Ubuntu 14 MySQL/Jetty server - uninitiated shutdown
I have a 16gb droplet running ubuntu 14.04.1. It runs MySQL 5.7, NGINX 1.6, and jetty 9. It has 0 security updates waiting to be installed and iptables block access to everything except for 80 and 443.
This server has had no new software installed, cron jobs activated, or serious updates in the past year, but all of a sudden it has started shutting down completely unexpectedly. Twice now in the past two days, I have been notified by external server monitoring tools that the server is not responding. I have to manually go into the DO control panel and startup the droplet, which has completely shut down.
I cannot find anything related to the shutdown in syslog, kern.log, nginx logs, mysql logs, or jetty logs. Nothing about "Shutdown initiated" or "Stopping services" or "Disconnecting ports" etc. There are no errors or panics written to any log that match any time close to the time that the system is shutdown. There are no logs in the DO control panel related to the shutdown events.
I have New Relic server monitoring installed, and all system utilizations (cpu, network, disk, memory) are way under max - cpu is less than 5% utilized, disk is 80% free and access is extremely minimal with very small spikes, and network is less than 5mbps. Here is a snapshot of my new relic control panel http://note.io/1F9jg2Y
I have brought this up with Digital Ocean support, who has said that they cannot see anything wrong in the hypervisor, but I am having a lot of trouble identifying what the cause could be. This is the master database server in a production system, so downtime is a serious problem.
Have any of you seen something similar, or know a step that I should take next?