Elevated Timeouts

Update timeline

1. resolved May 20, 2026, 08:26 AM UTC

   We have identified some major components that have been contributing to these downtime events, and we're going to be working to get them resolved as soon as we can. In this instance, we had multiple data-intensive things happening simultaneously. As they piled up, they started to get in eachother's way and create a situation where many requests took much longer than normal to complete. In order to combat this, we temporarily pulled back how much background processing was happening—things like broadcast sending speed and sequence delivery were intentionally slowed to a halt to allow the system to recover. We were able to collect a lot of data during this downtime event, and will be using it to further improve performance to prevent these issues from recurring. If you have any questions, please contact [email protected] and we'll be happy to answer them. David Director of Engineering

2. monitoring May 20, 2026, 08:26 AM UTC

   Application performance is returning to normal. We are looking into the root cause of this issue, and will update with a postmortem once we know more.

3. investigating May 20, 2026, 08:26 AM UTC

   We are currently experiencing an elevated number of request timeouts. We are investigating the issue and will update the status as soon as we have more information.