Harsh Kumar
11/06/2025, 5:27 AMkapa.ai
11/06/2025, 5:27 AMabctl, there are a few important steps you need to follow after loading your custom image. According to the [connector development documentation](https://docs.airbyte.com/platform/contributing-to-airbyte/developing-locally#connector-contributions), after loading your image with kind load docker-image, you need to: 1. Update the connector version in the Airbyte UI: - Visit http://localhost:8000/ - Log in - Click Workspace settings > Destinations - Click the edit button for your S3 destination - Update the version number to use your docker image tag (default is dev) - Click Change to save the changes 2. Clear the connector specification cache: The Airbyte Server caches connector specifications for performance reasons. If you've updated your connector, you need to clear this cache: - Go back to Workspace settings > Destinations - Click the edit button for your S3 destination - Leave the version set to dev - Click Change to save the changes, which will refresh the dev connector's spec This cache clearing step is crucial and might be why your code changes aren't being triggered even though the pod is using your new image. Additionally, make sure you've verified that your image was built correctly by running: docker images ls | grep airbyte/destination-s3:dev After following these steps, when you run a sync with that connector, it should use your local docker image with your custom code changes. If you're still experiencing issues after these steps, the knowledge sources don't contain specific troubleshooting information for this scenario, so you may need to check your connector logs more carefully or reach out to the Airbyte community with specific error messages.