Page History

...

Upgrading XperienCentral in a clustered production environment is very similar to upgrading a standalone version of XperienCentral except that there are multiple instances of XperienCentral that must be upgraded (masterread/write node(s) and one or more slavesread-only nodes) and these must be upgraded in a specific order. To upgrade a clustered XperienCentral deployment, follow these steps:

1. Stop XperienCentral on all slave machinesread-only nodes.
2. Stop XperienCentral on the masterread/write node(s).
3. Upgrade XperienCentral on the masterread/write node(s) and all slaves read-only nodes by following the steps in the previous section .above.
4. See Upgrade Notes per Version Refer to the upgrade.txt file, located in the root directory of your XperienCentral installation, for further tasks (if any) that you need to perform for the newly installed version of XperienCentral.

...

1. Start XperienCentral on a master server read/write node with the setting
   
   -Dwebmanager.clustering.filestore=false.
   
   XperienCentral is completely started when the following messages appears in the Tomcat log:
   

   <date><time>.gx.webmanager.startup.impl.Startup start
INFO: XperienCentral started successfully in x ms



2. When you have determined that the upgrade was successful, stop the masterread/write node.

3. On the master read/write node that you started and confirmed that the upgrade was successful in the steps above, change the setting clustering.filestore to true
   (-Dwebmanager.clustering.filestore=true).

4. Start XperienCentral on that masterread/write node.
   
   After the master read/write node is started, the file store will be initialized by the File Distribution Service (files from the upload and upload_mm directories). Depending on the number of/size of files on the website, this process could take some time. During this process, the server node is available for use but could be slower due to the impact from the file synchronization process. Editors should be able to continue working. The following message appears in the log when the File Distribution Service has completed its task:

   INFO: Finished scanning the web roots for files for the File Distribution Service



5. Start XperienCentral on the second master read/write noder if you are running a dual master read/write node environment. In a dual master environment, wait Wait until the second master read/write node has started successfully before starting XperienCentral on all slavesread-only nodes. This could also take some time because the second master read/write node needs to synchronize files from the central file store to the local disk. Until the file synchronization process is complete, the second master read/write node will not be available for use.

6. In a single master read/write node environment, start XperienCentral on all slavesread-only nodes.

Environment that Does not use the File Distribution Service

1. Start XperienCentral on the master serverread/write node. XperienCentral is completely started when the following messages appears in the Tomcat log
   
   <date><time> nl.gx.webmanager.startup.impl.Startup start
INFO: XperienCentral started successfully in x ms
   
   
2. Confirm that the upgrade was successful.
3. Start XperienCentral on all slavesread-only nodes.

Back to top