6.2 Recover - Platform

When the CMS receives an exit code from the monitor script that indicates failure on the Platform, the following measures apply:

1. Stop the Platform by calling the offline script.

   Example

   $ ./offline <$JAVA_HOME> <$MZ_HOME> mzadmin platform 
    Shutting down platform...done.

2. The Platform should be down, but to make sure it is completely down, call the clean script.

   Example

   $ ./clean <$JAVA_HOME> <$MZ_HOME> mzadmin platform

3. Start the Platform in an alternative Platform Container. The database, including its listener, must be started before the Platform, since the Platform depends on it. The CMS must execute the database online script in an alternative container. 

   Example

   $ ./online <$JAVA_HOME> <$MZ_HOME> mzadmin platform
Starting platform...done.

   


   Note!

   The database and its corresponding monitor/online/offline-functionality is not part of the HA solution.

4. Due to the reconnection behavior of the EC/ECSAs, you do not need to restart these pico instances unless they are unavailable.  However, it is recommended that you check that the workflows are behaving as expected.
   
   

5. Services, running in an SC (e.g. Kafka), must be restarted when the Platform has recovered. You can update the online script to do so by uncommenting the following line:

   #cmd=$cmd;mzsh service restart --publish-only

   Next:

   6.3 Recover - EC, ECSA, and SC