Issue Summary

Starting Crowd immediately after the stop command causes duplicate nodes to appear in cluster

Environment

(Optional - If Applicable)

• Tested with Postgres as data source connection
  *

  Steps to Reproduce

1. Stop Crowd by issuing stop_crowd.sh
2. After the execution, again run start_crowd.sh e.g

   6-Aug-2019 17:34:19.006 INFO [main] org.apache.catalina.core.StandardServer.await A valid shutdown command was received via the shutdown port. Stopping the Server instance
   2019-08-16 17:35:00,235 localhost-startStop-1 INFO [com.atlassian.crowd.startup] Starting Crowd Server, Version: 3.4.5 (Build:#1070 - 2019-05-23) in Data Center mode
   

3. After some time node is automatically removed

Expected Results

There shouldn't be duplicate nodes present in the configuration and clustering page

Actual Results

The below exception is thrown in the xxxxxxx.log file:

2019-08-16 17:35:00,245 localhost-startStop-1 INFO [crowd.console.listener.StartupListener] Joined a cluster with nodes:
2019-08-16 17:35:00,245 localhost-startStop-1 INFO [crowd.console.listener.StartupListener] 1.  1726c1f4-5204-4f07-870b-3018e410bcf6
2019-08-16 17:35:00,245 localhost-startStop-1 INFO [crowd.console.listener.StartupListener] 2.  79ed6362-83a3-411b-8d1b-8ec0bf031c0a

Notes

(Optional - If Necessary)

Workaround

Required, if there is no workaround please state:
A sufficient time of around 60 seconds between start and stop doesn't replicate the problem