Preflight commands are run before every update. These commands are here to ensure the cluster has the resources needed.
A diagnosis tool is also available on the Admin Console. With it, you can generate a support bundle that can identify common issues. You can also send it to GitGuardian to be analyzed. Sensitive information can be automatically redacted.
Once the analysis is done, you can get a preview of it by browsing the different files it contains. That is the support bundle.
In case of a Helm-based installation, the KOTS Admin Console isn't available. In such scenario, you need to install a client-side utility, packaged as a kubectl plugin and distributed through the krew package manager.
Follow these steps to get the support bundle for a Helm installation:
- Install the krew plugin.
- Install the support-bundle utility with
kubectl krew install support-bundle.
kubectl support-bundle --load-cluster-specs --namespace gitguardian-namespace, replacing
gitguardian-namespacewith the namespace where the application is installed.
This will generate the support bundle as a .tar.gz archive in your current directory. You can then send it to GitGuardian support.
When you need to debug an offline Kubernetes cluster, you can use host collectors to generate a support bundle without the Admin Console available.
First, you need to install the support bundle on a host in the cluster you need to debug:
curl -L https://github.com/replicatedhq/troubleshoot/releases/latest/download/support-bundle_linux_amd64.tar.gz | tar xzvf -
You can generate the support bundle:
./support-bundle --interactive=false https://raw.githubusercontent.com/replicatedhq/troubleshoot-specs/main/host/cluster-down.yaml
You may need to run the above command with
sudo if your current user doesn’t
have the access to gather information for a specific collector.