Skip to main content

Troubleshooting

Short fixes for problems we have reproduced on real builds. Each entry: what you see, what it means, what to do.

"Cloud import is not available in this build" (slim build)​

The slim package bundles only kubectl and Helm. Cloud Provider Import uses only the bundled CLIs, not the ones on your PATH, so on slim it does not run: choosing AWS EKS shows "Cloud import is not available in this build: the aws CLI is not bundled." and no List clusters button, even if aws works in your terminal.

What to do: install the full-offline package, which includes the cloud CLIs. Or skip the import: Connect Cluster > Paste kubeconfig YAML > Import clusters. The cluster then appears under Managed Clusters.

openSUSE: "Package header is not signed!"​

The .rpm is not signed, so zypper stops at the signature check. This is expected for current builds. If you trust the download source, install it with:

sudo zypper install --allow-unsigned-rpm ./<file>.rpm

Linux: the app does not update itself​

Linux builds do not update themselves. When a new version is out, the app shows a notification: "Rozoom 0.26.3 is available" (the number is the new version). It says: "Rozoom doesn't update itself on Linux. Download the new .deb from the release page and install it the way you installed this one." (on an .rpm install it says .rpm). Click Open release page, download the installer and install it over the old one.

A cluster shows "Unreachable" after Check​

Check asks the cluster's API server whether it answers. "Unreachable" means it did not. The line under the badge gives the reason: "Connection refused", "No answer from the server" or "Host name not found".

An address that never answers can keep the row on "Checking..." for about 25 seconds before the result appears. This is normal.

What to do: check the server: address in your kubeconfig, then your VPN or network, and click Retry.

On the cluster card the same state shows as UNREACHABLE with the alert Cluster unreachable, where the reason can also be "No route to the cluster address", and Last successful contact gives the time the cluster last answered. See the health card.

Where the logs are​

Help > Open Logs Folder opens the folder with logs.log. Attach this file when you report a problem. A kubeconfig that fails to parse is logged as the position and the error name only, without the content of the file.