The stop script shuts down the Appian engines and associated other services, like Service Manager and the Internal Messaging Service.
|Short Name||Long Name||Required||Meaning|
|-h||--help||No||Show usage information|
|-p||--password||Yes||Password for Admin REST API|
|-s||--services||No||A comma-separated list of services to stop. See below for values. Excluding this flag will result in stopping the cluster.|
|-nk||--no-kafka||No||Skip stopping Internal Messaging Service|
|-par||--parallel||No||Stop all engines at once (deprecated; this will have no effect)|
|-q||--quick||No||Skip checkpointing when stopping an engine or engines|
Available values for the
If you are shutting down a single server you should use
-s all. Otherwise, you do not need to use the
-s flag and by default the cluster will be shut down.
When using the default shutdown behavior, you must run the stop script on all servers that are configured to run engines at the same time as the script will pause and wait for other servers to indicate that they have shut down their services before exiting.
By default, and unless you use the
--quick option, the stop script will checkpoint the engines as they shut down. This results in .kdb files that contain up-to-date data (i.e. no data is persisted solely in the Internal Messaging Service - Kafka) which are suitable for performing an upgrade. Using the
--quick option will result in .kdb files that do not contain all of the engine's data and the remaining transactions will need to be replayed as part of the subsequent startup. Kdb files like this are not suitable for performing an upgrade.
This will stop all services (including the Internal Messaging Service)
1 ./stop.sh -p <password> -s all
This script can be used for a single server environment or against one server for a High Availability or distributed deployment. Only the services on the machine where the script is being run will affected.
This must be run on all nodes.
1 ./stop.sh -p <password>
To be the least disruptive for end users when you need to restart nodes in a Highly Available environment for maintenance. This is not recommended for distributed environments.
1 ./checkpoint.sh -p <password> -s all
1 ./stop.sh -p <password> -s all -q