Documentation survey

Extension Troubleshooting

This section provides troubleshooting steps for any interoperability issues between {stackstate-product-name} and the Rancher UI extension.

Collect a HAR (Http ARchive)

Http ARchive collects details on requests that are sent by the browser to Rancher and {stackstate-product-name}. It also logs the responses received. This archive is specific to the page/browser tab with the error. It does not need access to the backing Rancher or {stackstate-product-name} instances (though their logs may be useful at a later stage).

  • Open the Network tab in the "Developer Tools" panel. You can access this by using F12. The network tab shows a log of all requests that are made.

  • Clear the log and do a (hard) refresh of the page that is displaying the problematic behavior.

  • There should now be a large number of requests to both Rancher and SUSE Observability visible.

  • Export/Download/Save the HAR - different browsers use different names here.

Cluster is unobserved

For the cluster health status to be visible in Rancher, there needs to be alignment between these configurations:

  • The (display) name of the cluster in the Rancher UI.

  • The stackstate.cluster.name set on the agent deployed on the cluster

  • The cluster name in the configuration of the Kubernetes stackpack in SUSE Observability

A mismatch will tend to show up as "no data received" in the cluster overview in Rancher.

Connection failed

The configuration for SUSE Observability may sometimes not yield a valid connection in the Rancher UI due to the following reasons:

  • The Rancher host is not configured as an allowed origin for SUSE Observability. You can fix this by adding it to the stackstate.allowedOrigins list in the helm values, or by passing --set stackstate.allowedOrigins={"https://<rancher-url>"} to the deployment command.

  • SUSE Observability is not accessible from the browser. For example, the root certificate has not been installed.