If you locate and inspect the deferred.py of the sdk files, it's actually a pretty simple and brilliant routine
When a task overflows the taskqueue size limit, it puts a _DeferredTaskEntity to the db to utilize it for storage *My Problem:* I see 30-40 failed defer tasks at production (out of a burst operation that spanned 100.000's of tasks) They fail with: "X-Appengine-Taskretrycount:1, X-Appengine-Taskretryreason:Instance Unavailable" The exception is a PermanentTaskFailure from run_from_datastore Does this mean that the tasks actually executed before, but faultily re-triggered, and threw PermanentTaskFailure's because the previous execution deleted the _DeferredTaskEntity entity? (I'm trying to pinpoint the cause and the trigger so it doesn't happen again, since these exceptions all happened once and have a low count, I'm guessing this is a very edge behaviour) If this is the case, they would probably cause a silent re-execution of the task, if the db entity / overflow scenario wasn't used -- 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. To view this discussion on the web visit https://groups.google.com/d/msgid/google-appengine/4b0ba6ff-0fec-4f94-b033-35f58d3dd85c%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
