I’ve set my app’s concurrency hard limit to 2 and a minimum instance count of 3 however when I load test with a concurrency of 12, I expected to see another vm allocated however this does not happen.
Perhaps I’m missing something?
I’ve set my app’s concurrency hard limit to 2 and a minimum instance count of 3 however when I load test with a concurrency of 12, I expected to see another vm allocated however this does not happen.
Perhaps I’m missing something?
Strange. I guess the first thing to check would be your autoscale settings: fly autoscale show
I believe by default autoscaling is disabled. So you will want to check it is enabled and also your max value is above 3 (since naturally if your min is 3 but also your max is 3, a new vm won’t be started):
Assuming it’s neither of those, you would also need to consider that auto-scaling won’t be instant. But assuming the concurrent connections persist, and auto-scaling is enabled, and there is capacity etc, the vm should then be started. I’m not sure what the time threshold is as I don’t believe that can currently be set.
Hi @artyroip,
Just letting you know we just recently tracked down an internal bug in the autoscaling service that was causing scaling events to not get triggered properly, sorry for the inconvenience. It should be working correctly now so please give it another try!
Ooh. Cool. I’ll give it a try!
Can confirm, the autoscaling is much better. I almost left Fly for another provider but I’m glad I stayed as the autoscaling is now much more reactive and results in pretty much no errors.