Increased Datastore latencies causing timeout errors for 30 minutes, September 14, 2010

[App Engine Downtime Notify]

At 4:30am PDT, App Engine's datastore experienced a sharp increase in
request latency which resulted in increased error rates as a result of
request timeouts. We have determined this issue was caused by
contention for underlying Google infrastructure and was unrelated to
yesterday's scheduled maintenance period.

By 4:51am, the oncall team had diagnosed the issue and took corrective
action. The Datastore worked through the backlog of requests by 5:00am
and latencies returned to expected values. We sincerely apologize for
the impact this outage had on your applications. We continue to make
improved reliability our number one goal.