Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Prefect Community

<@ULVA73B9P> Is there a supported way to suppress the per-node logs/events produced by
  PrefectDbtRunner.invoke() when running dbt ls?

  Redirecting stdout/stderr does not suppress them because they appear to be emitted through
  Prefect’s logging callbacks. We want to keep using PrefectDbtRunner, rather than calling
  dbt’s native dbtRunner directly. Is there a setting or callback configuration to disable
  these logs while still obtaining result.result?

:white_check_mark: thought for 85.9 seconds

Yes — *if you’re on a recent Prefect version*, `PrefectDbtRunner` has a way to suppress those callback-driven per-node logs/events:

```python
from prefect_dbt import PrefectDbtRunner

runner = PrefectDbtRunner(_disable_callbacks=True)
result = runner.invoke(["ls"])

# You still get the dbt result object
print(result.result)
```

A few important notes:

- `_disable_callbacks=True` prevents `PrefectDbtRunner` from registering the Prefect/dbt event callbacks that emit the per-node logs.
- You still get the invocation result, including `result.result`.
- This is an *internal/private* constructor argument, not a documented public setting, so treat it as version-sensitive.

I verified this from the source:
- <https://github.com/PrefectHQ/prefect/blob/main/src/integrations/prefect-dbt/prefect_dbt/core/runner.py#L157|`_disable_callbacks` in `PrefectDbtRunner` constructor>
- <https://github.com/PrefectHQ/prefect/blob/main/src/integrations/prefect-dbt/prefect_dbt/core/runner.py#L1221-L1236|callback creation logic in `invoke()`>

So the short answer is:

- *Supported public setting/config?* No, I couldn’t find a documented public callback/logging config for this.
- *Practical way while still using `PrefectDbtRunner`?* Yes: `PrefectDbtRunner(_disable_callbacks=True)`.

Example:

```python
from prefect_dbt import PrefectDbtRunner

runner = PrefectDbtRunner(
    _disable_callbacks=True,
)

result = runner.invoke(["ls", "--select", "my_selector"])

# Consume dbt output as usual
print(result.result)
```

If you want, I can also help you:
- check whether your installed `prefect-dbt` version includes this flag, or
- suggest a safe fallback pattern if you’re on an older release where `_disable_callbacks` is not available.