Updating in a cluster group

If the updated FirstSpirit Server is a cluster master server in a cluster group, the participating cluster slave servers will also be updated and restarted automatically. This means that if the master server is shut down, all slave servers will also be shut down and restarted after a waiting time of 90 seconds (default) as soon as they can access the master server again. (The value of the waiting time can be configured via the cluster.slaveRestartDelay parameter in the fs-server.conf file (see FirstSpirit Server (fs-server.conf)), e.g., cluster.slaveRestartDelay=120).

Note: Updating the slave servers is only carried out for the default configuration if the cluster master server and cluster slave servers use a shared file system (see Clustering). In this case, the slave servers access the master server's fs-isolated-server.jar file and are thus also updated when the master server is updated. To prevent conflicts during the update process, the slave servers are started from the fs-update.jar for the waiting time stated above. This means that the shared fs-isolated-server.jar is not being used when the master server is restarted and can therefore be updated.

Mixed operation not possible in cluster environments: Mixed operation of isolated servers and legacy servers is not permitted in a cluster group. This means that an isolated master server cannot be operated in a group with a legacy slave server (or vice versa), for example.