# Airflow DAG start_date and Manual Triggers

> When manually triggering a DAG, Airflow may skip task execution if the trigger

- Published: 2026-01-27
- Updated: 2026-03-22
- Category: devops
- Tags: devops, airflow
- Page: https://brandonwie.dev/posts/airflow-dag-start-date

I manually triggered a backfill DAG and it showed "success" instantly — 0.02 seconds, no task logs, no Docker containers started, no Slack notifications. The DAG appeared to run but did absolutely nothing. I checked task dependencies, the Docker image, the Celery worker — everything looked fine. The problem was much simpler: the DAG's `start_date` was set to a future date.

When Airflow's trigger date falls before the DAG's `start_date`, it silently marks the run as successful without executing any tasks. No error, no warning, no indication that anything went wrong.

## The Trap

Here's the code that caused the issue:

```python
with DAG(
    dag_id="my_dag",
    start_date=datetime(2026, 1, 29),  # Future date
    ...
) as dag:
```

When manually triggered on January 27th:

- Trigger date (2026-01-27) is before `start_date` (2026-01-29)
- Airflow marks the DAG run as success **without executing tasks**
- Total duration: ~0.02 seconds (no actual work done)

## How to Spot This

The symptoms are distinctive once you know what to look for:

- DAG run shows "success" immediately (under 1 second duration)
- No task execution logs at all
- No Docker containers started (for DockerOperator tasks)
- Scheduler logs show no "tasks up for execution" entries
- Callbacks don't fire (no Slack notifications)

The key giveaway is the duration. A real DAG run takes seconds to minutes. A skipped run completes in milliseconds.

## The Fix

Use a safe past date that will always be before any trigger date:

```python
with DAG(
    dag_id="my_dag",
    start_date=datetime(2024, 1, 1),  # Safe past date
    ...
) as dag:
```

That's it. Change one line. The `start_date` just needs to be far enough in the past that no manual trigger will ever fall before it.

## Which Approach to Use

| Approach               | Pros                               | Cons                       |
| ---------------------- | ---------------------------------- | -------------------------- |
| `2024-01-01`           | Simple, always works               | Looks arbitrary            |
| Actual deployment date | Semantically meaningful            | May break manual triggers  |
| Future date            | Prevents accidental scheduled runs | **Breaks manual triggers** |

I recommend `datetime(2024, 1, 1)` as a standard convention across all DAGs. It's predictable, always safe, and eliminates this class of bug entirely.

## Won't This Cause Unwanted Scheduled Runs?

No. The `start_date` primarily controls when **scheduled** runs begin, but for DAGs with `catchup=False`, Airflow won't create runs for past dates anyway. Using a past `start_date` doesn't trigger a flood of backfill runs — it only ensures manual triggers always work.

## Takeaway

If an Airflow DAG shows "success" in under a second with no task logs, check the `start_date`. A future `start_date` causes Airflow to silently skip task execution on manual triggers. Set `start_date` to a safe past date like `datetime(2024, 1, 1)` across all your DAGs. Combined with `catchup=False`, this prevents both the silent-skip bug and unwanted historical runs.

## References

- [Airflow DAG Runs Documentation](https://airflow.apache.org/docs/apache-airflow/stable/core-concepts/dag-run.html)
