Make Bamboo more resilient to network/database failures

On a remote agent Bamboo uses a wrapper program that monitors and restarts the agent java process if it becomes unresponsive.

There is currently no such mechanism on Bamboo server/data center node.

In a scenario where multiple Bamboo data center nodes are connected to a remote database and the network between the database and Bamboo nodes goes offline for a long time, say, 10 minutes, the Bamboo nodes do not recover.

This feature request is to add better resilience to external network/ database and have Bamboo continue after the outage.

Can we get a wrapper or similar functionality so that the Bamboo instance can automatically restart/recover from a network/database outage?