Displays a list of all Runs for the task engine. The latest Runs are shown at the top of the list, but they can be sorted by clicking on the column headers. The Runs can be filtered by Source, Group, Name, Type (Tree, Local Routine, or Global Routine), Source Id, and Run Id. To filter the Runs, select the desired values in the filter form and press Apply. To remove all of the filters and show all runs press clear.
Clicking on the Run # will open a page displaying the details for that Run.
The page contains the following sections:
- Outputs (shown for routines only).
Timeline is shown by default and it contains a combination of the Triggers, Tasks, and Errors in one table sorted by the order of creation. The Inputs tab will show the Run Inputs (for routines) and source data (if relevant). The Triggers, Tasks, and Errors tabs show a table for each type of Run data. Finally, if the Run was a Routine, the Outputs tab contains the outputs returned from the Routine.
Clicking on each row of a Trigger, Task, or Error shows more detail depending on the type of record and node. To see details of an error, click the error row and then click the Error #.
From this screen, you can diagnose and fix the error (see below). Clicking the Builder button opens the tree builder for the Source Tree. Run Triggers represent actions that the task engine takes. For example, Triggers are created to start or continue Tasks. Triggers normally come from two places, outside sources that want to start or continue a Tree or create Trigger nodes on trees that are made to interact with other Trees.
Selecting Errors from the upper left of the console switches the view to the Errors list. The newest Errors are displayed first. All columns can be filtered. You can filter the list based on Source, Group, Tree Name, Type, Source Id, and Error Id. Click Apply to apply the filter and Clear to restore the default list.
Only active Errors are shown in this list, if an Error has been resolved it will not be displayed here. Also note that only Errors that occurred during a run of a tree or routine are shown here, system Errors are shown on the Admin tab.
Clicking on the Error # takes you to the Error Details page.
This console shows the details about a single Error. If the Error is active, it will include a Resolve Error dialog (show in blue below), otherwise it will display information about how the Error was resolved.
To resolve an Error, fill out the form and click Resolve Error. Some types Errors allow you to perform additional actions when resolving, like retrying the failed Task. The types of Errors and resolution actions are discussed below in the Additional Actions section.
The Run the Error occurred in is shown (clicking on the link will take you to the Run details for that particular Run), as well as the status and type of Error. Beneath those fields the Error details are shown. The details section includes the actual error raised from the code as well as any causes and stack traces.
Almost every type of Error has the Do Nothing action which marks the Error as resolved but does not do any additional processing. The Run will be left at whatever status it was in when the Error occurred.
Errors that occur when processing Nodes and Handlers have the following actions:
- The Retry Task action will tell the Task Engine to attempt to re-run that Node and Handler. This is helpful when the issue with the Node or Handler has been resolved and you want to continue the rest of the run by restarting at the failed spot.
- The Skip Task action will tell the Task Engine to continue the Run but it will not attempt to re-run the failed Node/Handler. This is helpful when the action the Handler was supposed to take was performed manually and you want to continue the rest of the process normally.
Errors that occur when processing Connectors have the following actions:
- The Continue Branch action will tell the Task Engine to restart processing of the Run as if the Connector had evaluated to true.
- The Cancel Branch action will tell the Task Engine to restart processing of the run as if the Connector had evaluated to false.
The Staged page shows work that is pending processing by the Task Engine. There are three different categories that Staged work is divided into: Backlog, Scheduled, and Paused. Each of these is a tab shown below.
Backlog displays work that is ready to be processed but the Task Engine is currently backlogged and will process once the Backlog has been cleared.
Scheduled shows work that is Scheduled to happen at a future time, which can be done with the Wait Handler.
Paused shows work that has started but is currently Paused because the Task Engine encountered a Paused Handler or Paused Tree.
Clicking on a row will reveal more details about that particular Run/Trigger (shown below).