Wait for etcd to startup before trying to connect to it

[bodom0015]

Problem

etcd sometimes takes a few seconds to startup. In the meantime, the apiserver goes into CrashLoopBackoff while attempting to connect to etcd.

Approach

Use a simple bash loop to wait for etcd to startup before attempting to connect to it

How to Test

1. Deploy the new image via the Helm chart
2. Check the logs of the apiserver: kubectl logs -f deploy/workbench -c apiserver
   • You should see that the apiserver waits for etcd to start before continuing its own startup

[bodom0015]

Included as a part of #336