sosso edited a comment on issue #14205: URL: https://github.com/apache/airflow/issues/14205#issuecomment-821326846
+1 for us on this issue as well, I think? Very strangely, we see the most recent run for a DAG have its *run* be set to 'running', but the only task in the DAG be a clear success:  This is a catchup=False DAG, where the only task runs in a pool, and there is *nothing* in the Scheduler log for this DAG for two hours (the DAG runs is supposed to run every 5 minutes) about why it can't schedule this DAG. No "max active runs reached", no "no slots available in pool", nothing. It's like the scheduler forgot this DAG existed until we rebooted it. *edit* This has happened again, here are the relevant log lines (cherry-picked via grep): ``` [2021-04-16 17:44:01,418] {{base_executor.py:82}} INFO - Adding to queue: ['airflow', 'tasks', 'run', 'mls_ivbor_smart_incremental_v1', 'streaming-importer', '2021-04-16T17:30:00+00:00', '--local', '--pool', 'ivbor', '--subdir', '/efs/airflow/dags/mls_incrementals_i.py'] [2021-04-16 17:44:04,638] {{scheduler_job.py:1206}} INFO - Executor reports execution of mls_ivbor_smart_incremental_v1.streaming-importer execution_date=2021-04-16 17:30:00+00:00 exited with status queued for try_number 1 [2021-04-16 17:44:04,661] {{scheduler_job.py:1226}} INFO - Setting external_id for <TaskInstance: mls_ivbor_smart_incremental_v1.streaming-importer 2021-04-16 17:30:00+00:00 [queued]> to 2c9ee22a-ad2b-4255-846a-85896fa517ed [2021-04-16 17:44:43,326] {{scheduler_job.py:1206}} INFO - Executor reports execution of mls_ivbor_smart_incremental_v1.streaming-importer execution_date=2021-04-16 17:30:00+00:00 exited with status success for try_number 1 [2021-04-16 18:00:18,075] {{dagrun.py:445}} INFO - Marking run <DagRun mls_ivbor_smart_incremental_v1 @ 2021-04-16 17:30:00+00:00: scheduled__2021-04-16T17:30:00+00:00, externally triggered: False> successful ``` So the task succeeded at 17:44, but the dagrun wasn't set to success until 16 minutes later? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
