This is mind boggling. There is a simple, brain-dead fix that does NOT require the (presumably complex) scheduling logic to change one bit. Simply NEVER expose a user-facing request to a loading request, if the developer configures it that way. Let the scheduler spin up and down instances however it sees fit, but don't let a new instance serve requests until it successfully completes initialization.
This was beat to death months ago in a series of threads, but obviously resulted in no action. On Tuesday, December 18, 2012 12:03:19 PM UTC-5, Cesium wrote: > > I changed my settings to have 3 resident instances. > > I still see long latency when the scheduler creates new instances, rather > than using the resident instances. > > The GAE scheduler behavior at low request rates is odd, and > counterintuitive. > > But we knew that didn't we? > David > > -- 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/-/mibjVETUupIJ. 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.
