Argocd-related issues and solutions

breaking port-forward issues, this is not just for argocd alone, when you encounter breaking port forward issues on kubectl

just open another terminal and put the service up regardless using the following syntax

while true ; do nc -vz 127.0.0.1 8080 ; sleep 10 ; done

so 127.0.0.1 is your localhost you are port forwarding to and the 8080 port is something you can change to any port you are trying to port forward to.

so the command is while true keeps the netcat connection on a loop non-stop, and the sleep 10 is needed because the while true is a hot loop, so nc -vz host port, helps keep querying the server for you to see if the connection is dead or successful, if not it continues.

hence keeping the connection alive

failed to replace object: Service "argocd-server"

fixed by just remove force_update = true

Argo CD error="server.secretkey is missing"

fixed by kubectl rollout restart deploy/argocd-server -n argocd

Argo CD error="error getting cached app resource tree: cache: key is missing"

Actually it's enough to restart only applicationController's statefulset, ie. kubectl rollout restart statefulset -n argocd argocd-application-controller