Scale node pool to zero throws HTTP 500 about doks HOT 2 CLOSED

Hey there @tedmiston. It looks like you have only one node pool running with that cluster. For now, we require at least one node to exist for a cluster or otherwise return an error. If you were to add another node pool and try that operation scaling down to a zero count for that first pool again, things should work.

Fully agree that the HTTP 500 response is not very helpful. Thanks for pointing out, we'll get that addressed shortly and return a meaningful error message. Keeping the ticket open at least until an improvement has shipped.

from doks.

We should ship a meaningful error code and message by now. Also, our documentation should describe the constraint of having at least one non-empty node pool.

I'll be closing the issue, but don't hesitate to post again / file a new bug report if you feel like the way we handle this could still be improved.

from doks.