Hi Bryan,
Your nodes memory is independent of each other so you can bring one node down, increase the heap and then bring that node back up until all your nodes have the memory increase that you need. As an admin, you definitely want to ensure that JIRA can keep up if a node fails and all of the load of the failed node is dumped on the remaining live nodes.
For more information on how memory is used in a cluster please see Node Sizing in a Clustered JIRA Environment.
Cheers,
Branden
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.