I've got deadline exceeded errors cased when setting memcache.   I've 
stared the issue and attached my graph.  I'm on an HDR app, not M/S

On Tuesday, May 22, 2012 9:57:23 PM UTC-7, Rishi Arora wrote:
>
> If you guys think this may be because of high memcache API latency, please 
> star this issue:
> http://code.google.com/p/googleappengine/issues/detail?id=7554
>
> Although it appears Google is being stubborn about it and has marked this 
> issue "WontFix", blaming it all on Master/Slave datastore.  An issue that 
> lasts this long can't be simply blamed on M/S.
>
> On Tue, May 22, 2012 at 11:41 PM, Daniel <[email protected]> wrote:
>
>> BTW, here's my milliseconds per request chart.  The problems started 
>> about 3 days ago and the data is dramatic.. here's a 7 day graph
>>
>>
>> <https://lh3.googleusercontent.com/-AuVmd45bJwY/T7xqVl-khiI/AAAAAAAAAzc/yOxaoYxWP9s/s1600/skywords.jpg>
>>
>>
>> On Tuesday, May 22, 2012 9:27:05 PM UTC-7, Daniel wrote:
>>>
>>> For the last 2-3 days I'm seeing the same thing.  I've gotten several 
>>> mails from unhappy users that my game server is too slow.. I don't normally 
>>> have complaints like this.   I'm also seeing Deadline Exceeded errors which 
>>> I don't normally see.   In addition my server costs are over double as 
>>> well, probably for the same reason. 
>>>
>>> You're not alone.. do you have any leads from Google on this?
>>>
>>> Daniel
>>>
>>> On Tuesday, May 22, 2012 8:25:44 PM UTC-7, Eric Ka Ka Ng wrote:
>>>>
>>>> from around last 12 hours, we see an exceptionally high latency on some 
>>>> of our gae apps. before requests on average takes 200ms-1000ms. now it can 
>>>> takes 2000ms - 5000ms. I try to measure the time as local client. like 
>>>> make 
>>>> 10 requests, 1-2 times the latency is normal <1s, half of the time is like 
>>>> 2s-5s, sometimes it even takes 10-20s!! 
>>>>
>>>> as a result, it also doubles the number of instances started by GAE, 
>>>> and so as the frontend cost. 
>>>>
>>>> does anyone observe similar problem recently, or it's only me? is it 
>>>> google's infrastructure prb, or anything i can do? if it's google's 
>>>> infrastructure prb, i dont think it makes sense to transform such bad user 
>>>> experiences (high latency) into doubling our hosting cost....
>>>>
>>>> - eric
>>>>
>>>  -- 
>> You received this message because you are subscribed to the Google Groups 
>> "Google App Engine" group.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msg/google-appengine/-/MJCF02agu5oJ.
>>
>> To post to this group, send email to [email protected].
>> To unsubscribe from this group, send email to 
>> [email protected].
>> For more options, visit this group at 
>> http://groups.google.com/group/google-appengine?hl=en.
>>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google App Engine" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/google-appengine/-/aV5lK9vm15YJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-appengine?hl=en.

Reply via email to