How to manage Qlik Replicate during a Windows Cluster Rolling OS upgrade

The document attached describes how to manage your Qlik Replicate servers in the scenario where the OS needs to be upgraded on each node in the cluster. This guide was created with a Windows Rolling Upgrade in mind. This assumes that one node will continue to run Qlik Replicate while the other node will be upgraded.

Option 1 - Qlik Replicate continues to run on the active node while the passive node is upgraded. Qlik Replicate is then moved over to the newly upgraded node and the tasks are resumed, while the passive node's OS is upgraded. This option offers the highest availability.

Option 2 - All Qlik Replicate tasks will be stopped, and the services will be disabled during the OS upgrade on both nodes. Qlik Replicate will only be started on each node to test that the upgrade was successful on a certain node, then it will be switched back off for the upgrade of the next node. I believe this is the safest option but has the greatest downtime.

I suggest Option 2. However, if relatively longer downtime is an issue then Option 1 should be chosen. Either way, both options will have some downtime. If you need assistance with this or further information, contact IBT Helpdesk.