Tanzu Community Edition

Documentation

Helm Controller

The helm-controller is a Kubernetes operator, allowing one to declaratively manage Helm chart releases. It is part of a composable GitOps toolkit and depends on source-controller to acquire the Helm charts from Helm repositories.

Configuration

The Helm controller package has following configurable properties.

ValueRequired/OptionalDescription
namespaceOptionalSets namespace for k8s objects where resources of helm-controller with be created.
limits_cpuOptionalSets maximum usage of cpu for helm-controller deployment.
limits_memoryOptionalSets maximum usage of memory for helm-controller deployment.
namespace: helm-system
limits_cpu: 1000m
limits_memory: 1Gi

Installation

To install FluxCD helm-controller from the Tanzu Application Platform package repository:

  1. Prerequisites:-

    Install fluxcd-source-controller package:-

    Helm controller package requires source-controller package to be installed for acquiring the Kubernetes manifests from the sources.

    tanzu package install fluxcd-source-controller -p fluxcd-source-controller.community.tanzu.vmware.com -v VERSION-NUMBER

    here VERSION-NUMBER is the version of the package

    For example:

    $ tanzu package install fluxcd-source-controller -p fluxcd-source-controller.community.tanzu.vmware.com -v 0.21.2
\ Installing package 'fluxcd-source-controller.community.tanzu.vmware.com'
| Getting package metadata for 'fluxcd-source-controller.community.tanzu.vmware.com'
| Creating service account 'fluxcd-source-controller-default-sa'
| Creating cluster admin role 'fluxcd-source-controller-default-cluster-role'
| Creating cluster role binding 'fluxcd-source-controller-default-cluster-rolebinding'
| Creating package resource
/ Waiting for 'PackageInstall' reconciliation for 'fluxcd-source-controller'
\ 'PackageInstall' resource install status: Reconciling

Added installed package 'fluxcd-source-controller'

  2. List version information for the package by running:

    tanzu package available list helm-controller.fluxcd.community.tanzu.vmware.com

    For example:

    $ tanzu package available list helm-controller.fluxcd.community.tanzu.vmware.com
/ Retrieving package versions for helm-controller.fluxcd.community.tanzu.vmware.com...  
NAME                                                    VERSION  RELEASED-AT  
helm-controller.fluxcd.community.tanzu.vmware.com        0.17.2   2022-02-23 16:44:08 +0530 IST

  3. Configure helm-controller package:

    User can optionally provide the configuration parameters with –values-file flag while installing the package. Download the values.yaml file from values.yaml.

    $ cat values.yaml  
namespace: helm-system
limits_cpu: 1000m
limits_memory: 1Gi

  4. Install the package by running:

    tanzu package install fluxcd-helm-controller -p helm-controller.fluxcd.community.tanzu.vmware.com -v VERSION-NUMBER

    here: VERSION-NUMBER is the version of the package

    For example:

    $ tanzu package install fluxcd-helm-controller -p helm-controller.fluxcd.community.tanzu.vmware.com -v 0.17.2
/ Installing package 'helm-controller.fluxcd.community.tanzu.vmware.com'
| Getting package metadata for 'helm-controller.fluxcd.community.tanzu.vmware.com'
| Creating service account 'fluxcd-helm-controller-default-sa'
| Creating cluster admin role 'fluxcd-helm-controller-default-cluster-role'
| Creating cluster role binding 'fluxcd-helm-controller-default-cluster-rolebinding'
| Creating package resource
/ Waiting for 'PackageInstall' reconciliation for 'fluxcd-helm-controller'
\ 'PackageInstall' resource install status: Reconciling

Added installed package 'fluxcd-helm-controller'

  5. Verify the package install by running:

    tanzu package installed get fluxcd-helm-controller

    For example:

    - Retrieving installation details for fluxcd-helm-controller...
NAME:                    fluxcd-helm-controller
PACKAGE-NAME:            helm-controller.fluxcd.community.tanzu.vmware.com
PACKAGE-VERSION:         0.17.2
STATUS:                  Reconcile succeeded
CONDITIONS:              [{ReconcileSucceeded True  }]

    Verify that STATUS is Reconcile succeeded

    kubectl get pods -n helm-system

    For example:

    $ kubectl get pods -n helm-system
NAME                               READY   STATUS    RESTARTS   AGE
helm-controller-7748954977-cgv59   1/1     Running   2          157m

    Verify that STATUS is Running

Try helm-controller

  1. Verify all the objects are installed:

    This package would create a new namespace where all the elements of fluxcd will be hosted called helm-system you can verify the main components of helm-controller were installed by running:

    $ kubectl get all -n helm-system
NAME                                   READY   STATUS    RESTARTS   AGE
pod/helm-controller-7748954977-cgv59   1/1     Running   0          3m28s

NAME                              READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/helm-controller   1/1     1            1           3m28s

NAME                                         DESIRED   CURRENT   READY   AGE
replicaset.apps/helm-controller-7748954977   1         1         1       3m28s

    you should get something really similar!

  2. Verify all the CRD were installed correctly:

    The way you would communicate with helm-controller would be through its CRDs, this will be your main action point.

    In order to check all the CRDs were installed you can run:

     $ kubectl get crds | grep ".fluxcd.io"
 buckets.source.toolkit.fluxcd.io                                               2022-04-18T07:29:09Z
 gitrepositories.source.toolkit.fluxcd.io                                       2022-04-18T07:29:09Z
 helmcharts.source.toolkit.fluxcd.io                                            2022-04-18T07:29:09Z
 helmreleases.helm.toolkit.fluxcd.io                                            2022-04-18T07:40:57Z
 helmrepositories.source.toolkit.fluxcd.io                                      2022-04-18T07:29:09Z

    Apart from helmreleases.helm.toolkit.fluxcd.io all the other CRDs are installed as part of the fluxcd-source-controller (which is a pre-requisite) to get helm-controller work as expected.

  3. Try one quick example yourself, so you can check everything is working as expected

    • Let’s consume a HelmRelease and HelmRepository object.

    • Create the following sample.yaml file:

      Note: flux-system namespace is created as part of the pre-requisite - installing fluxcd-source-controller

    ---
apiVersion: helm.toolkit.fluxcd.io/v2beta1
kind: HelmRelease
metadata:
  name: podinfo
  namespace: flux-system
spec:
  interval: 5m
  chart:
    spec:
      chart: podinfo
      version: '4.0.x'
      sourceRef:
        kind: HelmRepository
        name: podinfo
        namespace: flux-system
      interval: 1m
  values:
    replicaCount: 2
---
apiVersion: source.toolkit.fluxcd.io/v1beta1
kind: HelmRepository
metadata:
  name: podinfo
  namespace: flux-system
spec:
  interval: 1m
  url: https://stefanprodan.github.io/podinfo
    • Apply the created conf
    $ kubectl apply -f sample.yaml
helmrelease.helm.toolkit.fluxcd.io/podinfo created
helmrepository.source.toolkit.fluxcd.io/podinfo created
    • Check HelmChart was pulled successfully
    • Check the HelmRepository was fetched successfully
    • Check the HelmRelease is reconciled successfully
    $ kubectl -n flux-system get helmreleases.helm.toolkit.fluxcd.io
NAME      READY   STATUS                             AGE
podinfo   True    Release reconciliation succeeded   27s

$ kubectl -n flux-system get helmcharts.source.toolkit.fluxcd.io
NAME                  CHART     VERSION   SOURCE KIND      SOURCE NAME   READY   STATUS                                         AGE
flux-system-podinfo    podinfo   4.0.x     HelmRepository   podinfo       True    Pulled 'podinfo' chart with version '4.0.6'.   42s

$ kubectl -n flux-system get helmrepositories.source.toolkit.fluxcd.io
NAME      URL                                      READY   STATUS                                                                               AGE
podinfo   https://stefanprodan.github.io/podinfo   True    Fetched revision: debd9e30ecc98721b7aa3404eb64db4032786e0572be4629359aa7b6cf0f5fe7   57s

$ kubectl get all -n flux-system
NAME                                     READY   STATUS    RESTARTS   AGE
pod/podinfo-7ddf7dd478-5d7dw             1/1     Running   0          12m
pod/podinfo-7ddf7dd478-rdk6k             1/1     Running   0          12m
pod/source-controller-66ccb7ff5c-j4q4l   1/1     Running   2          168m

NAME                        TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)             AGE
service/podinfo             ClusterIP   10.96.42.37    <none>        9898/TCP,9999/TCP   12m
service/source-controller   ClusterIP   10.96.133.58   <none>        80/TCP              168m

NAME                                READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/podinfo             2/2     2            2           12m
deployment.apps/source-controller   1/1     1            1           168m

NAME                                           DESIRED   CURRENT   READY   AGE
replicaset.apps/podinfo-7ddf7dd478             2         2         2       12m
replicaset.apps/source-controller-66ccb7ff5c   1         1         1       168m

    You can find more examples checking out the samples folder on fluxcd/helm-controller/samples

Documentation

For documentation specific to helm-controller, check out the main repository fluxcd/helm-controller.

Join us!

Our open community welcomes all users and contributors

Community