Troubleshooting

Overview

This page contains a list of common errors you may encounter using the Humanitec platform and how to resolve them.

I can’t see my GitHub repositories

Here are some reasons why your module repositories may not appear when you click the Refresh the list button:

GitHub organization setup

If you can’t see any of your repositories, then the administrator responsible for connecting your organization to the platform may have registered without providing the proper third-party access to your organization. See the GitHub organizations page for more information.

Repository requirements

Does your repository meet the following requirements?

  1. Dockerfile: Each module repository must include a Dockerfile.
  2. Git tag: Your repository must have a tagged commit that includes the Dockerfile.

I can’t add my repository to my app

Repository requirements

Does your repository meet the requirements? See the section above for more details.

GitHub API connection error

Even if you got all of the configurations for your module right, the platform may still fail to connect to the GitHub API.

  • If this happens, then refresh the page and try again.
  • Check https://githubstatus.com to see if GitHub is available.
  • If the error persists, then please send a help request in the platform interface.

Deployment failed (pre-configured Humanitec cluster)

Here are some common reasons why your environment could fail to deploy on the platform using the platform-managed cluster. See below for troubleshooting around deploying to your own cluster.

Kubernetes connection error

The “Kubernetes connection error” may occur if the platform fails to connect to the Kubernetes API, or if the connection succeeds but Kubernetes fails to execute the deployment request.

  • Try clicking the deployment button again.
  • If the error persists, then please send a help request in the platform interface.

Build error

If no image exists for a module, then the platform must build it first. Check to see if an image was successfully built for your module by going to your app, clicking on the module, and going to its Build log section.

Broken pod

When Kubernetes begins to create pods for your modules, the pod could break due to an improper configuration in the module’s Dockerfile or its Platform configurations.

Check to see if your module’s Dockerfile runs properly. You can see this in the platform interface by going to your app, clicking on the module, and going to its Container log section.

Failure to connect to external resource

If your app depends on a resource not deployed on the platform, such as an external database or API, and the platform cannot connect to it, then the deployment will fail.

  • We recommend enabling logging in your app. You can view these logs in the platform interface by going to your app, clicking on the module, and then going to its Container log section.

Deployment failed (self-hosted cluster)

Please double check to make sure that your cluster follows the prerequisites set forth in the documentation:

You should use the provided cluster preparation script to install ingress and certmanager.

I can’t access my module

If you exposed your module with a public IP but get an error when you try to access the URL, then check the module’s Details page to see if the ports you exposed match the ones exposed in the module’s Dockerfile.

“No build found” for my module

This message may display when you add a module to your app and the platform cannot find an image for that module in its registry.