The incident states that it's a power issue (like they said last night), but none of my servers were rebooted last night, so it's clearly something affecting the network level, and it seems to be affecting all servers.
I look forward to their full postmortem about the issue (they promised that after the first outage, but the second happened before they'd published anything).
Sounds like they had enough battery to safely pause all the instances and then resume them when power came back.
I wouldn't expect you to be able to monitor your provider's power but you can monitor your own power and perhaps you have your own UPS in a CoLo style setup or at least a promise of some sort. Perhaps your stuff isn't important enough for that - that's your risk assessment.
Your monitoring should be able to discern power vs networking.
https://news.ycombinator.com/item?id=32325224
"Linode London data-center down for 2.5 hours in 26"
