The GitHub Actions job "Tests (AMD)" on airflow.git/fix-ti-history-retry-metadata has failed. Run started by GitHub user kalluripradeep (triggered by kalluripradeep).
Head commit for run: 6a272edb658de2be1c01008ef7f5750f9046754a / Pradeep Kalluri <[email protected]> Fix KubernetesExecutor retry losing hostname When a task fails from a non-RUNNING state (QUEUED, SCHEDULED, or DEFERRED) — e.g. a Kubernetes pod OOMKilled before the task process starts — the scheduler's executor-event path calls handle_failure(). Previously, prepare_db_for_next_try() was only invoked when ti.state == RUNNING, so TaskInstanceHistory was never recorded for non-RUNNING retries, losing the hostname and start_date and causing 'http://:8793 No host supplied' in the log UI. Fix: change the guard in fetch_handle_failure_context() from if ti.state == RUNNING to if ti.state != RESTARTING This ensures history is recorded for every retry-eligible failure path (RUNNING, QUEUED, SCHEDULED, DEFERRED) while still excluding RESTARTING, where prepare_db_for_next_try() was already called during the clear operation. Also fix taskinstancehistory.py to preserve an existing end_date instead of always overwriting it with utcnow(), and update the execution-API retry handler to capture end_date and rendered_map_index before calling prepare_db_for_next_try(). closes: #65366 closes: #67238 Report URL: https://github.com/apache/airflow/actions/runs/28033404676 With regards, GitHub Actions via GitBox --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
