How KubViz works

Kubviz client can be installed on any Kubernetes cluster. Kubviz agent runs in a kubernetes cluster where the changes/events need to be tracked. The agent detects the changes in real time and send those events via NATS JetStream and the same is received in the kubviz client.

Kubviz client receives the events and passes it to Clickhouse database. The events present in the Clickhouse database can be visualized through Grafana.

KubViz’s event tracking component provides comprehensive visibility into the changes and events occurring within your Kubernetes clusters.

KubViz offers a seamless integration with Git repositories, empowering you to effortlessly track and monitor changes that occur within your codebase. By capturing events such as commits, merges, and other Git activities.

KubViz also monitors changes in your container registry, providing visibility into image updates. By tracking these changes, KubViz helps you proactively manage container security and compliance.

How to install and run Kubviz

Prerequisites

• A Kubernetes cluster
• Helm binary

Prepare Namespace

This command will create a new namespace for your cluster.

kubectl create namespace kubviz

Client Installation

helm repo add kubviz https://intelops.github.io/kubviz/
helm repo update

The following command will generate a token. Please make sure to take note of this token as it will be used for both client and agent installation purposes.

token=$(openssl rand -base64 32 | tr -dc 'a-zA-Z0-9' | fold -w 32 | head -n 1)

helm upgrade -i kubviz-client kubviz/client -n kubviz --set "nats.auth.token=$token"

NOTE:

• If you want to enable Grafana with the client deployment, add --set grafana.enabled=true to the helm upgrade command.

• If grafana already exist use the same upgrade command without –set grafana.enabled=true flag.

helm upgrade -i kubviz-client kubviz/client -n kubviz --set "nats.auth.token=$token" --set grafana.enabled=true

• The KubViz client will also install NATS and Clickhouse. The NATS service is exposed as a LoadBalancer, and you need to note the external IP of the service kubviz-client-nats-external and pass it during the KubViz agent installation.

The following command will retrieve the IP address. Please make sure to take note of this IP address as it will be used for agent installation if your agent is located in a different cluster.

kubectl get services kubviz-client-nats-external -n kubviz --output jsonpath='{.status.loadBalancer.ingress[0].ip}'

NOTE:

• Kubviz-client pod is in a CrashLoopBackOff state, installing the Kubviz-agent will bring it back up and running.

Agent Installation

Deploying Agent on the Same Kubernetes Cluster as kubeviz Client:

1. Make sure you have the KubViz client running on your Kubernetes cluster.
2. Run the following command to deploy the KubViz agent:

helm upgrade -i kubviz-agent kubviz/agent -n kubviz \
  --set "nats.auth.token=$token" \
  --set git_bridge.enabled=true \
  --set "git_bridge.ingress.hosts[0].host=<INGRESS HOSTNAME>",git_bridge.ingress.hosts[0].paths[0].path=/,git_bridge.ingress.hosts[0].paths[0].pathType=Prefix,git_bridge.ingress.tls[0].secretName=<SECRET-NAME>,git_bridge.ingress.tls[0].hosts[0]=<INGRESS HOSTNAME> \ 
  --set container_bridge.enabled=true \
  --set "container_bridge.ingress.hosts[0].host=<INGRESS HOSTNAME>",container_bridge.ingress.hosts[0].paths[0].path=/,container_bridge.ingress.hosts[0].paths[0].pathType=Prefix,container_bridge.ingress.tls[0].secretName=<SECRET-NAME>,container_bridge.ingress.tls[0].hosts[0]=<INGRESS HOSTNAME>

3. Replace “INGRESS HOSTNAME” with the desired hostname for the Git Bridge and Container Bridge Ingress configurations.
4. Replace “SECRET-NAME” with the desired secretname for the Git Bridge and Container Bridge Ingress configurations.

NOTE:

• Default Annotations for Ingress

By default, this Helm chart includes the following annotations for the git bridge and container bridge ingress resource:

annotations:
  cert-manager.io/cluster-issuer: letsencrypt-prod-cluster
  kubernetes.io/force-ssl-redirect: "true"
  kubernetes.io/ssl-redirect: "true"
  kubernetes.io/tls-acme: "true"
...

If you do not want to use the default value, you can modify the annotation in values.yaml and execute the following command:

helm upgrade -i kubviz-agent kubviz/agent -f values.yaml -n kubviz

Deploying Agent on a Different Kubernetes Cluster:

1. Run the following command to deploy the KubViz agent:

helm upgrade -i kubviz-agent kubviz/agent -n kubviz --set nats.host=<NATS IP Address> --set "nats.auth.token=$token"   

2. Replace “” with the IP address of your NATS service kubviz-client-nats-external.

How to Verify if Everything is Up and Running

After completing the installation of both the client and agent, you can use the following command to verify if they are up and running.

kubectl get all -n kubviz

Configuration

Once everything is up and running, you need to perform additional configurations to monitor git repository events and container registry events.

To ensure that these events are sent to KubViz, you need to create a webhook for your repository. This webhook will transmit the event data of the specific repository or registry to KubViz.

To set up a webhook in your repository, please follow these steps

How to View Event Data in Grafana

1. Retrieve your Grafana login password by running the following command:

kubectl get secret --namespace kubviz kubviz-client-grafana -o jsonpath="{.data.admin-password}" | base64 --decode ; echo

2. Get the Grafana URL to visit by running these commands in the same shell:

export POD_NAME=$(kubectl get pods --namespace kubviz -l "app.kubernetes.io/name=grafana,app.kubernetes.io/instance=kubviz-client" -o jsonpath="{.items[0].metadata.name}")

kubectl --namespace kubviz port-forward $POD_NAME 3000

3. Access “localhost:3000” in your web browser, where you’ll be prompted to enter your credentials. Utilize the username “admin” and the password obtained from step 1 to proceed.