Date: February 6, 2018
Time: 10:57 am - 11:52 am PST
Wait times for Windows and Linux VMs were high, which led to some tests timing out.
Why did it happen:
A transient network issue led to a temporary drop in capacity, causing wait times to increase.
What did we do to fix it:
The problem recovered on its own.
What we are doing to prevent this from happening again:
We’re reviewing our network configurations and adding more network logging to facilitate further investigation. Additionally, we’re adding logging and instrumentation to the service responsible for managing the cloud to more quickly predict network issues and react more effectively when they occur.