Learn how to – How To Manage Kubernetes Cluster with Kubevious.
At this point, we can all agree that containerization is everywhere. This is because many organizations have decided to modernize their existing applications for cloud use. Containerization can be defined as the packaging of software and all the required libraries in a lightweight executable referred to as a container. The most popular open-source tools used to run containers are docker, Podman, and Kubernetes.
Kubernetes/k8s is a widely used open-source tool, used in container orchestration. It works by distributing the workload across a server farm, automating the container networking needs, storage, and persistent volumes while maintaining the desired container state
Managing Kubernetes clusters can be hard, especially in a large environment. Therefore the need for Kubernetes management tools increases. The popularly used tools to manage Kubernetes clusters are K9s, Weave scope, Dashboard + Kubectl + Kubeadm, KubeSpray, Kontena Lens, Konstellate, WKSctl, Portainer, Rancher, Headlamp e.t.c
What is Kubevious?
Kubevious is an open-source graphical interface for Kubernetes. It makes the Kubernetes cluster, and application configuration state safe and easy to understand. Kubevious eliminates disasters and costly outages by validating the application manifests and configurations periodically. It normally detects and prevents misconfigurations, conflicts, inconsistencies, and violations of Kubernetes practices. It operates inside the clusterand can be accessed via the web application or CLI for CI/CD pipelines integration.
From the Kubevious app-centric user interface, you can get intuitive insight, introspection, and troubleshooting tools for cloud-native applications. The features offered by Kubevious are:
- Investigate: It lets users travel back in time, investigate cluster configuration and state, audit applications, root cause outages, and recover good and working manifests. The Time Machine executes resources from the Kubernetes API server making it easy to keep track of all the changes made.
- Search Engine: This is a full-text search engine that helps users find and return Kubernetes manifests matching various search criteria.
- Introspect: It provides insights optimized for specific roles and responsibilities as well as correlated app-centric views. These are:
- Correlated RBAC: helps understand which permissions are granted to Applications through ServiceAccounts, RoleBindings, and Roles.
- Correlated Network Policies: defines how NetworkPolicies are affecting applications and which traffic is allowed.
- Identifying Blast Radius: helps identify shared resources within the cluster.
- Radioactive Workloads: used to identify applications that have excessive permissions.
- Capacity Planning and Resource Usage Optimization: helps identify how much of cluster resources are used by Applications and Namespaces.
- Validate: It lets users validate cluster configuration and state for misconfigurations and violations of best practices using:
- Built-in Validations that include a comprehensive library of rules to detect and prevent DevOps/SRE-focused misconfigurations.
- Rules Engine which provides an intuitive policy language for custom extensions of validation logic using a JavaScript–like(if-then-else) syntax. The Rules Engine allows the enforcement of complex cross-manifest policies.
This guide demonstrates how to manage Kubernetes Cluster with Kubevious.
#1. Setup kubectl and helm
It is required that you have a Kubernetes cluster set up. The below guides can be used to create a Kubernetes cluster on your Linux system.
- Deploy HA Kubernetes Cluster on Rocky Linux 8 using RKE2
- Deploy Kubernetes on KVM using Flatcar Container Linux
- Run Kubernetes on Debian with Minikube
- Deploy Kubernetes Cluster on Linux With k0s
- Install Kubernetes Cluster on Ubuntu using K3s
- Install Kubernetes Cluster on Rocky Linux 8 with Kubeadm & CRI-O
- Deploy k0s Kubernetes on Rocky Linux 9 using k0sctl
- Install Minikube Rocky Linux 9 and Create Kubernetes Cluster
Once a Kubernetes cluster has been set up as desired, confirm that you can use kubectl command:
View the available nodes in the cluster.
$ kubectl get nodes
NAME STATUS ROLES AGE VERSION
master Ready control-plane 80m v1.24.4+k0s
worker1 Ready <none> 80m v1.24.4+k0s
worker2 Ready <none> 80m v1.24.4+k0sYou also need to install Helm on your system. The guide below can be used to achieve this:
Verify the installation.
$ helm version
version.BuildInfo{Version:"v3.10.2", GitCommit:"50f003e5ee8704ec937a756c646870227d7c8b58", GitTreeState:"clean", GoVersion:"go1.18.8"}#2. Create a Persistent Volume for Kubevious
Kubevious runs with a MySQL Pod that requires a persistent volume claim named data-kubevious-mysql-0
Begin by creating an isolated namespace for Kubevious:
kubectl create namespace kubeviousSet it as the default namespace:
kubectl config set-context --current --namespace kubeviousYou’ll need StorageClass configured in your cluster. See our guides below on how to setup persistent storage in your Kubernetes cluster.
- Configure NFS as Kubernetes Persistent Volume Storage
- How To Deploy Rook Ceph Storage on Kubernetes Cluster
Confirm the storage class available for use.
$ kubectl get sc
NAME PROVISIONER RECLAIMPOLICY VOLUMEBINDINGMODE ALLOWVOLUMEEXPANSION AGE
nfs-01 (default) nfs-provisioner-01 Delete Immediate true 10d
The storage class that we’ll use in this guide is named nfs-01 as seen in output above.
You can set the SC as the default storage class:
Storage_Class=nfs-01
kubectl patch storageclass $Storage_Class -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'#3. Deploy Kubevious on Kubernetes
With a Kubernetes cluster setup and Helm v3.2+ installed, you can easily deploy Kubevious in the cluster.
Add the Kubevious Helm repo:
helm repo add kubevious https://helm.kubevious.ioRun the command below to perform an update before using it.
$ helm repo update
Hang tight while we grab the latest from your chart repositories...
...Successfully got an update from the "kubevious" chart repository
...Successfully got an update from the "nfs-subdir-external-provisioner" chart repository
...Successfully got an update from the "stable" chart repository
Update Complete. ⎈Happy Helming!⎈
Install Kubevious on Kubernetes with the command:
helm install kubevious kubevious/kubeviousSample Output:
NAME: kubevious
LAST DEPLOYED: Tue Sep 13 10:02:38 2022
NAMESPACE: kubevious
STATUS: deployed
REVISION: 1
TEST SUITE: None
NOTES:
NKXW
WXx:,;oKW
Nk;.....,oKW
No'.......,oKW
Nkc'.......,o0W
WKxkNNkc'.......,oKW
Kc.'ckNNkc'.......,oKW
WWWNNNNWW Nk;...'ckNNkc'.......cK
WX0kdocc:::::clodk0NW NOc'......'ckNNkc'...'cON
N0dc;'...............:ONNOc'..........'ckNNkl:ckN
WKx:'.................'c0WKl'...........,,;oX WNW
WKd;.....................;d0X0d;........;d0KXW
Nk:.........................'ckXKx;....;xXW
Xd,............................':kXKo,:xXW
Xo'...............................'l0NKXNXW
Wd'..................................;kXOc:O
Nkooooooooooooooooooooooooooooooooooooxkdod0W
WNK0000000000000000000000000000000000KK0000KKK0000KKXW
Kl,''''''''''''''''''''''''''''''''''''''''''''''''';xW
NkdoooooooooooooooooooooooooooooooooooooooooooooooooxKW
WNKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKNW
Nd,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,dN
Nl....................''''....................oN
Nl................';ok000Oko;.................oN
Nl...............'oXWKkxxkKWXo'...............oN
Nl...............lXWO;....;kWXl...............oN
Nl..............'dWWd'.....lNWd'.............'oN
WKOkkkkkkkkkkkkkOXWWKOkkkkkKW XOOOOOOOOOOOOOOOKW
Kubevious v1.0.13 installed!
Access Kubevious by setting up port forwarding
$ kubectl port-forward $(kubectl get pods -n kubevious -l "app.kubernetes.io/component=kubevious-ui" -o jsonpath="{.items[0].metadata.name}") 8080:80 -n kubevious
and visiting http://127.0.0.1:8080Verify the deployment:
$ kubectl get deploy
NAME READY UP-TO-DATE AVAILABLE AGE
kubevious-backend 1/1 1 1 53s
kubevious-collector 1/1 1 1 53s
kubevious-guard 1/1 1 1 53s
kubevious-parser 1/1 1 1 53s
kubevious-ui 1/1 1 1 53s
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
kubevious-backend-59fd8bc777-hpf6k 1/1 Running 0 4m17s
kubevious-collector-7cb5b9c87d-5ztc2 1/1 Running 0 4m17s
kubevious-guard-669c466bfb-67ctc 1/1 Running 0 4m17s
kubevious-mysql-0 1/1 Running 0 4m16s
kubevious-parser-64c898fbbb-9tbwv 1/1 Running 0 4m17s
kubevious-redis-0 1/1 Running 0 4m16s
kubevious-ui-59b8b84c9-swsck 1/1 Running 0 4m17sView if the PV is bound:
$ kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
kubevious-local-pv 10Gi RWO Retain Bound kubevious/data-kubevious-mysql-0 kubevious-local-storage 118sAlso, view the services created:
$ kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubevious-backend-clusterip ClusterIP 10.106.16.225 <none> 4000/TCP 2m45s
kubevious-collector-clusterip ClusterIP 10.100.198.48 <none> 4000/TCP 2m45s
kubevious-guard-clusterip ClusterIP 10.101.41.10 <none> 4000/TCP 2m45s
kubevious-mysql ClusterIP 10.105.79.115 <none> 3306/TCP 2m45s
kubevious-parser-clusterip ClusterIP 10.104.85.82 <none> 4000/TCP 2m45s
kubevious-redis ClusterIP 10.109.170.171 <none> 6379/TCP 2m45s
kubevious-ui-clusterip ClusterIP 10.102.156.222 <none> 80/TCP 2m45s#4. Exposing the Kubevious Service.
On Kubernetes, there are 3 ways of exposing the service externally. These are NodePort, Loadbalancer, and Ingress.
A. NodePort Service
To expose the service using NodePort, we will execute the command:
$ kubectl expose deployment kubevious-ui --type=NodePort --port=80 --name=kubevious1
service/kubevious1 exposedVerify the creation:
$ kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubevious-backend-clusterip ClusterIP 10.106.16.225 <none> 4000/TCP 5m32s
kubevious-collector-clusterip ClusterIP 10.100.198.48 <none> 4000/TCP 5m32s
kubevious-guard-clusterip ClusterIP 10.101.41.10 <none> 4000/TCP 5m32s
kubevious-mysql ClusterIP 10.105.79.115 <none> 3306/TCP 5m32s
kubevious-parser-clusterip ClusterIP 10.104.85.82 <none> 4000/TCP 5m32s
kubevious-redis ClusterIP 10.109.170.171 <none> 6379/TCP 5m32s
kubevious-ui-clusterip ClusterIP 10.102.156.222 <none> 80/TCP 5m32s
kubevious1 NodePort 10.106.51.191 <none> 80:32167/TCP 8sWe have the Kubevious service exposed on port 32167, allow the port through the firewall if you have it active.
#For UFW
sudo ufw allow 32167
##For Firewalld
sudo firewall-cmd --add-port=32167/tcp --permanent
sudo firewall-cmd --reloadNow access the Kubevious WebUI using the URL http://node_IP:32167
B. LoadBalancer and Ingress (MetalLB)
If you don’t have MetalLB installed already check out our guide below:
Now after the above config, you can expose the Loadbalancer service with the command:
$ kubectl expose deployment kubevious-ui --type=LoadBalancer --port=80 --name=kubevious2
service/kubevious2 exposedVerify the creation:
$ kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubevious-backend-clusterip ClusterIP 10.106.16.225 <none> 4000/TCP 12m
kubevious-collector-clusterip ClusterIP 10.100.198.48 <none> 4000/TCP 12m
kubevious-guard-clusterip ClusterIP 10.101.41.10 <none> 4000/TCP 12m
kubevious-mysql ClusterIP 10.105.79.115 <none> 3306/TCP 12m
kubevious-parser-clusterip ClusterIP 10.104.85.82 <none> 4000/TCP 12m
kubevious-redis ClusterIP 10.109.170.171 <none> 6379/TCP 12m
kubevious-ui-clusterip ClusterIP 10.102.156.222 <none> 80/TCP 12m
kubevious1 NodePort 10.111.165.67 <none> 80:32167/TCP 2m30s
kubevious2 LoadBalancer 10.101.61.192 192.168.205.40 80:30562/TCP 9sNow you can access the service using the provided IP address. For example http://192.168.205.40
- Ingress
For Ingress, we will install Traefik Ingress. First, add the Helm repo:
helm repo add traefik https://helm.traefik.io/traefik
helm repo updateNow install the Traefik Ingress Controller:
helm install traefik traefik/traefikRetrieve the Loadbalancer IP address for the Traefik service:
# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubevious-backend-clusterip ClusterIP 10.96.194.236 <none> 4000/TCP 8m6s
kubevious-collector-clusterip ClusterIP 10.99.102.242 <none> 4000/TCP 8m5s
kubevious-guard-clusterip ClusterIP 10.97.177.127 <none> 4000/TCP 8m5s
kubevious-mysql ClusterIP 10.111.89.154 <none> 3306/TCP 8m5s
kubevious-parser-clusterip ClusterIP 10.103.7.152 <none> 4000/TCP 8m5s
kubevious-redis ClusterIP 10.103.242.85 <none> 6379/TCP 8m5s
kubevious-ui-clusterip ClusterIP 10.107.80.18 <none> 80/TCP 8m5s
kubevious1 NodePort 10.100.213.150 <none> 80:30789/TCP 3m6s
kubevious2 LoadBalancer 10.110.66.189 192.168.205.40 80:30618/TCP 118s
traefik LoadBalancer 10.110.13.33 192.168.205.41 80:30135/TCP,443:30545/TCP 15sAdd a DNS entry for the Traefik service:
$ sudo vim /etc/hosts
192.168.205.41 kubevious.ownyourlife.com.ngNow create and Ingress using the hostname above:
kubectl apply -f - <<EOF
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: kubevious
annotations:
traefik.ingress.kubernetes.io/router.entrypoints: web
spec:
rules:
- host: kubevious.ownyourlife.com.ng
http:
paths:
- path: /
pathType: Prefix
backend:
service:
name: kubevious-ui-clusterip
port:
number: 80
EOFVerify the creation:
$ kubectl get ing | grep kubevious
kubevious <none> kubevious.ownyourlife.com.ng 80 27sYou can now access the Kubevious WebUI using the domain name //kubevious.ownyourlife.com.ng
#5. Access Kubevious Web UI
Once the service has been exposed using any of the 3 methods, you can access the Kubevious WebUI as shown.
Kubevious works with 4 components:
- Observe
- Guard
- Introspect
- Validate
Under Observe,(Browser tab) you can view the objects in the cluster that include Logic, images, packages e.t.c in cluster with the related warnings and errors.
To use Guard, a CLI extension to validate changes for conflicts, misconfigurations, typos, and violations of best practices for Kubernetes clusters and applications.
This can be done by piping the manifest through the Guard script before applying it with the command:
cat manifests.yaml | sh <(curl -sfL https://run.kubevious.io/validate.sh)You can view validators.
View the available rules.
You can view the affected objects
Verdict
This guide has provided you with knowledge on how to manage Kubernetes Cluster with Kubevious. We have also demonstrated how to get started with Kubevious. There are more other use cases not provided in this guide, You are now free to explore them on your own.
See more:
- Managing Kubernetes Cluster using Portainer Web Dashboard
- How To Install Kubernetes Dashboard with NodePort
- How To Create Admin User to Access Kubernetes Dashboard
- Install Lens – Best Kubernetes Dashboard & IDE
Leave a Reply
You must be logged in to post a comment.