Q: 4
How does the Oracle Cloud Infrastructure Container Engine for Kubernetes (OKE) Cluster Autoscaler
determine when to create new nodes for an OKE cluster?
Options
Discussion
I remember seeing similar questions in the official guide and practice exams, and I went with A. My reasoning was that autoscalers often trigger on resource utilization, like CPU or memory going above a set level. Not totally sure though, official docs can be a bit tricky on how node vs pod scaling is triggered. Anyone else use labs for this?
A
B
B tbh
B . A trips people up, but autoscaler looks at pending pods, not just CPU spikes.
Why not A? I thought most cloud autoscalers trigger when CPU or memory goes above a certain threshold.
A for me, since cluster autoscalers often kick in when CPU or memory usage goes above a set amount. That's the usual trigger in most cloud platforms I think, not 100% on OKE specifics though. Someone else try this in a lab?
B
Not A, it's B. Cluster Autoscaler doesn't react to CPU/memory utilization spikes, it actually looks for unschedulable pods because resource requests can't be met. I've seen similar questions trip folks up since A sounds right at first glance, but that's more for pod autoscaling. Correct me if I'm missing something here.
A is wrong, B. Cluster autoscaler checks if pod resource requests can't be scheduled, not just CPU or app traffic.
Probably B, that's how autoscalers in Kubernetes work. Official docs and hands-on labs help clarify this if you want extra confirmation.
Be respectful. No spam.
