Troubleshooting¶

Obscure task failures¶

Task state changed externally¶

There are many potential causes for a task’s state to be changed by a component other than the executor, which might cause some confusion when reviewing task instance or scheduler logs.

Below are some example scenarios that could cause a task’s state to change by a component other than the executor:

If a task’s Dag failed to parse on the worker, the scheduler may mark the task as failed. If confirmed, consider increasing core.dagbag_import_timeout and dag_processor.dag_file_processor_timeout.
The scheduler will mark a task as failed if the task has been queued for longer than scheduler.task_queued_timeout.
If a task instance’s heartbeat times out, it will be marked failed by the scheduler.
A user marked the task as successful or failed in the Airflow UI.
An external script or process used the Airflow REST API to change the state of a task.

Process terminated by signal¶

Sometimes, Airflow or some adjacent system will kill a task instance’s TaskRunner, causing the task instance to fail.

Below we discuss a few common cases.

Dag run timeout¶

A dag run timeout can be specified by dagrun_timeout in the dag’s definition. The task process would likely be killed with SIGTERM (exit code -15).

Out of memory error (OOM)¶

When a task process consumes too much memory for a worker, the best case scenario is it is killed with SIGKILL (exit code -9). Depending on configuration and infrastructure, it is also possible that the whole worker will be killed due to OOM and then the tasks would be marked as failed after failing to heartbeat.

Lingering task supervisor processes¶

Under very high concurrency the socket handlers inside the task supervisor may miss the final EOF events from the task process. When this occurs the supervisor believes sockets are still open and will not exit. The workers.socket_cleanup_timeout option controls how long the supervisor waits after the task finishes before force-closing any remaining sockets. If you observe leftover supervisor processes, consider increasing this delay.