This is very helpful. I’m trying to build a “Job Scheduling” system that will enable and disable tasks when they are ready to run based on another system’s scheduling.
I have it “retry” each task 5 times, so that if the pipeline fails it will auto retry… but if it succeeds, then it will disable its own task and shouldn’t run again. Unfortunately even after the task has been disabled, the pipeline will run a second time – so it appears that scheduling is added to a queue, and if it’s within 1-2 minutes of running… disabling a task doesn’t remove it from the queue.
Does anyone have any ideas? I thought about using the Task Execute, but that’s being deprecated (documentation says so). I don’t want to use the pipeline execute because I don’t want them all related, because I’ll have 1 master scheduler pipeline related to every single execution.
EDIT: Maybe it’s that I’ve got the retry interval set to 3 minutes? I’m dropping it to 5 and will see if that fixes it.