Partial probe network outage

Update timeline

1. identified Aug 17, 2026, 07:00 AM UTC

   We are investigating a disruption that affects a subset of our global checking infrastructure. Checks that run from these locations produce no results and send no alerts for the duration of this incident. For the duration of this incident, checks from these locations produced no results and sent no alerts. Treat this period as a missing data interval in your check history. Checks from all other locations, and all other platform functions, operate normally. We will give an update in an hour

2. monitoring Aug 17, 2026, 08:29 AM UTC

   We identified the cause. A fault in our scheduling layer dispatched check runs too late, so the runs expired before the affected probe locations could start them. This is why the locations produced no results and sent no alerts. We are applying a correction to the scheduling layer and we are clearing the affected runs. Checks from all other locations continue to operate normally. We will give an update in 30 minutes.

3. resolved Aug 17, 2026, 09:17 AM UTC

   We have doubled capacity of our scheduling layer to keep up with an increasing load of active check runs. The system is back to normal