Probe network · quality

repose.sh →

Every number on this page is computed live from the same tables our alert engine writes. If our probes are unhealthy, you get to know before we do damage control.

Engine decisions · last 7 days
0
pages sent — consensus confirmed
10
single-region flaps suppressed
0
held — upstream provider incident
Region
Checks 24h
Success
Avg latency
ap-southeast
4,319
99.95%
444ms
eu-west
4,559
100.00%
353ms
us-east
4,435
100.00%
260ms

Beta honesty note: three probe machines, one per continent — Amsterdam, Virginia and Singapore. That is fewer than anyone else advertises, and it means the default threshold is unanimity: all three have to agree before anyone is paged. A published false-positive rate follows once there's enough incident history to audit.

We don't cry wolf — and you shouldn't have to take our word for it.