Install the Administrator Daemon on the nodes where you have installed the Informatica domain.
The following situations describe how the
restarts or fails over to a standby machine:
- Informatica domain becomes unavailable
- If the Informatica domain becomes unavailable on the master gateway node, it fails over to a backup node. The Administrator Daemon failover depends on the domain failover. When the domain fails over to the backup node, the
Administrator Daemon that is running on the same node also fails over to the backup node.
- Administrator Daemon process becomes unavailable
If the Administrator Daemon process becomes unavailable on the master gateway node, the process can restart.
On Linux, the
Administrator Daemon process script
admind.sh runs the
processmonitor-admind.sh script which detects if the
Administrator Daemon, LBMRD, and Apache ZooKeeper are running. If one of the processes stop unexpectedly, the
processmonitor-admind.sh script restarts only the process that has stopped.
On Windows, the
Administrator Daemon, LBMRD, and Apache ZooKeeper run as services. The recovery action for the services is set as
Restart the Service. If one of the services stop unexpectedly, the service that has stopped is restarted.
The LBMRD runs on the same machine as the
Administrator Daemon. When the Informatica domain fails over to a backup node, the
Administrator Daemon detects that the backup node is the new master gateway node and uses the LBMRD that is running on the master gateway node for topic resolution.