2 | 
3 | 
4 | 
5 | 
6 | 
7 | 
8 | 
9 | 
10 | 
11 | 
12 | 
13 | 
14 |
15 |
16 |
17 |
18 | # Kube Sentry
19 |
20 | Kube Sentry an innovative open-source tool reshaping Kubernetes cluster management. Designed for developers, it offers advanced features for early detection of potential issues, aiming to notably reduce bug occurrences. Seamlessly integrated with Prometheus, it enhances tracking and alerting capabilities, providing a holistic, user-friendly approach to Kubernetes management.
21 |
22 |
23 | ## Quickstart
24 | Clone this repository, install dependencies and run:
25 |
26 | ```
27 | git clone https://github.com/oslabs-beta/KubeSentry.git
28 | cd KubeSentry
29 | npm install
30 | npm run
31 | ```
32 |
33 |
34 | The Sentry Dashboard will be available at [localhost:3000](http://localhost:3000/dashboard).
35 |
36 | # Setting up your Kubernetes cluster
37 |
38 | - Install kubectl. Run `kubectl version` to check if it has been sucessfully downloaded.
39 |
40 | - KubeSentry will connect to the Kubernetes cluster configured by `kubectl config current-context`. For testing purposes, you can run a local Kubernetes cluster either with [Docker Desktop](https://www.docker.com/products/docker-desktop/) or [minikube](https://minikube.sigs.k8s.io/docs/start/).
41 |
42 | * The following commands will follow managing your K8s cluster on Docker Desktop.
43 | - To start Kubernetes on docker-desktop, navigate to Settings > Kubernetes and **Enable Kubernetes**.
44 | - run `kubectl cluster-config` to ensure your Kubernetes cluster is operating.
45 |
46 | # Set up metrics server
47 |
48 | - [Kubernetes Metrics Server](https://github.com/kubernetes-sigs/metrics-server) is required to get additional information from your K8s cluster.
49 | - Follow instructions on the Github repo to install the Kubernetes metrics-server.
50 | - ** Note the `--kubelet-insecure-tls` flag to bypass the CA authenication for kubelets. **
51 | ```yaml
52 | spec:
53 | containers:
54 | - args:
55 | - --cert-dir=/etc/metrics-server-auth
56 | - --secure-port=4443
57 | - --kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname
58 | - --kubelet-use-node-status-port
59 | - --metric-resolution=15s
60 | - --kubelet-insecure-tls
61 | image: registry.k8s.io/metrics-server/metrics-server:v0.6.4
62 | ```
63 | - If your metrics server is installed correctly, you will be able to run `kubectl top pods` and `kubectl top nodes` to get pod and node metrics of your cluster.
64 |
65 | # Configure Prometheus to collect cluster metrics
66 |
67 | If you have not already, install [Helm](https://helm.sh/). Helm is a repository of published yaml files that are used to configure deployments used in Kubernetes.
68 |
69 | Add the Prometheus Helm chart to your Helm repo
70 | `helm repo add prometheus-community https://prometheus-community.github.io/helm-charts`
71 |
72 | - you can check if the Helm chart has been sucessfully added by running `helm list` in you CLI
73 |
74 | Install Prometheus to your Kubernetes cluster.
75 | `helm install prometheus prometheus-community/prometheus`
76 |
77 | - If you run `kubectl get all` you should see the deployments, configmaps, services Prometheus has installed.
78 | - If you are running Kubernetes on docker-desktop, you will not be able to current access the Prometheus server API that is hosted on port 9090. This is because we are running Kubernetes in a containerized environment. To work around this, we will open up a **nodeport** service that will act as a middle man to set up a connection with the prometheus server with the outside world
79 | - `kubectl apply -f https://github.com/oslabs-beta/KubeSentry/Yamlfiles/prometheus-nodeport.yaml`
80 | - Now you should be able to access the Prometheus server API via `localhost:31302`
81 |
82 | # instrument your application
83 |
84 | To instrument your appplication for Prometheus to scrape, you need to require [Prom-client](https://github.com/siimon/prom-client) onto your application. Prometheus' default scraping endpoint is `/metrics`
85 |
86 | ```js
87 | const client = require('prom-client');
88 |
89 | //create a custom counter
90 | const customCounter = new client.Counter({
91 | name: 'my_custom_counter',
92 | help: 'Description of my custom counter',
93 | });
94 |
95 | //increment counter
96 | customCounter.inc();
97 |
98 | //endpoint to provide metrics for prometheus to scrape
99 | app.get('/metrics', async (req, res) => {
100 | const data = await client.register.metrics();
101 | res.header('Content-Type', 'text/plain').status(200).send(data);
102 | });
103 | ```
104 |
105 | As an alternative, you can install [express-prom-bundle](https://www.npmjs.com/package/express-prom-bundle) which grants you the **metricsMiddleware** to expose all your metrics to the /metrics endoint with one line.
106 |
107 | ```js
108 | const promBundle = require('express-prom-bundle');
109 | const client = require('prom-client');
110 | const metricsMiddleware = promBundle({ includePath: true });
111 |
112 | app.use(metricsMiddleware);
113 | ```
114 |
115 | # Create an image of your application
116 |
117 | Build an image of your application to be deployed on Kubernetes.
118 |
119 | # Deploy your application on Kubernetes
120 |
121 | Create a deployment yaml file to deploy your application. This will handle any horizontal scaling required for your app.
122 |
123 | > Example of a deployment configuration yaml file
124 |
125 | ```yaml
126 | apiVersion: apps/v1
127 | kind: Deployment
128 | metadata:
129 | name: sentry-tests
130 | namespace: sentry-tests #optional. Will create in current namespace if left out
131 | spec:
132 | selector:
133 | matchLabels:
134 | app: sentry
135 | replicas: 4
136 | template:
137 | metadata:
138 | labels:
139 | app: sentry
140 | spec:
141 | containers:
142 | - name: sentry-test
143 | image: sentrytest4:1.0 #application image
144 | ports:
145 | - containerPort: 3000
146 | ```
147 |
148 | Run `kubectl apply -f path/to/deployment.yaml` to deploy your application.
149 |
150 | Create a service yaml configuration to handle load balancing for your applications:
151 |
152 | ```yaml
153 | apiVersion: v1
154 | kind: Service
155 | metadata:
156 | name: sentrytest-nodeport
157 | spec:
158 | selector:
159 | app: sentry #must match the label of the pod
160 | ports:
161 | - name: web
162 | port: 3000 #port to access the service node
163 | targetPort: 3000 #must match the container port of the application
164 | nodePort: 30011 #optional nodePort for open access if running in a docker conatiner
165 | type: NodePort
166 | ```
167 |
168 | Run `kubectl apply -f path/to/service.yaml` to deploy your application.
169 |
170 | Now you should be able to access your application at `localhost:30011`
171 |
172 | # Configure Prometheus to scrape metrics
173 |
174 | Add scrape jobs for your application and the metrics server. Learn more at [Prometheus](https://prometheus.io/docs/prometheus/latest/getting_started/)
175 |
176 | ```yaml
177 | global:
178 | evaluation_interval: 15s
179 | scrape_interval: 10s
180 | scrape_timeout: 10s
181 | rule_files:
182 | - /etc/config/recording_rules.yml
183 | - /etc/config/alerting_rules.yml
184 | - /etc/config/rules
185 | - /etc/config/alerts
186 | - /etc/prometheus/rules/*.yaml #will be used later for
187 | scrape_configs:
188 | - job_name: 'kubernetes-metrics'
189 | static_configs:
190 | - targets: ['metrics-server.kube-system.svc.cluster.local:443']
191 | - job_name: 'sentry-test'
192 | static_configs:
193 | - targets: ['sentrytest-nodeport.default.svc.cluster.local:3000']
194 | - job_name: prometheus
195 | static_configs:
196 | - targets: [localhost:9090]
197 | ```
198 |
199 | # The Kube Sentry
200 |
201 | ```
202 | git clone
203 | npm install
204 | npm run
205 | ```
206 | http://localhost:3000/dashboard should display your Kubernetes cluster.
207 |
208 | # Set up the Alert Manager
209 |
210 | Coming soon!
211 |
212 | # Learn More
213 |
214 | Head over to kubesentry.app!
215 |
216 |
--------------------------------------------------------------------------------
/__tests__/react.test.tsx:
--------------------------------------------------------------------------------
1 | /**
2 | * @jest-environment jsdom
3 | */
4 |
5 |
6 | import React from 'react'
7 | import { PodCard, PodCardProps } from "../src/app/ui/PodCard";
8 | import '@testing-library/jest-dom';
9 | import { render, screen, waitFor, fireEvent } from '@testing-library/react';
10 |
11 | describe('Unit testing React components', () => {
12 | describe('PodCard', () => {
13 | const mockOnClick = jest.fn();
14 | const mockPodData: PodCardProps = {
15 | podName: 'coredns-5dd5756b68-425cd',
16 | podStatus: 'Running',
17 | nameSpace: 'kube-system',
18 | handleClick: mockOnClick,
19 | }
20 |
21 | test('PodCard displays Name, Status, and Namespace', () => {
22 | render( 40 | TEAM Eat Sleep Kubernetes 41 | 44 | Join us and be part of our crew 45 |
46 | 50 | I wanna be join "KUBE SENTRY" team 51 | 52 |You clicked {count} times
9 | 10 |Could not find requested resource
11 | Return Home 12 |
7 | {'('}
8 | {alert.startsAt}
9 | {') '}
10 | Name:
12 | Status:
13 | {/* conditional coloring based on status */}
14 |
19 | {' '}
20 | {alert.status.toUpperCase()}{' '}
21 |
22 | Instance: {alert.labels.instance}
23 |
25 | Summary:
{namespace}
68 |69 | {status} 70 |
71 |PodName: {name}
85 |Home
45 | 46 |Pods
66 | 67 |Status
87 | 88 |Resource
110 | 111 |Report
167 | 168 |
176 | Graph View
177 | 178 |Alert
216 | 217 |