NOTE: The information in this article is not release-specific. You can reference it for any release.
To make sure you complete all the tasks necessary for upgrading platform components after a release, we recommend following the upgrade checklist instructions below.
Before upgrading to a newer version of the platform, complete the tasks outlined in the Update Product Components documentation.
To ensure all data is transferred properly, perform your upgrades in the specific recommended order for the components you’re using:
Review the Upgrade the Enterprise Console documentation first.
To determine the requirements, review the Controller system requirements and increase your hardware resources if necessary.
The Enterprise Console supports the installation of the Events Service on a Windows environment for a single node install. If you have several remote nodes, you will need to do a manual upgrade of the Events Service cluster and set the keys manually. For instructions on connecting to the Events Service, see the End-User Monitoring Connection Settings documentation.
Note: Even though the Enterprise Console will set and sync the keys between the Controller and Events Service, we still recommend confirming that the keys were synced correctly:
appdynamics.on.premise.event.service.key == ad.accountmanager.key.controller
Note: If you're upgrading the Events Service using the Enterprise Console, you don't need to stop the Events Service.
Follow the Upgrade the Events Service documentation to complete the Events Service upgrade.
See Upgrade the EUM Server for instructions on upgrading an EUM Server.
Complete a full backup of the Controller. Reference Back up the Existing Controller for instructions.
Before completing the upgrade process, review the following and implement as appropriate:
If you are upgrading an HA pair, we recommend upgrading via CLI using the Phased Upgrade Method (new in v4.5.6). This allows you to retain the HA roles of the Controllers so that the primary and secondary roles do not switch after upgrade and you can check the status of your primary Controller before upgrading the secondary.
If you encounter any issues during or post-upgrade, you can roll back by promoting and failing over to the preserved, secondary Controller.
It is also possible to use GUI or the old CLI method to upgrade your HA pair, but we recommend the Phased Upgrade Method.
You can directly upgrade from small/demo deployments to large/extra large deployments. See the Tuning for Large Profile Controller documentation for instructions on preparing a large deployment.
Check the Controller&rs
database.log for any errors during this process. You can find the log at
<controller_home>/db/logs/database.log>. There should not be any
InnoDB: Error lines in the log from the last startup. If any errors are found, contact AppDynamics Support before attempting the upgrade. Upgrading the Controller with a corrupt database may put the Controller in a bad state with a high recovery time.
Once you’ve completed preceding steps and considerations, refer to the Upgrade the Controller using the Enterprise Console documentation to perform and complete the upgrade.
In the event of a failure, follow the troubleshooting steps in the Troubleshoot the Upgrade documentation.
In the event of a failure when upgrading a platform component, please contact AppDynamics Support.