I take my words back, my particular case was my own fault. It was caused by denormalized entities that we did not clean up and they gradually became so big that consumed all the memory. Instances did not die as i would have expected, instead the latency dropped and gae spinned more instances somehow limiting even those that did not run out of memory yet. Its a shame, but gae fault was the first thing that came to my mind when the problem emerged, it was not like that a year ago :(
пятница, 28 июня 2013 г., 9:14:12 UTC+3 пользователь Aleksei Rovenski написал: > > We are having now 20-40 instances where normally there are 2-4 for this > load, so went up about 10 times. > Just look at this pathetic screenshot http://clip2net.com/s/5iO2VD > And look how nicely billing instances grow http://clip2net.com/s/5iO47p while > requests per second just a bit http://clip2net.com/s/5iO5j2 > What a brilliant business model.... Service breaks, customers pay. > Any clue somebody is fixing this right now? > > четверг, 27 июня 2013 г., 18:02:49 UTC+3 пользователь Kyle Baley написал: >> >> We've been having severe latency issues for about an hour now. Haven't >> made any changes to our system for some time. We have a couple of apps and >> it *appears* to be affecting only one at the moment. ( >> bookedin-prod.appspot.com) >> >> Is anyone else having trouble? We've got some pretty testy clients on our >> hands at the moment and I need to tell them something. >> >> >> <https://lh4.googleusercontent.com/-UG942Im8YgM/UcxUD0QgN8I/AAAAAAAAA7A/F5SU7LzABSU/s1600/Latency.png> >> >> -- 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.
