Since the Clustered Active Directory is supported for Jira, the Incremental Synchronisation often failed when the DC change to a different LDAP node.

The following stack trace appear in Jira application logs:

2021-06-24 15:54:02,472+0300 Caesium-1-2 INFO ServiceRunner [c.a.c.d.synchronisation.cache.UsnChangedCacheRefresher] Last incremental synchronization took place for AD instance with invocation id 'xxxxx-xxxxx-xxxxx-xxxx1', current instance has invocation id of 'xxxxx-xxxxx-xxxxx-xxxx0', falling back to full

When this happen, a Jira instance that contain large user base will experience performance issue due to high load from the full sync.

Therefore, it will be very helpful to have an option to allow incremental synchronisation when the Clustered Active Directory change to a different node.