I am having a problem where task queues stop for 30 minutes - 2 hours, and then restart (using a large number of instances to pick up the slack). This is causing us great distress, since we were hoping we could rely on task queues for timely processing of some workload.
Does anybody know what can cause this stalling?
Is anybody else seeing it?
Luca