Multiple Sources for Applications in Argo CD

OpenShift GitOps

- Thomas Jungbauer Thomas Jungbauer ( Lastmod: 2024-06-02 ) - 5 min read

Argo CD or OpenShift GitOps uses Applications or ApplicationSets to define the relationship between a source (Git) and a cluster. Typically, this is a 1:1 link, which means one Application is using one source to compare the cluster status. This can be a limitation. For example, if you are working with Helm Charts and a Helm repository, you do not want to re-build (or re-release) the whole chart just because you made a small change in the values file that is packaged into the repository. You want to separate the configuration of the chart with the Helm package.

The most common scenarios for multiple sources are (see: Argo CD documentation):

  • Your organization wants to use an external/public Helm chart

  • You want to override the Helm values with your own local values

  • You don’t want to clone the Helm chart locally as well because that would lead to duplication and you would need to monitor it manually for upstream changes.

This small article describes three different ways with a working example and tries to cover the advantages and disadvantages of each of them. They might be opinionated but some of them proved to be easier to use and manage.

Option 1: Multisource by Argo CD

Specifying multiple sources for an application is a beta feature.

The first option I would like to demonstrate is the "official way" by Argo CD. It was introduced in Argo CD version 2.6 and while this option is still in the Beta phase, it is one of the most requested features and was added to the release candidate of version 2.11. The article by Argo CD gives more details of this release. The official documentation by Argo CD can be found here

In my repository I am using this option as an example to create the App-of-Apps. The App-of-Apps created an Application, that uses multiple sources.

  1. The values files from https://github.com/tjungbauer/openshift-clusterconfig-gitops/blob/main/base/argocd-resources-manager/values.yaml

  2. The Helm Chart helper-argocd from https://charts.stderr.at/

Let’s see the working example:

apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
  name: argocd-resources-manager
  namespace: openshift-gitops
spec:
  destination:
    namespace: openshift-gitops
    server: 'https://kubernetes.default.svc'
  info:
    - name: Description
      value: >-
        This is the starting point which will initialize all applicationsets or
        argocd applications
  project: default
  sources: (1)
    - chart: helper-argocd (2)
      helm:
        valueFiles: (3)
          - $values/base/argocd-resources-manager/values.yaml
      repoURL: 'https://charts.stderr.at/' (4)
      targetRevision: 2.0.28 (5)
    - ref: values (6)
      repoURL: 'https://github.com/tjungbauer/openshift-clusterconfig-gitops'
      targetRevision: main
  syncPolicy:
    automated:
      prune: true
      selfHeal: true
1Using sources instead of singular source
2I want to use the Helm Chart helper-argocd
3The chart will use the values file(s) from github.com/tjungbauer/openshift-clusterconfig-gitops/base/argocd-resources-manager/values.yaml. $value (can only be specified at the beginning of the path) resolves to the root of the values file repository
4The URL of the Helm Chart repository
5The version of the Helm chart helper-argocd
6The reference to the values file, defining the repository URL and target revision.
Argo CD does not currently support using another Helm chart as a source for value files.

With this option, it is possible to separate the values files from the Helm Chart itself. Whenever I want to change something in the configuration, I simply change the values file without being required to release a new Chart version.

Advantages

  • Allows easily to define an Application with multiple sources, using Argo CD features.

  • No additional tool (see other options below) is required

  • Was added to the release candidate of Argo CD v2.11

Disadvantages

  • Currently a Beta feature and thus is not supported and has some limitations, such as lacking support of CLI and UI.

  • Does not allow using additional local specifications. At least as far as I know.

  • Can be complex to configure and manage

Option 2: Wrapper Helm Chart

With this option, which I extensively use, a wrapper Helm Chart is used to define local values files while calling additional (sub) Helm Charts as a dependency. This wrapper could simply define the values files and nothing else (being empty instead) or even define additional files, such as Sealed Secrets or things that are not provided by the sub-charts.

The first one is using sub-charts to build the required specifications:

dependencies:
  - name: rhacs-setup
    version: ~1.0.0
    repository: https://charts.stderr.at/
  - name: helper-operator
    version: ~1.0.23
    repository: https://charts.stderr.at/
  - name: helper-status-checker
    version: ~4.0.0
    repository: https://charts.stderr.at/
    condition: helper-status-checker.enabled

The values file specifies the configuration for these sub-charts.

The second example also uses sub-charts, but additionally defines local files such as a SealedSecret for htpasswd.

As you can see throughout my repository I am using this option almost all the time. It proved to be quite simple, especially if you prefer working with Helm Charts such as I do. However, you must take care of the settings in the values file. Specifications you would like to be presented in a sub-chart must be put into the correct place.

For example:

Everything underneath

helper-operator:
  operators:
    rhacs-operator:
[...]

will be used by the chart helper-operator. While everything underneath:

helper-status-checker:
  enabled: true

will be used by the chart helper-status-checker.

Advantages

  • Easy to use, at least for myself

  • Allows defining additional, local files that are not provided by the sub-charts

Disadvantages

  • A wrapper Chart must be created, that at least defines: Chart.yaml, templates folder and values.yaml

  • The configuration must be done correctly and all settings for a sub-chart must be forwarded to the sub-chart.

Option 3: Using Kustomize with Helm enabled

The third option I would like to show is using Kustomize. This tool can be used to call a Helm Chart when the option --enable-helm is activated in Argo CD. I am using one example for the IngressController.

Here the values file is placed into the local folder and the Kustomize.yaml is configured as:

apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization
helmCharts:
- name: ingresscontroller
  repo: https://charts.stderr.at
  valuesFile: values.yaml

This simply defines the use of the chart ingresscontroller from the Helm repository with the local values file.

Unlike Option #2 you do not need to take care about the settings for sub-charts and which settings are passed to which chart. Like Option #2 you can also define additional files that shall be rendered using out-of-the-box Kustomize possibilities.

Advantages

  • Easy to use

  • Allows defining additional, local files that are not provided by the sub-charts

  • No need to take care of sub-charts and correctly pass the settings to a sub-chart

Disadvantages

  • combines two different tools, which might become confusing

  • requires specific option to be enabled --enable-helm

Conclusion

As this short article demonstrates, there are multiple ways to work with multiple sources and therefore to separate the values file from the actual Helm Chart. There might be even more options, but these are the ones I was seeing at customers.

Which one do I use? When you look at my repository you see that I mainly use Option #2. Actually, I completely moved from Option #3 to Option #2 a few months ago, because this proved to be clearer for customers, especially when they are new to Kustomize and Helm. That way, only one tool is used and must be managed. Option #3 proved to be more complex in such a case.

What about Option #1? While I am using it to showcase this feature it is still in a TechPreview phase. However, I do not think that it will completely replace the other options, because it is more complex to configure.

However, in the end, it is all about personal preferences. Use the tool that you feel most comfortable with :).

See Also:

  • Managing Certificates using GitOps approach –

    The article SSL Certificate Management for OpenShift on AWS explains how to use the Cert-Manager Operator to request and install a new SSL Certificate. This time, I would like to leverage the GitOps approach using the Helm Chart cert-manager I have prepared to deploy the Operator and order new Certificates. I will use an ACME Letsencrypt issuer with a DNS challenge. My domain is hosted at AWS Route 53. However, any other integration can be easily used.

  • Update Cluster Version using GitOps approach –

    During a GitOps journey at one point, the question arises, how to update a cluster? Nowadays it is very easy to update a cluster using CLI or WebUI, so why bother with GitOps in that case? The reason is simple: Using GitOps you can be sure that all clusters are updated to the correct, required version and the version of each cluster is also managed in Git. All you need is the channel you want to use and the desired cluster version. Optionally, you can define the exact image SHA. This might be required when you are operating in a restricted environment.

  • Installing OpenShift Logging using GitOps –

    OpenShift Logging is one of the more complex things to install and configure on an OpenShift cluster. Not because the service or Operators are so complex to understand, but because of the dependencies logging has. Besides the logging operator itself, the Loki operator is required, the Loki operator requires access to an object storage, that might be configured or is already available. In this article, I would like to demonstrate the configuration of the full stack using an object storage from OpenShift Data Foundation. This means: Installing the logging operator into the namespace openshift-logging Installing the Loki operator into the namespace openshift-operators-redhat Creating a new BackingStore and BucketClass Generating the Secret for Loki to authenticate against the object storage Configuring the LokiStack resource Configuring the ClusterLogging resource All steps will be done automatically. In case you have S3 storage available, or you are not using OpenShift Data Foundation, the setup will be a bit different. For example, you do not need to create a BackingStore or the Loki authentication Secret.

  • Configure Buckets in MinIO using GitOps –

    MinIO is a simple, S3-compatible object storage, built for high-performance and large-scale environments. It can be installed as an Operator to Openshift. In addition, to a command line tool, it provides a WebUI where all settings can be done, especially creating and configuring new buckets. Currently, this is not possible in a declarative GitOps-friendly way. Therefore, I created the Helm chart minio configurator, that will start a Kubernetes Job, which will take care of the configuration. Honestly, when I say I have created it, the truth is, that it is based on an existing MinIO Chart by Bitnami, that does much more than just set up a bucket. I took out the bucket configuration part, streamlined it a bit and added some new features, which I required. This article shall explain how to achieve this.

  • Setup & Configure Advanced Cluster Security using GitOps –

    Today I want to demonstrate the deployment and configuration of Advanced Cluster Security (ACS) using a GitOps approach. The required operator shall be installed, verified if it is running and then ACS shall be initialized. This initialization contains the deployment of several components: Central - as UI and as a main component of ACS SecuredClusters - installs a Scanner, Controller pods etc. Console link into OpenShift UI - to directly access the ACS Central UI Job to create an initialization bundle to install the Secured Cluster Job to configure authentication using OpenShift Let’s start …​

Copyright © 2020 - 2024 Toni Schmidbauer & Thomas Jungbauer