Suppose GAE only supported a maximum of two versions per instance at 128MB max memory per instance on the default F1 config. Suppose Version 1 of your app uses exactly 64MB memory at startup. In that case, you can't start Version 2 of you app if it uses so much as one byte more memory than version 1, since 64MB + 64MB + 1 byte > 128MB. (Yes, it's a soft limit, but here's my point-->). What's the *purpose* of versions then? It isn't testing two versions side by side on the same instance in this case because one can't.
I actually have this problem now. I have an app that starts up at 50MB and have added a feature that increases it to 100MB (some large in-memory hash lookups were added). However, I can't run them side by side on one F1 instance as I exceed my 128MB capacity, when *both individually* are under 128MB. I should be able to do this. What exactly is the point of GAE versions otherwise? I should be able to stand up both and route traffic to both via the provided features to do so without production failing due to this problem. Also posted to Stackoverflow: http://stackoverflow.com/questions/37402912/google-app-engine-memory-quotas-should-be-per-version-per-instance-not-just-per -- 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 https://groups.google.com/group/google-appengine. To view this discussion on the web visit https://groups.google.com/d/msgid/google-appengine/6080da18-48be-4a9c-8afe-aee10e22de53%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
