Ipregistry - Notice history

A surge in load for a replicated database cluster saturated available connections. A recent backend library change introduced an overly aggressive connection timeout, which led to cascading timeouts in a subset of regions, while others were unaffected.

We sincerely apologize for the disruption caused. We adjusted connection settings, and put additional safeguards in place to reduce the risk of similar issues. We are also improving testing and monitoring around connection handling to better detect and mitigate such situations in the future.