patrick0057 / cluster-agent-tool Goto Github PK

1. Create function / option to delete all rancher agents.

docker ps -a | grep rancher-agent | awk '{print $1}' | xargs docker stop; docker stop kubelet ; docker stop kube-proxy && sleep 1
docker ps -a | grep rancher-agent | awk '{print $1}' | xargs docker rm

tool does not correct detect cluster id for rancher local cluster

Fork working copy of jq so that upstream changes don't potentially break the script

make check for dos2unix

remove jq dependency

setup regex to cleanup input for -s so that you can use just the hostname, have a trailing / or not and also use https:// or http://

add the ability to use api keys

At the moment I have baked in most of kubecert.sh into this script to make things easier. This isn't really all that efficient as if something changes in how we get our kubeconfig for the local cluster, I need to update two scripts now.

I need to set up the script so that it downloads kubecert from github and runs it rather than performs the function itself.

-l option should also make jq a local binary

Create a friendlier error message when the Rancher URL returns default backend instead of the expected reply.

This can happen when a user is updating their server URL but only changed it on the helm upgrade command and not in the web interface "settings" section.