Idan Noti
12/09/2023, 4:13 PMMarvin
12/09/2023, 4:14 PMIdan Noti
12/09/2023, 4:15 PMMarvin
12/09/2023, 4:15 PMIdan Noti
12/09/2023, 4:18 PMMarvin
12/09/2023, 4:18 PMIdan Noti
12/09/2023, 4:20 PMMarvin
12/09/2023, 4:20 PM/api/flow_runs/count
REST API endpoint and whether it differentiates between parent flow runs and subflow runs in its count.
However, based on the behavior of subflow runs being treated like normal flow runs, as mentioned in the documentation, it is possible that the endpoint counts both parent flows and subflows together. If you require a count that is specific to only parent flows or subflows, you may need to look for or request additional filtering capabilities in the API or handle the differentiation in your own code after fetching the flow run data.Idan Noti
12/09/2023, 4:27 PMIdan Noti
12/09/2023, 4:34 PMMarvin
12/09/2023, 4:35 PM@flow
decorator in Python. They are templates or blueprints for the processes you want to run. A flow on its own is just a definition; it does not execute any logic until it is run.
- Flow Runs: A flow run represents a single execution of a flow. Every time a flow is triggered to run, whether manually or through a scheduler, a flow run is created. Flow runs are instances of the flow that can be monitored and tracked. They capture the state of the workflow execution for observability. The Prefect API monitors flow runs, tracking each task's state within the flow and any subflows that might be called.
In short, a flow defines what to do, and a flow run is an instance of doing it.