Skip to content
Pydantic Docs

Troubleshooting

Self-hosted Logfire sends its own internal telemetry to the logfire-meta organization. Start there when the deployment is running but behavior is unclear.

Access the Meta Organization

On first install, the chart creates a frontend token for the logfire-meta organization.

Read the token from the release namespace:

Terminal
kubectl -n logfire get secret logfire-meta-frontend-token \
  -o "jsonpath={.data.logfire-meta-frontend-token}" | base64 -d

Open the meta project with your hostname and token:

https://logfire.example.com/logfire-meta/logfire-meta#token=LOGFIRE_META_FRONTEND_TOKEN

After you get access, create an invite link for your own user from Settings > Invite and give it the Admin organization role.

To find errors in the meta project, use this query filter:

level >= 'error'

No Traces in the Meta Organization

If the meta project has no traces, check the services involved in ingestion and internal telemetry export:

Terminal
kubectl -n logfire get pods
kubectl -n logfire logs statefulset/logfire-ff-ingest
kubectl -n logfire logs deployment/logfire-ff-ingest-processor
kubectl -n logfire logs deployment/logfire-otel-collector
kubectl -n logfire logs deployment/logfire-ff-query-api

Common causes are PostgreSQL connectivity, object storage connectivity, image pull failures, or incorrect public URL/TLS settings.

ErrImagePull or ImagePullBackOff

If pods cannot pull images, check that:

  • The image pull secret exists in the same namespace as the Helm release.
  • imagePullSecrets in your values file references that secret name.
  • The secret was created from the current image pull key provided by Pydantic.

The production requirements page shows the expected secret shape.

PostgreSQL Migration Failures

Backend and FusionFire migrations run before the main workloads are ready. If Helm times out or reports a failed migration job, inspect the migration logs:

Terminal
kubectl -n logfire logs job/logfire-backend-migrations
kubectl -n logfire logs job/logfire-ff-migrations

Check that:

  • postgresDsn points at the crud database.
  • postgresFFDsn points at the ff database.
  • Dex storage points at the dex database.
  • Database users have owner permissions for their databases.
  • The Kubernetes cluster can connect to the PostgreSQL host and port.

Support

When asking for support, include:

  • Chart version.
  • Kubernetes version.
  • A sanitized copy of your values file.
  • Relevant pod, job, or Helm error messages.