Using Helm Charts for Deployment

1. Understanding Helm

Concept	Description	Usage
Helm Chart	Package format for Kubernetes applications	Used to define, install, and upgrade even the most complex Kubernetes applications
Helm Release	An instance of a Helm chart	Represents a specific deployment of a chart in a Kubernetes cluster
Values File	YAML file containing configuration values for a chart	Allows customization of default settings during chart installation
Helm Repository	Remote location storing Helm charts	Used to store and share charts (e.g., Artifact Hub, private repositories)

Example: Installing a Chart with Values

helm install my-release stable/elasticsearch \
  --values ./my-values.yaml

Note: Helm charts are composed of templates, values, and a chart.yaml file.

Key Features of Helm

Template engine for Kubernetes manifests
Version control and release management
Dependency management between charts
Supports custom value files for configuration

2. Adding Elastic Helm Repository

Command	Description	Notes
helm repo add elastic https://helm.elastic.co	Add the official Elastic Helm repository	Required before installing any Elastic stack components via Helm

Example: Adding the Elastic Helm Repository

helm repo add elastic https://helm.elastic.co
helm repo update

Note: Ensure you have helm CLI installed and configured to interact with your Kubernetes cluster.

3. Installing Elasticsearch Chart

Parameter	Default Value	Description
elasticsearch.enabled	`true`	Enables or disables the Elasticsearch chart
elasticsearch.image	`docker.elastic.co/elasticsearch/elasticsearch:8.15.3`	Image to use for Elasticsearch pods
elasticsearch.replicas	`1`	Number of Elasticsearch nodes to deploy
elasticsearch.resources.limits.cpu	`1000m`	Maximum CPU limit for Elasticsearch pods
elasticsearch.resources.limits.memory	`2Gi`	Maximum memory limit for Elasticsearch pods
elasticsearch.persistence.enabled	`true`	Enables persistent volumes for data storage
elasticsearch.persistence.size	`10Gi`	Size of the persistent volume for Elasticsearch data

Example: Installing Elasticsearch Chart with Custom Values

elasticsearch:
  enabled: true
  image: docker.elastic.co/elasticsearch/elasticsearch:8.15.3
  replicas: 3
  resources:
    limits:
      cpu: 1000m
      memory: 2Gi
  persistence:
    enabled: true
    size: 20Gi

Note: Ensure that the Kubernetes cluster has sufficient resources to support the specified Elasticsearch configuration.

4. Installing Kibana Chart

Parameter	Default Value	Description
image.repository	`docker.elastic.co/kibana/kibana`	Kibana Docker image repository
image.tag	`8.15.3`	Kibana Docker image tag (latest stable)
service.type	`ClusterIP`	Kubernetes service type for Kibana
resources.limits.cpu	`1000m`	CPU limit for Kibana pod
resources.limits.memory	`2Gi`	Memory limit for Kibana pod
ingress.enabled	`false`	Enable Ingress for Kibana access

Example: Install Kibana Chart with Custom Values

helm repo add elastic https://helm.elastic.co
helm install my-kibana elastic/kibana \
  --set image.tag=8.15.3 \
  --set service.type=LoadBalancer \
  --set ingress.enabled=true \
  --set ingress.hosts[0]=kibana.example.com

Note: Ensure Elasticsearch is running and accessible before installing Kibana.

Warning: The Kibana chart requires Elasticsearch to be deployed with the same Helm release or accessible via a known service endpoint.

5. Installing Filebeat Chart

Parameter	Default Value	Description
image.repository	`docker.elastic.co/beats/filebeat`	Filebeat Docker image repository
image.tag	`8.15.3`	Filebeat Docker image tag
image.pullPolicy	`IfNotPresent`	Docker image pull policy
serviceAccount.create	`true`	Whether to create a service account for Filebeat
serviceAccount.name	`filebeat`	Name of the service account to use or create
tolerations	`[]`	Tolerations for scheduling Filebeat pods

Example: Installing Filebeat Chart with Custom Values

image:
  repository: docker.elastic.co/beats/filebeat
  tag: 8.15.3
  pullPolicy: Always

serviceAccount:
  create: true
  name: filebeat-sa

tolerations:
  - key: "node-role.kubernetes.io/master"
    operator: "Equal"
    value: "true"
    effect: "NoSchedule"

Note: Ensure that the Kubernetes cluster has the necessary permissions and that the Elastic Stack is properly configured before installing Filebeat.

6. Customizing Values

Example: Overriding Elasticsearch Configuration

elasticsearch:
  config:
    elasticsearch.yml: |
      network.host: 0.0.0.0
      http.port: 9200
      discovery.type: single-node

Value Path	Description	Default	Notes
elasticsearch.config.elasticsearch.yml	Configuration file content for Elasticsearch	`network.host: localhost`	YAML formatted configuration
elasticsearch.resources.limits.cpu	Maximum CPU limit for Elasticsearch pod	`1000m`	Resource limit in millicores
elasticsearch.resources.requests.memory	Memory request for Elasticsearch pod	`1Gi`	Resource request in bytes
kibana.config.kibana.yml	Configuration file content for Kibana	`server.host: "0.0.0.0"`	YAML formatted configuration
logstash.config.logstash.yml	Configuration file content for Logstash	`pipeline.threads: 1`	YAML formatted configuration

Note: Values can be overridden using helm install --set key=value or by providing a custom values.yaml file.

7. Upgrading Releases

Example: Upgrade a Release

helm upgrade my-release elastic/elasticsearch

Option	Values	Description
--set	`key=value`	Set values on the command line
--values	`file.yaml`	Specify values in a YAML file
--version	`1.2.3`	Specify a version constraint for the chart
--atomic	`true`	Rollback on failure
--timeout	`300s`	Time to wait for any individual Kubernetes operation

Note: Ensure that the new version of the chart is compatible with your existing configuration.

Warning: Upgrading may cause downtime if not configured properly.

8. Rolling Back Releases

Action	Description	Command
View release history	Lists all previous versions of a release	`helm history <release-name>`
Rollback to previous version	Revert to a specific revision number	`helm rollback <release-name> <revision>`
Rollback to last successful version	Revert to the most recent working release	`helm rollback <release-name> -1`

Example: Rolling Back a Release

helm history my-elk-stack
helm rollback my-elk-stack 3

Note: Ensure that the target revision has been successfully deployed before rolling back to avoid downtime or data loss.

9. Configuring Persistent Storage

Parameter	Description	Default Value	Notes
elasticsearch.persistence.enabled	Enables persistent storage for Elasticsearch	`false`	Set to `true` to enable persistent volumes
elasticsearch.persistence.storageClass	Storage class to use for persistent volumes	`""`	If not set, uses default storage class
elasticsearch.persistence.size	Size of the persistent volume claim	`10Gi`	Adjust based on data size requirements
elasticsearch.persistence.accessModes	Access modes for the persistent volume	`["ReadWriteOnce"]`	Valid values: `ReadWriteOnce`, `ReadOnlyMany`, `ReadWriteMany`

Example: Enabling Persistent Storage

elasticsearch:
  persistence:
    enabled: true
    storageClass: "fast-ssd"
    size: "50Gi"
    accessModes:
      - ReadWriteOnce

Note: Persistent storage is crucial for data durability in Elasticsearch clusters.

Warning: Changing storage size after deployment may not be supported in all environments.

10. Setting Resource Requests

Resource Type	Property	Default Value
requests.cpu	`500m`	Half a CPU core
requests.memory	`512Mi`	512 Mebibytes of RAM
limits.cpu	`1000m`	One full CPU core
limits.memory	`1Gi`	1 Gibibyte of RAM

Example: Helm Values Configuration

elasticsearch:
  resources:
    requests:
      cpu: "500m"
      memory: "512Mi"
    limits:
      cpu: "1000m"
      memory: "1Gi"

kibana:
  resources:
    requests:
      cpu: "250m"
      memory: "256Mi"
    limits:
      cpu: "500m"
      memory: "512Mi"

Note: Setting appropriate resource requests and limits is crucial for stable deployment in Kubernetes environments.

11. Configuring Ingress

Property	Description	Default Value	Notes
ingress.enabled	Enables or disables the Ingress resource	`false`	Set to `true` to enable Ingress
ingress.hosts[0].host	Hostname for the Ingress rule	`elk-stack.local`	Must be a valid DNS name or IP address
ingress.hosts[0].paths[0]	Path for the Ingress rule	`/`	Can be any valid path pattern
ingress.annotations	Annotations to be added to the Ingress resource	`{}`	Used for configuring TLS, authentication, etc.
ingress.tls	Enables TLS for the Ingress resource	`[]`	Must include secret name and host names

Example: Basic Ingress Configuration

ingress:
  enabled: true
  hosts:
    - host: elk-stack.example.com
      paths:
        - path: /
          pathType: Prefix
  annotations:
    kubernetes.io/ingress.class: nginx
    cert-manager.io/cluster-issuer: letsencrypt-prod
  tls:
    - secretName: elk-stack-tls
      hosts:
        - elk-stack.example.com

Note: Ensure that the Ingress controller is installed and configured in your cluster before enabling this feature.

12. Managing Helm Releases

Action	Description	Command
Install Release	Deploy a new release from a chart	`helm install [RELEASE_NAME] [CHART]`
Upgrade Release	Update an existing release with new values or chart version	`helm upgrade [RELEASE_NAME] [CHART]`
Rollback Release	Revert to a previous release version	`helm rollback [RELEASE_NAME] [REVISION]`
List Releases	Show all installed releases	`helm list`
Show Release Info	Display details of a specific release	`helm show [RELEASE_NAME]`
Delete Release	Uninstall a release completely	`helm uninstall [RELEASE_NAME]`

Example: Install ELK Stack Release

helm install elasticsearch stable/elasticsearch
helm install kibana stable/kibana

Example: Upgrade Release with New Values

helm upgrade elasticsearch stable/elasticsearch --set replicas=3

Managing Helm Release Lifecycle

Install chart using helm install
Verify deployment with helm list
Update configuration using helm upgrade
Revert changes if needed with helm rollback
Cleanup with helm uninstall when done

Note: Helm releases are tracked in the Kubernetes cluster under the helm namespace. Ensure proper RBAC permissions for managing releases.