Setting up

Kubirds is distributed as a Kubernetes Operator and, as such, will require a working Kubernetes Cluster.

In the next part of this guide, we will assume that your cluster is up and running, and that kubectl is properly configured.

Kubirds is cloud-agnostic, meaning it will work flawlessly on any cloud provider, such as:

• DigitalOcean Kubernetes
• Google Kubernetes Engine
• Amazon Elastic Kubernetes Service
• Microsoft Azure Kubernetes Service
• ...

NB: This guide describes the installation method for the Freemium version of Kubirds. If you are interested by the complete version, please contact us.

Requirements

In this guide, we will install Kubirds in the monitoring-system namespace, feel free to change it to your preferences. To create the namespace, run the command below:

kubectl create namespace monitoring-system

Kubirds relies on other technologies to implement the supervision behavior.

In the next part of this guide, we will set up those dependencies.

NB: The installation procedures for those dependencies described here are only one of many.

These instructions will help you deploy Kubirds on a brand new Kubernetes cluster.

If you already deployed those components, you can skip to the next part of this guide.

TektonCD

TektonCD is a Cloud Native CI/CD solution. Kubirds relies on it to build and run your supervision pipelines.

There are currently 2 methods to deploy TektonCD:

Deploy manually

To install the core component of Tekton, Tekton Pipelines, run the command below:

kubectl apply -f "https://storage.googleapis.com/tekton-releases/pipeline/latest/release.yaml"

It may take a few moments before the installation completes. You can check the progress with the following command:

kubectl get pods --namespace tekton-pipelines

Confirm that every component listed has the status Running.

NB: TektonCD is composed of 3 components:

• Tekton Pipelines: define and execute Tasks and Pipelines
• Tekton Triggers: listen for events, and trigger the execution of Tasks and Pipelines
• Tekton Dashboard: visualize your tasks and pipelines

Kubirds only needs Tekton Pipelines, but feel free to deploy the other components as well.

Deploy with the Tekton Operator

First, install the operator with the following command:

kubectl apply -f "https://storage.googleapis.com/tekton-releases/operator/latest/release.yaml"

Then, create the following resource to deploy the Tekton components:

---
apiVersion: operator.tekton.dev/v1alpha1
kind: TektonConfig
metadata:
  name: config
spec:
  profile: lite
  targetNamespace: pipeline-system

There are 3 profiles availables:

• lite: installs Tekton Pipelines
• basic: installs Tekton Pipelines and Tekton Triggers
• all: install all components

Deploying the Kubirds Operator

Kubirds is deployed through an Helm Chart.

Deploy the Helm Chart by running the following commands:

helm repo add link-society-incubator "https://charts.link-society.com/incubator"
helm repo update
helm upgrade --install \
    kubirds link-society-incubator/kubirds \
    --namespace monitoring-system \
    --wait

NB: For more information about the chart options, read this document.

Implement your supervision

In this example, we'll set up monitoring for this website.

We want to verify that the website is accessible, and be notified on Slack as soon as it's not the case.

We'll perform a simple curl every 5 minutes in a Unit and trigger a Reactor when things go bad, and once they are back to normal.

Your first unit

In order to schedule the curl command every 5 minutes, we'll use the curlimages/curl:latest Docker image.

Create the following resource on your Kubernetes cluster:

---
apiVersion: kubirds.com/v1
kind: Unit
metadata:
  name: check-kubirds-access
  namespace: default
  labels:
    app: kubirds-www
spec:
  schedule: every 5 minutes
  image:
    name: curlimages/curl:latest
    pullPolicy: Always
    command: >
            curl -L -v $HOST
  env:
    - name: HOST
      value: https://kubirds.com

Your first reactor

Now, we want to notify a Slack channel. Make sure you created a Slack Webhook.

Once your Slack Webhook is created, create a secret on your Kubernetes cluster containing its URL, for example:

---
apiVersion: v1
kind: Secret
metadata:
  name: my-slack-webhook
  namespace: default
stringData:
  URL: "<YOUR WEBHOOK URL>"

In order to notify Slack, we'll also use curl to push the notification.

Create the following resource on your Kubernetes cluster:

---
apiVersion: kubirds.com/v1
kind: Reactor
metadata:
  name: slack-notifier
  namespace: default
spec:
  unitSelector:
    app: kubirds-www
  triggers:
    success: no
    failure: no
    fixed: yes
    regression: yes
  image:
    name: curlimages/curl:latest
    pullPolicy: Always
    command: >
      curl -X POST $HOST -H "Content-Type: application/json" \
           -d "{ \"text\": \"$UNIT_NAME reported state $UNIT_STATE\" }"      
  env:
    - name: HOST
      valueFrom:
        secretKeyRef:
          name: my-slack-webhook
          key: URL

Once the site is down, you'll receive the following notification on Slack:

And once it's back to normal:

What's next

Kubirds comes with a lot of integrations with other systems out of the box. Those will help you build your supervision with ease.

If you wish to dive deeper into Kubirds features and inner workings, please consult the Documentation.