreciated much for your time and help.
Thanks & Regards