Open omri-shilton opened 4 months ago
I have no idea about that and didn't explore that option and the possibilities. I suspect we have only the basic version of it in the chart here.
I would suggest exploring that option x and documenting it in the wiki for the others, plus the updated charts. Check also the past PRs regarding HPA, maybe you'll find some hints of its application.
I've been playing with queue mode and autoscaling, but it seems that the chart only autoscales (with hpa) the main pod, and not the worker pods. even though in the worker deployment there is an if-else that checks with autoscaling is enabled. Am I missing something? What is the correct approach to scaling worker pods based on executions/resources/other metrics?
Thanks!