latency started spiking up dramatically today

49 views
Skip to first unread message

Wenzhe Gao

unread,
Sep 29, 2016, 8:27:28 PM9/29/16
to Google App Engine
Hi,

Our app has been running fine with low latency for months now. Latency started spiking up dramatically at around 6:50am this morning. No real changes on our side. I see quite a few datastore RPC deadline exceeded errors in the logs. Ex:

...
rpc = datastore_rpc.MultiRpc.wait_any(self.rpcs) DeadlineExceededError: The overall deadline for responding to the HTTP request was exceeded.

See latency graph attached for reference. Any ideas what's going on?
Screen Shot 2016-09-29 at 4.04.38 PM.png

Nick (Cloud Platform Support)

unread,
Oct 3, 2016, 10:37:33 AM10/3/16
to Google App Engine
Hey Wenzhe, 

We have no incidents reported on the Cloud Status Dashboard for Datastore as of this moment. Sometimes performance may degrade for a short period of time dependent on Data-Center conditions. You should check out SLA on Datastore and consider implementing some exponential back-off logic in your code to handle the possibility of deadline exceeded errors. We have a document on this topic titled "Dealing with DeadlineExceededErrors".

I'll check to see if I can find any other explanation and will report back to this thread. 

Cheers,

Nick
Cloud Platform Community Support

Nick (Cloud Platform Support)

unread,
Oct 3, 2016, 10:38:47 AM10/3/16
to Google App Engine
Hey Wenzhe,

Apologies for not including this in my last message, as I had believed your original message included the necessary information. Can you email me your project ID directly at pay...@google.com? I'll be glad to take a look for you after that. Is this still occurring, by the way?

Regards,


Nick
Cloud Platform Community Support

Reply all
Reply to author
Forward
0 new messages