Upgrade clusters

Do you require assistance?

Before starting an upgrade, best practice is determining whether you need help from Axway. Answer the following:

  • Did Axway professional services help in the initial deployment or a previous upgrade?
  • Are you upgrading to a version that is six or more versions later than your current version?
  • Have you implemented changes that make your environment atypical or unique?
  • Are you unsure whether in-house resources have the necessary technical expertise?

If you answer "yes" to one or more of these questions, or are unsure, contact your Axway account executive or technical support to discuss options.

 Important:  All nodes within an Axway Decision Insight (DI) cluster must use the same version of DI.

Upgrade steps

To limit downtime, the following steps are recommended.

1. Stop consumer nodes and upgrade them

  • Stop all replica nodes.
  • Stop HA backup node.
  • Upgrade each node like a standalone deployment: see Upgrade the deployment.

2. Stop producer node and upgrade it

  • Stop primary (for P/R cluster) or HA main node (for HA cluster).
  • Upgrade this node like a standalone deployment: see Upgrade the deployment.

    If you're upgrading a HA cluster, HA will not be usable while you're upgrading your DI nodes, since all nodes must run the same DI version. 

3. Start all nodes

Restart the producer node before the consumer nodes so that the database is migrated from producer node first.

Check the following page to get instructions on how to start Decision Insight:

If needed, the consumer nodes will download the last checkpoint migrated from the producer node.

If a node fails to start, check the <logging directory>/migration-error.log and the <logging directory>/node-error.log.



Related Links