Symptoms

Preview of APS Connector fails:

"Failed to deploy backend for <connector_name>. Readiness check failed. Verify that correct health check URL is provided and that it responds with status code 200. Connector's standard output: \n "

Cause

There can be multiple reasons.

Resolution

  1. Make sure that your connector backend meets all the requirements specified at http://help.aps.odin.com/publish e.g.:

    a. It is important that request-response (access.log style) logs must be written to STDOUT. In this case it will be easier to understand why the deployment fails and how the error can be fixed.

    b. Such error can be shown if connector requires more than 256Mi RAM. To check it, you may deploy connector locally (or on your kubernetes cluster) and check its memory usage via docker ps.

  2. Deploy the docker image locally and make sure that it is accessible via /connector path (all the methods need to be now under /connector/...). Detailed info about API requirements can be found here: https://help.aps.odin.com/api

  3. If you created a new version of docker image for connector backend, it is necessary to upload it with a new tag, not replace the older version with a new one. Otherwise, cached older version will be still used at APS Connect kubernetes cluster, and the preview will fail

  4. If you develop you Connector on the basis of fallball-connector, make sure that you are using the latest version mentioned in release notes: http://help.aps.odin.com/releases/. To find it click the latest release and find the version of Fallball Connector (Simulator).

Internal content