Today, all of the sudden, we've been getting excessive amounts of DeadLineExceeded errors during URL fetch operations. We leave the default 5s deadline for a url fetch call, but through local testing and other benchmarks, we don't think our url fetch() calls should be taking longer than 1s.
Is app engine experiencing any latency issues? We had problems earlier this week and I saw an "elevated" status on the system dashboard for other services but not for url fetch, although there is clearly a large spike:
We started getting these latency hits around 9am EST.