Hey Vinny P, thanks for the answer.
I'm investigating it since a couple of days, but without any results so far - so I need your help and ideas, since the billing support is ignoring me, probably because of the technical nature of the problem. Facts: Date / DAU / Bandwidht 15.11 / 385/ 6.5 GB 16.11 / 585 / 9 GB 17.11 / 660 / 10 GB 18.11 / 451 / 12 GB 19.11 / 455 / 46 GB 20.11 / 438 / 42 GB 21.11 / 429 / 44 GB There is a huge increase of outgoing bandwidth, but when we examine the charts from the dashboard, it's not clearly evident, why this happens (charts attached). On the 19th, we didn't deploy a new version or changed the configuration of the application. We examined also the bandwidth correlated components (blob, mail, channel api) but nothing has changed on that day. As next, I downloaded the logs from all the days and summed up all response sizes, I got following result: 18.11: 3.9 GB 19.11: 4.2 GB 20.11: 3.8 GB 21.11: 4.1 GB So there is a huge discrepancy in the bandwidth. I deployed then the application on other app-id and did some unit testing: Clientside: Firebug measured ~20MB download (some images and JS) Serverside: in the logs the response size of every GET of a ressource is 0 with status 200 (...3.cache.js HTTP/1.1" 200 0 ...) and the total size of one game session according to logs is 715kB. App Engine Dashboard: Outgoing bandwidth 0,11GB! AppStats: none urlFetch, a couple of channel send message (we have a chat in our app, which I tested). Tried it with 3 browsers and accumulated 0.33GB outgoing bandwidth, although the logs say 2,5MB and according to the summed result of the clients about 65MB (what I expect). Caching works, because joining the second time, I only download 30kB and also the bandwidth counter doesn't go up in this situation. Any other help and ideas hreatly appreciated! Robert Am Dienstag, 26. November 2013 00:13:52 UTC+1 schrieb Vinny P: > > On Tue, Nov 19, 2013 at 2:10 PM, robs <[email protected] <javascript:> > > wrote: > >> The usage statistics, load, instances and the traffic graphs doesn't show >> up any anomalies compared to last days (they are even smaller), but the >> bandwidth counter just shots through the roof, showing by now (13h to next >> reset) outgoing bandwidth three times as high as in the last whole days! >> >> Do you maybe have an idea how to determine the cause? >> > > > Outbound bandwidth is consumed by either (1) the application serving data > in response to incoming HTTP requests or (2) the application initiating an > outbound request (urlfetch, sockets, etc). In the first instance, you > should see incoming requests in the Logs section of the App Engine console. > Make sure you look at every log and every part of the logging page; you > have to use the dropdown box to view the logs of different modules and > backends. In the second case (outbound requests such as urlfetch) you > should audit your code and attached libraries to figure out where outbound > requests are originating from. If you can't find the source, use AppStats ( > https://developers.google.com/appengine/docs/java/tools/appstats ) - > it'll show you when and where outbound requests are coming from. > > > ----------------- > -Vinny P > Technology & Media Advisor > Chicago, IL > > App Engine Code Samples: http://www.learntogoogleit.com > > -- You received this message because you are subscribed to the Google Groups "Google App Engine" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/google-appengine. For more options, visit https://groups.google.com/groups/opt_out.
