Date: October 16, 2018
Time: 7:28 am - 8:45 am PDT
*There was an increased failure rate for live and automated tests running on our U.S Real Device Cloud (RDC).
*Why it happened:
*A surge in long-running tests caused excessive memory usage and degraded the quality of service for RDC in the U.S.
*How we fixed it:
*We allocated additional resources for the U.S. RDC Service to ensure that the quality of service returned to normal levels.
*What we are doing to prevent it from happening again:
*We are further optimizing our services so that they make more efficient use of memory.