The outage sucked for everyone. The root cause also feels like something they could have caught much earlier in a canary rollout from my reading of this.
All that said, to have an outage reported turned around practically the same day, that is this detailed, is quite impressive. Here's to hoping they make their changes from this learning, and we don't see this exact failure mode again.
All that said, to have an outage reported turned around practically the same day, that is this detailed, is quite impressive. Here's to hoping they make their changes from this learning, and we don't see this exact failure mode again.