My Jobs were running just fine, but for some reason all of a sudden they all started failing.  When I looked into it I saw it was an error due to an init script error (we do use an init script).

run failed with error message
 Cluster 1117-045226-lfvc8fde was terminated while waiting on it to be ready: Cluster 1117-045226-lfvc8fde is in unexpected state Terminated: INIT_SCRIPT_FAILURE(CLIENT_ERROR): instance_id:i-009209e91e8cc95e2,databricks_error_message:Cluster scoped init script /test | 290d381e-8eb4-4d6a-80d4-453d82897ecc/testinit.sh failed: Script exit status is non-zero.

What is strange is, I tried just putting in a blank init script and I still got the error above. 

If I completely remove the init script, then everything works fine.

Why would a blank init script cause this error too?  Is something broken in the backend?