Download PDF
Download page Start or Stop the Controller.
Start or Stop the Controller
To start or stop the Controller:
- Linux users can use the Enterprise Console GUI or CLI.
- Microsoft Windows users can use the Enterprise Console GUI or following Windows services:
AppDynamics Database
AppDynamics Controller Application Server
AppDynamics Reporting Service
To check the Controller health, Linux and Windows users can use the Enterprise Console GUI or CLI.
Scripts Used to Start or Stop the Controller
The scripts to start and stop the Controller and other AppDynamics platform processes are located in the platform-admin/bin
directory for standalone or secondary HA Controllers. Using the platform-admin
script, you can start the platform processes individually or all at once.
To avoid the possibility of data corruption errors, ensure to stop the application server and database gracefully by using the stop scripts before shutting off or rebooting the machine on which the Controller is running.
You can start and stop the Controller and Controller services on the Controller page in the Enterprise Console GUI or from the Enterprise Console command line. When you start and stop the Controller, the services and processes related to the Controller also start and stop, including the Reporting Service. If you use the GUI to start and stop the Controller, specify that you want to stop MySQL if you want to also stop the Controller Database.
If your Enterprise Console manages multiple platforms, to distinguish the Controller platform, you must use the command line for standalone or secondary HA Controllers and specify the --platform-name <name_of_the_platform>
.
For example:
platform-admin.sh start-controller-db --platform-name <name_of_the_platform>
platform-admin.sh start-controller-appserver --platform-name <name_of_the_platform>
To see all options, run platform-admin.sh list-jobs --service controller --platform-name <platform-name>
from the command line. For more information, see Enterprise Console Command Line or Administer the Enterprise Console.
The Enterprise Console has a max wait time of 45 minutes when starting or stopping the Controller. You can set a timeout which exits the command and returns a failure by appending --args controllerProcessTimeoutInMin=<minutes>
to the end of your start or stop command.
How to Start or Stop a Standalone Controller
The following commands apply to standalone Controllers. See Starting or Stopping a Controller for information on how to start or stop Controllers in High Availability (HA) pairs.
Start and Stop the Controller App Server
When using the Enterprise Console, starting or stopping the Controller will also start or stop the Reporting Service.
Start the Controller App Server
The start-controller-appserver
command starts the database automatically.
platform-admin.sh start-controller-appserver
On Windows, the Controller processes are installed as a service, named the AppDynamics Controller Application Server service and AppDynamics Database service.
Start the Windows Service > AppDynamics Controller Application Server to start the Controller App server.
The database starts automatically when you start the Controller App server.
Stop the Controller App Server
The stop-controller-appserver
command does not stop the database.
platform-admin.sh stop-controller-appserver
Stop the app server:
Stop the Windows Service > AppDynamics Controller Application Server to stop the Controller app server.
The database does not stop when you stop the Controller app server.
Start and Stop the Controller Reporting Service
Start the Reporting Service
To start the Controller Reporting Service:
You must log in to the Controller host before running the scripts to start the controller.sh
processes.
bin/controller.sh start-reporting-service
Stop the Reporting Service
You must log in to the Controller host before running the scripts to start the controller.sh
processes.
bin/controller.sh stop-reporting-service
Start and Stop the Controller Database
Start the Controller Database
platform-admin.sh start-controller-db
Stop the Controller Database
platform-admin.sh stop-controller-db
How to Start or Stop a Controller in High Availability Pairs
Use the following commands to start or stop Controllers in HA pairs:
- You should disable auto-failover before restarting a primary Controller. Do not forget to reenable auto-failover afterward.
- If you enabled auto-failover in the Enterprise Console, and you stopped the app server, the Watchdog will trigger a failover if it takes longer than five minutes to update.
- If you are using a combination of the Enterprise Console with the High Availability Toolkit (HATK), then you can start or stop the Controller using services.
Start and Stop the Controller
Start the Controller
You must log in to the primary Controller host before running the scripts to start the controller.sh
processes.
bin/startController.sh
service appdcontroller start
Stop the Controller
bin/stopController.sh
service appdcontroller-db stop
Start and Stop the Controller App Server
You must log in to the primary Controller host before running the scripts to start the controller.sh
processes.
Start the App Server
bin/controller.sh start-appserver
service appdcontroller start
Stop the App Server
bin/controller.sh stop-appserver
service appdcontroller stop
Start and Stop the Controller Database
You can start and stop the Controller database either on the primary or the secondary host based on which database you require to restart.
Start the Database
To start the database, run this script on the server where you have hosted the database:
bin/controller.sh start-db
service appdcontroller-db start
Stop the Database
To stop the database, run this script on the server where you have hosted the database:
bin/controller.sh stop-db
service appdcontroller-db stop
How to Check Controller Health
To check the Controller health status from:
- Enterprise Console GUI, navigate to Platform > <name of the platform> > Controller. The right pane displays the health status of the Controller. The red color represents a critical status and the green color represents a normal status.
You can check the health status of each node from the Controller table. The Running State column displays the status for the corresponding Host. - Enterprise Console server run:
bin/platform-admin.sh check-controller-health
The following output shows the status of the Controller and its uptime:
Controller status : HEALTHY; Started 6 seconds ago.
To check on the health of the Controller, run:
bin\platform-admin.exe cli check-controller-health
The Powershell Execution Policy must be set correctly for the command to work. See https://docs.microsoft.com/en-us/powershell/module/microsoft.powershell.core/about/about_execution_policies?view=powershell-7.2.