A major outage afflicted our hosted network controllers, our web sites, and the hosted ZeroTier Central API today. Root servers (and thus private controllers) were not affected. Some users may not have noticed unless they tried to use our web or API, while others might have experienced problems connecting to networks from devices offline long enough to no longer have up to date network certificates of membership.
This one resulted from a very serious failure at our cloud provider, making it tempting for us to say it's not our fault. We're not going to do that. Our goal is to offer enterprise class reliability with all our products and services. That means any outage is our "fault" and is a chance to improve. We want to build a system where these kinds of things just don't matter.
That's all for now. This will be discussed in more detail in our upcoming blog on plans for 2018 and beyond.