I'd be lying if I said this maintenance was fun, but the plan was almost perfect and the skills and experience of our engineers covered the gaps. It was certainly exciting, but I think we all agree that's not such a good thing. Regardless, the maintenance was completed in line slightly ahead of our expectations.
We don't have good numbers on the exact outage time, but it was significantly under the hour predicted. The reason we don't know for sure is that certain network bisects that we had to perform cause instability in our applications. Our current "best guess" is that complete service was restored after a total of 40 minutes of downtime and most service was restored after closer to 30 minutes. We're examining the cause of that instability which should improve the overall reliability of the service.
All other elements of the maintenance went as expected. Our tests indicate that everything is functioning correctly and we've removed the problematic hardware.
If you have any questions about this maintenance, please don't hesitate to reach out to us.