Capturing logs in authentik

When troubleshooting issues in authentik, reviewing the event logs can be invaluable. These logs provide continuous output, helping to diagnose problems effectively.

Adjusting log levels

The server and worker containers support multiple log levels: debug, info, warning, and error. By default, the log level is set to info.

To modify the log level, follow the steps for your platform

docker-compose

1. Add the following environment variable to your docker deployment:

   AUTHENTIK_LOG_LEVEL=debug

2. Recreate your containers to apply the changes.

Kubernetes

1. Add the following configuration to your values.yml file:

   authentik:
       log_level: debug

2. Recreate your containers to apply the changes.

Enabling trace mode

danger

The trace log level provides deeper insights, but be aware that using trace logs can expose sensitive information, including session cookies. Handle these logs with extreme caution and avoid using trace unless absolutely necessary.

To enable trace logging, follow the platform-specific steps below:

docker-compose

1. Add the following environment variable to your docker deployment:

   AUTHENTIK_LOG_LEVEL=trace

2. Recreate your containers to apply the changes.

Kubernetes

1. Modify your values.yml file:

   authentik:
       log_level: trace

2. Recreate your containers to apply the changes.

Viewing past logs

To review historical logs, you can use the --since option with both docker logs and kubectl logs. This option allows you to specify either a duration (e.g., 1m30s, 3h) or a specific timestamp (e.g., 2006-01-02T07:00, 2006-01-02) to view logs generated after that point in time.

For more details, see the docker logs documentation and kubectl logs documentation.

docker

To retrieve logs from a specific timeframe, use:

docker logs <container_name_or_id> --since 5m

Kubernetes

To fetch logs from a Kubernetes pod:

kubectl logs --since 5m <pod_name>

Streaming logs in real-time

To continuously monitor logs, use the --follow (-f) option. This will stream log output to your terminal until manually stopped (Ctrl + C or closing the terminal).

docker

To follow logs in real time:

docker logs <container_name_or_id> -f

Kubernetes

To stream logs from a Kubernetes pod:

kubectl logs -f <pod_name>