Whilst testing the Hot-Cold HA deployment we did the following:
1. Deployed a node on one machine (devMode=false)
2. Registered the node properly with the doorman
3. Copied files to cold backup node.
4. Stopped primary node and started secondary node.
5. This seemed to create a fresh node identity??? How in devMode=false????
6. Then correctly copied identity from primary node and restarted.
7. Now DB contains two identities for itself and the node cannot be started and crashes out!!!!
Good logging of error is required when this happens so the node operator / administrator knows what the issue is
Attached the log of this sordid tale
Fix should focus on improved logging of the reason for the crash
I am working on a fix
Moving to Core since Stefano seems to have claimed it