Incidents
An incident is an error that stops a process instance from continuing, for example when a job or an external task has run out of retries. The mechanism is described in the Operaton documentation: Incidents.
The Incidents view shows open incidents of the connected engine across all processes. To open it, select Incidents in the application menu. The incidents of a specific process instance are also shown on the Incidents tab of the detailed instance view, and such instances are marked with an icon in the instance list.
Incident list
For each incident the list shows:
-
Id: the incident identifier, which opens the incident details;
-
Type: the incident type, for example a failed job or a failed external task;
-
Message: the error text;
-
Process and Process instance;
-
Activity: the process element where the incident occurred;
-
Creation date.
You can filter the list:
-
by process and process version;
-
by incident type, including a custom type (the Custom value);
-
by message text;
-
by activity;
-
by process instance identifier;
-
by creation date.
Retrying failed steps
If an incident was caused by a job or an external task that ran out of retries, you can run the step again: Control increases the number of remaining retries, and the engine tries to execute the step once more.
To retry several incidents:
-
Select the incidents in the list.
-
Click the retry button above the list.
-
In the Retry failed activities dialog, enter the new number of retries in the Retries field and confirm.
The number of retries is increased asynchronously. Refresh the list to see the result.
For a single incident, retrying is available in the incident details: the Retry failed job or Retry external task dialog opens, depending on the incident type.
| If the cause of the error has not been fixed, the incident appears again once the new retries are used up. Check the error message and the stack trace first. |
Incident details
To open an incident, click its identifier, or select it and click View. The window has three groups:
-
General: the identifier, type, message, creation date, and payload (for example, the identifier of the job or external task);
-
Process & activity: the process, the process instance, the activity, and the failed activity;
-
Related incidents: the cause incident and the root incident, if this incident was caused by another one.
From this window you can:
-
open the stack trace of the error;
-
go to the process, the process instance, or the related job or external task;
-
open the cause incident or the root incident;
-
copy the incident identifier and the payload;