we don't have any such automation for Server to get notified if the agent process goes down - in that case, you would need to implement monitoring on the infrastructure level, e.g. tracking logs and metrics on your agent's VM and sending something like Grafana alert if the agent no longer emits health checks or agent process dies