KB-1340 How to automate the restart of stuck nodes within a process model

Purpose

DISCLAIMER: The application as well as the AppMarket plug-ins provided in this article are provided as-is with no express guarantee of functionality by Appian Technical Support.

Restarting process nodes when they fail is generally a manual process, with multiple process instances having failed process nodes that need to be restarted individually. There is no out-of-the-box way to restart multiple failed process nodes at once, but it can be achieved using a custom solution.

This article provides a high level example of how a custom solution could be developed using the following plug-ins available on the AppMarket:

• Process By Node Status
• Process Management Services

Instructions

1. Create a new process model.
2. Add the "Process By Node Status" smart service.
3. Open the smart service and enter the UUID of the process model that contains instances needing to be analyzed (this smart service can be looped over to support multiple process models).
4. Set the "status to search for" to 8 (Paused by Exception). This will search through the process model(s) from step 3) and return the process ids and node UUIDs for any nodes Paused by Exception.
5. Add the "Restart All Nodes" smart service
6. Pass in the node UUIDs and the process IDs obtained in step 4).

An application that demonstrates this is attached here: Restarting Failed Nodes Example.zip

Affected Versions

This article applies to all versions of Appian.

Last Reviewed: August 2020