Prometheus Adapter
The Prometheus adapter provides an implementation of Kubernetes resource metrics, custom metrics, and external metrics APIs.
This adapter is therefore suitable for use with the autoscaling/v2 Horizontal Pod Autoscaler in Kubernetes 1.6+. It can also replace the metrics server on clusters that already run Prometheus and collect the appropriate metrics.
Source: kubernetes-sigs/prometheus-adapter
Requirements
- Kubernetes >=
1.21.0
- Kustomize
= v3.5.3
- prometheus-operator
- prometheus-operated
Image repository and tag
- Prometheus adapter image:
registry.sighup.io/fury/prometheus-adapter/prometheus-adapter:v0.9.1
- Prometheus adapter repository: Prometheus adapter on GitHub
Configuration
Fury distribution Prometheus adapter is deployed with the following configuration:
- Resource limits are
250m
for CPU and 180Mi for memory - Listens on port 6443
- Metrics are scraped from Prometheus every
1m
Deployment
You can deploy prometheus-adapter by running the following command:
kustomize build katalog/prometheus-adapter | kubectl apply -f -