knative / serving

Kubernetes-based, scale-to-zero, request-driven compute
https://knative.dev/docs/serving/
Apache License 2.0
5.55k stars 1.16k forks source link

is there a way to have TTL for each replica ? #15538

Open vasudha-devarakonda opened 3 weeks ago

vasudha-devarakonda commented 3 weeks ago

I have 7 replica of each revision and would want each replica to be up for 10 minutes if they are inactive and then terminate. to 0 (not just the last replica, all the replicas )

skonto commented 2 weeks ago

Hi @vasudha-devarakonda would https://knative.dev/docs/serving/autoscaling/scale-bounds/#scale-down-delay work for your case?