Currently, there is only one PostgreSQL server, stucked in Resuming. Because of this error: postgres: max_wal_senders must be less than max_connections
You can fix it, and restart Scalarizr, which will bring Slave back into Running.
Another option is to terminate this server, and Scalr will launch new Master.
Can you please check logs in AWS CloudTrail? Was it API call to detach or not. There is a possible case when volumes got detached by Amazon for maintenance reasons. But would be great to check who actually sent a detach volume API call.