Cluster monitoring stack for clusters based on Prometheus Operator
Go to file
2018-03-13 18:47:34 -05:00
grafana-dashboards Updated dashboard 2018-03-13 18:47:34 -05:00
manifests Replaced temporarily the configmap-reload image to ARM version due to a bug in ARM64 image. https://github.com/jimmidyson/configmap-reload/issues/17 2018-03-13 10:45:09 -05:00
deploy Use internal service name for SMTP 2018-03-03 08:52:00 -05:00
Readme.md Initial import 2018-03-01 19:03:53 -05:00
teardown Initial import 2018-03-01 19:03:53 -05:00

Prometheus Operator for ARM platform

The Prometheus Operator for Kubernetes provides easy monitoring definitions for Kubernetes services and deployment and management of Prometheus instances.

This project aims on porting the official manifests and images to the ARM platform. This have been tested on a ARM64 Kubernetes cluster deployed as this article.

Changes to Kubeadm for Prometheus Operator

According to the official deployment documentation here, a couple of changes on the cluster are required:

We need to expose the cadvisor that is installed and managed by the kubelet daemon and allow webhook token authentication. To do so, we do the following on all the masters and nodes:

sed -e "/cadvisor-port=0/d" -i /etc/systemd/system/kubelet.service.d/10-kubeadm.conf
sed -e "s/--authorization-mode=Webhook/--authentication-token-webhook=true --authorization-mode=Webhook/"
systemctl daemon-reload
systemctl restart kubelet

In case you already have a Kubernetes deployed with kubeadm, change the address kube-controller-manager and kube-scheduler listens on master node in addition to previous kubelet change:

sed -e "s/- --address=127.0.0.1/- --address=0.0.0.0/" -i /etc/kubernetes/manifests/kube-controller-manager.yaml
sed -e "s/- --address=127.0.0.1/- --address=0.0.0.0/" -i /etc/kubernetes/manifests/kube-scheduler.yaml