TestingBot had 5 outages in the last 2 years totaling 3h 13m of downtime — averaging 0.2 incidents per month.
There were 5 TestingBot outages since September 5, 2025 totaling 3h 13m of downtime. Each is summarised below — incident details, duration, and resolution information.
Our upstream datacenter is experiencing issues causing services on TestingBot to fail. We are following up.
This incident has been resolved.
Maestro tests against physical Android devices were returning errors.
A previous deploy introduced a bug that failed the start of Maestro tests when some conditions were met. We have added a fix, test and alert.
TestingBot Storage has been experiencing issues today on February 3, 2026 from 16:20 UTC up to 19:00 UTC. This has been resolved.
Today between 10:19 and 10:29 we had an incident with our database which caused downtime. This was resolved and we will be making changes to prevent this from happening.
Since 1:00 UTC time we have been experiencing elevated wait times, causing some tests to wait for up to 5 minutes before they could start. We deployed a fix and are monitoring the waiting times. They are back to < 10 seconds since 04:30 UTC.
Today we experienced long waiting times for Windows \+ macOS machines. The issue was caused by a server issue, where memory issues were causing restarts of our hub \(h[ub.testingbot.com\)](http://hub.testingbot.com) This in turn caused the machines running the tests to fail to communicate with the hub, causing tests from clients to wait up to 5 minutes before a test could start. If your test runner is configured with a http connect timeout < 5 minutes, this may have caused issues where tests would fail to start. The memory issue was resolved and we are looking into a solution to prevent this from happening again.