- Introduction
- Getting started
- Process modeling
- Process implementation
- Process operations
- Understanding Process operations
- Working with instance management
- Variables and element filtering
- Process monitoring
- Process optimization
- Reference information
Maestro user guide
Working with instance management
The process overview homepage provides an operational summary of all processes instances, and includes an active instances dashboard, a table of all incidents, and a table of all processes.
Active instances dashboard
The active instances dashboard displays the count of all instances, broken down by running vs. faulted instances.
Incidents table
The incidents table displays all unresolved incidents, with each incident corresponding to a faulted instance. The table is organized by error message per process and includes the time elapsed since a specific error message was first seen for that process.
Processes table
The processes table contains a list of all processes in the tenant that the user has access to. The table includes the counts for the number of running instances, faulted instances, and total instances, as well as the number of versions and the folder location in Orchestrator.
The processes in this view are based on the permissions you have as a user in Orchestrator, more specifically a folder permission for process view, or you can use the predefined Automation User folder role.
Process details
When you select a specific process from the Processes table, a table of all instances for that process appears. Additionally, you can view a diagram representing the process model by selecting a single version in the table.
All process instances dashboard
The all process instances dashboard offers a real-time and historical view of every process execution across your workspace. Use this screen to monitor automation activity, troubleshoot failures, and understand performance trends.
-
Top processes by instances
See which processes are executed most frequently. This view helps identify high-volume workflows that may need optimization or additional monitoring.
- Displays total execution count per process.
- Links let you navigate directly to that process’s detailed historyExample:
Loan.Origination.and.Reviewwith 151 instances is the most active process in this workspace.
-
Top processes by duration
Identify processes with the longest aggregate execution times. This can uncover bottlenecks or long-running flows needing review.
- Aggregates total runtime across all instances.
- Useful for performance monitoring and optimization.
The same
Loan.Origination.and.Reviewprocess has over 201 days of combined runtime. -
Process instance status (bar graph)
A visual chart showing how many instances completed, faulted, or were cancelled over time.
- Completed (green)
- Faulted (red)
- Cancelled (gray)Use this to spot trends across days or pinpoint anomalies. A spike in faults and completions on April 22 might indicate a recent deployment or an upstream system failure.
-
Process table (with filters)
This table provides a detailed breakdown by process name, including:
Column Description Running Currently active process instances Completed Total completed successfully Faulted Total that failed due to error Total Combined count of all executions Versions Number of published process versions Location Orchestrator Folder where the process is deployed Tip:You can search, filter by location, or refresh this table in real-time.
-
Drilling into instance details
To view detailed information about individual process runs:
- Select the process name link (e.g.,
Loan.Origination.and.Review) in the list or charts. This opens the Instance Details View. - Use filters to view instances by:
- Status: Running, Completed, Faulted, Cancelled
- Date/time range
- Version number
- Custom value filters
- Specific elementsThis detailed view helps you debug, audit, and optimize with full transparency.
- Select the process name link (e.g.,
Single instance view
When you select a single instance from the All instances view, a diagram showing the execution status of each step appears. This diagram is accompanied by the instance execution trail and pertinent variable and action history data. The single instance view allows you to see the entire execution trail, action history, step details, and global variables in one centralized location.
The instance execution trail clearly delineates each step in the instance process’s execution, including hierarchical steps such as agents and subprocesses. To further investigate, you can select a specific node on the diagram or a step in the Execution trail. The selected component will be highlighted in blue, revealing details such as step ID, label, input/output variables, execution duration, and links to additional tasks, such as opening a job or app task.
The Action history tab displays details for instance management actions that have been taken,including details about the action performed, the individual responsible, and any comments provided. This exhaustive information ensures complete transparency and traceability of each step in the instance execution.
In the Execution trail tab, if you click the first row, you will display the Global variables, but if you click any other row you will trigger the display of the Variables.
- Global variables are shared across instances of the same process, they are persistent and maintain their value beyond a singe instance.
- Variables are local to the process instance, are defined within a certain process instance, and they represent the current state or data context of that particular instance.
Roles and permissions
The processes in this view are based on the permissions you have as a user in Orchestrator, more specifically:
- a folder permission for jobs view, or you can use the predefined Automation User folder role.
- a tenant permission for packages view, or you can use the predefined Allow to be Automation User tenant role.
Managing an instance
You can also manage the instance, including pause, resume, retry, cancel, move, and migrate. When you perform one of these actions, there is a comment section where you can log additional details. These comments are then shown in the action history section.
Pause
Performing the pause action on an instance pauses the execution after all currently running steps have completed.
Resume
Performing resume on an instance will resume the execution.
Retry
Performing the retry action on an instance restarts all the current activities in the workflow from the last completed step. Typically this is done after editing a variable or updating something in an external system for a faulted instance.
Migrate
Performing the migrate action on an instance upgrades it to a different version of the process. This allows you to update the instance with the latest logic and configurations from the target version.
Cancel
Performing the cancel action on an instance will stop all current activities and terminate their progress in the agentic process. This action is irreversible.
If you cancel an instance that contains a human-in-the-loop task, Maestro automatically deletes that task, so you don't have to go to Action Center and remove it manually.
Move
Executing the Move action on an instance enables you to alter the sequence by relocating the execution to a different process element. This is typically applied to faulted or paused instances when a prior step requires re-execution or when you want to bypass the current steps and proceed to later ones.
After performing the move, you must select either the Retry or Migrate option to resume execution.
This operation might result in disruption to the ongoing execution path and potentially impact process consistency. Certain variable values needed at the new location may not generate automatically, necessitating potential manual updates for accuracy.
