Re:Post-mortem for Partial Cloudflare Outage on October 25, ...
by londons_explore on Wednesday October 26, 2022 @04:19AM
(#33341105)
But there is a good chance that was a 100% request failure rate for a specific service. And when you have 100% of requests failing for a service, you ought to be thinking about doing a fast rollback for that service as long as you have confidence that such a fast rollback won't take other infrastructure down.
They had a 5% request failure rate at the peak of this outage. So a botched rollback could definitely do more harm than good if the impact isn’t 100% understood.