# Monitoring of infrastructure

> [HTML Version](monitoring_infrastructure.html)

Prometheus is a monitoring tool designed to collect and analyze system operation data. Monitoring facilitates maintaining the stability and reliability of the infrastructure.

Beforehand, install the [monitoring tools](install-monitoring-tools.md). Then configure the internal components of Kubernetes to transmit metrics to Prometheus.

The setup consists of four steps:

1. [Configure the kube-proxy component](#configure_kube_proxy_component).

2. [Configure the kube-scheduler component](#configure_kube-scheduler_component).

3. [Configure the etcd component](#configure_etcd_component).

4. [Configure the kube-controller-manager component](#configure_kube_controller_manager_component).

## Step 1: Configure the kube-proxy component

1. Make changes to the ConfigMap;

````
kubectl edit cm/kube-proxy -n kube-system

2. ````
Set the value \[OBJECT\] in hte line \[OBJECT\];

````
kind: KubeProxyConfiguration  
metricsBindAddress: "0.0.0.0:10249"

3. ````
Apply the changes.

````
kubectl -n kube-system rollout restart daemonset kube-proxy

## ````
Step 2: Configure the kube-scheduler component

1. Make changes to the **kube-scheduler.yaml** manifest file;

````
nano /etc/kubernetes/manifests/kube-scheduler.yaml

2. ````
Set the value \[OBJECT\] in the line \[OBJECT\]. The configuration will be applied automatically after saving the manifest.

````
\- --bind-address=0.0.0.0

## ````
Step 3: Configure the etcd component

1. Make changes to the **etcd.yaml **manifest file;

````
nano /etc/kubernetes/manifests/etcd.yaml

2. ````
Set the value \[OBJECT\] in the line \[OBJECT\]. The configuration will be applied automatically after saving the manifest.

````
\- --listen-metrics-urls=http://0.0.0.0:2381

## ````
Step 4: Configure the kube-controller-manager component

1. Make changes to the **kube-controller-manager.yaml **manifest file:

````
nano /etc/kubernetes/manifests/kube-controller-manager.yaml

2. ````
Set the value \[OBJECT\] in the line \[OBJECT\]. The configuration will be applied automatically after saving the manifest.

````
\- --bind-address=0.0.0.0

````
Once the component configurations are set, their metrics will be available in Prometheus.

## Data vizualisation

In the data visualization software Grafana, you can view charts with monitoring results.

Chart example 1

**[![monitoring\_infrastructure\_1](monitoring_infrastructure_1.png)]**

Chart example 2

**[![monitoring\_infrastructure\_2](monitoring_infrastructure_2.png)]**

Chart example 3

**[![monitoring\_infrastructure\_3](monitoring_infrastructure_3.png)]**

Chart example 4

**[![monitoring\_infrastructure\_4](monitoring_infrastructure_4.png)]**

Custom dashboards are also available. To activate them, enable **dashboards.enabled** in the **values**  file for the installation of monitoring tools.

````
\# Enable chart import  
  grafana-custom-dashboards:  
    dashboards:  
      enabled: true

````
Dashboard example 1

**[![monitoring\_infrastructure\_5](monitoring_infrastructure_5.png)]**

Dashboard example 2

**[![monitoring\_infrastructure\_6](monitoring_infrastructure_6.png)]**

Dashboard example 3

**[![monitoring\_infrastructure\_7](monitoring_infrastructure_7.png)]**

Dashboard example 4

**[![monitoring\_infrastructure\_8](monitoring_infrastructure_8.png)]**