A dry-run instance is an instance that the scheduling system marks as successfully run but never actually executes. No data is processed, no operational logs are generated, and no execution duration is recorded.
Node vs. instance: A node is a template that defines what runs. An instance is the runtime record created each time the scheduler triggers that node. Dry-run is a property of instances, not nodes — though a node's configuration directly determines whether its instances become dry-run.
Why dry-run instances exist
Five scheduling conditions produce dry-run instances. Use the table below to identify your situation before reading the scenario details.
| Trigger condition | Instance status shown | Expected or needs action |
|---|---|---|
| Node is scheduled weekly or monthly; today is not the scheduled day | (normal scheduling status) | Expected — by design |
| Node deployed with Start Instantiation = Immediately After Deployment; instance scheduled within 10 minutes of deployment | Deprecated real time generated task | Expected — by design |
| A failed instance was manually set to successful | Instance Set Successfully | Expected — intentional action |
| Node's Recurrence is set to Dry Run in the schedule configuration | (normal scheduling status) | Needs action — check your configuration |
| Node is excluded from a backfill workflow but lies in the dependency chain | Unselected instance in temporary workflow | Expected — by design |
Scenario 1: Weekly or monthly node — today is not the scheduled day
For nodes scheduled to run on a specific day each week or month, the scheduling system runs the node only on that day. On all other days, it generates dry-run instances instead of actually running the node.
To confirm the node's schedule, go to the node's Properties tab > Schedule section and check the configured recurrence day.
To avoid dry-run instances when backfilling data for a weekly or monthly node, use a data backfill node and set the data timestamp to the day immediately before the node's scheduled run day:
-
Monthly node (runs on the 1st of each month): set the data timestamp to the last day of each month
-
Weekly node (runs on Monday): set the data timestamp to Sunday of each week
For an auto triggered instance to actually run, its scheduled time must be more than 10 minutes after the node is deployed. Alternatively, set the data timestamp to yesterday and use a data backfill node — this lets the auto triggered instance run on schedule today.
View the scheduled time and data timestamp of the auto triggered instance on the current day. 
Scenario 2: Instance deprecated immediately after deployment
When Start Instantiation is set to Immediately After Deployment in the Schedule section on the Properties tab, the scheduling system runs only instances whose scheduled time is more than 10 minutes after deployment. Instances scheduled within 10 minutes of deployment are skipped — the system generates dry-run instances with the status Deprecated real time generated task instead.
For details on configuring this setting, see Configure time properties for a node to immediately generate an instance.
Scenario 3: Instance manually set to successful
After you set a failed instance's status to successful, the scheduling system does not re-run that instance. Instead, it skips the instance and continues running the downstream (descendant) node instances. The instance status shows as Instance Set Successfully (also displayed as Succeeded).
Scenario 4: Node's recurrence is set to dry run
If a node consistently generates dry-run instances, check its schedule configuration.
In DataStudio, go to the node's Properties tab > Schedule section and verify whether Recurrence is set to Dry Run. If it is, the scheduling system always generates dry-run instances for that node.
Scenario 5: Node excluded from a backfill workflow
When running a backfill on a subset of nodes in a dependency chain, any intermediate node that is not selected gets a dry-run instance with the status Unselected instance in temporary workflow (triggered when users manually select a subset of nodes in a temporary backfill workflow).
For example, given the dependency chain Node C → Node B → Node A (C depends on B, B depends on A): if you backfill data for Node A and Node C but not Node B, the scheduling system generates a dry-run instance for Node B.
Troubleshoot dry runs for daily nodes
If a node scheduled on a daily basis is generating dry-run instances unexpectedly, go to the node's Properties tab > Schedule section and check whether Recurrence is set to Dry Run.
T+1 means the scheduling system runs nodes on the second day using data generated on the current day. A daily node configured with T+1 runs today using yesterday's data.