Correct 🙂
polling_interval_time_min = the scaler interval for checking tasks in the queue

Why would it solve the issue? max_spin_up_time_min should be the param defining how long to wait after starting an instance, not polling_interval_time_min , right?

HI JitteryCoyote63 ,

can you try increasing the polling_interval_time_min to 5-7 minutes for the check? do you get double machines with it too?

Here is what happens with polling_interval_time_min=1 when I add one task to the queue. The instance takes ~5 mins to start and connect. During this timeframe, the autoscaler starts to new instances, then spin them down. So it acts as if max_spin_up_time_min=10 is not taken into account

Ok, I am asking because I often see the autoscaler starting more instances than the number of experiments in the queues, so I guess I just need to increase the max_spin_up_time_min

I want to verify it doesn’t start more than one instance for the same task

👍 checking it

Yes, it did spin two instances for the same task

I will try with that and keep you updated

whats the clearml version?

1.1.2