So... yeah... we found a guest to talk to us about zero day markets and exploit pricing. He's... credible, I think?
(David and I tacked an "SCW: After Dark" bit at the end because he couldn't make the original interview).
@rennerocha GRU had a status-paged incident last night --- it wasn't out of capacity, but rather a deployment broke placement logic. The problem was exclusive to GRU. It'll be on the infra log this week.
I understand that this would bother if you were deploying more instances of a critical app that absolutely had to be in GRU, but it's the nature of what we do that individual regions --- we have almost 40 --- are going to experience transient issues. This was one we had a lot of control over, but >half of what we run into isn't like that.