Should be back online now. I'm going to schedule nightly reboots for each server, which should take care of most downtime issues without a human having to push a reset button. I do like the idea of ordaining some folks to help monitor and kick boxes when needed. I have a half built docker system that was supposed to replace the current infrastructure (which would be much better at auto healing, isolating instances, allowing userland filesystems, etc...), but time has been at a premium lately - gotta feed the family 🙂