Deployment Template

Introduction

In Devtron, a Deployment Template defines how your application should run by defining its specifications. Devtron uses Helm charts to manage these deployments, allowing you to control everything from Resource Allocation to environment variables.

You can use default Deployment Charts provided by Devtron or Custom Deployment Charts developed by a super-admin to suit your needs.

This guide covers how to:

• Select a Deployment Chart

• Choose a Chart Version

• Configure the Deployment Template

• Enable Application Metrics

• Perform a Dry Run

• Extra: Edit a Protected Deployment Configuration

Select a Deployment Chart Type

1. Go to the Configurations page of your application.

1. Click Base Configuration → Deployment Template.

1. Select the Chart drop-down box. The following tabs are displayed:

• Charts by Devtron - Displays the default deployment charts provided by Devtron

• Custom charts - Displays your custom deployment charts (if available). To create a custom deployment chart, refer to Deployment Charts.

Choose a Chart Version

Devtron maintains multiple chart versions for each chart type. Additionally, each chart version has a supporting README file that you can use to know more about the features and variables.

Once you select a deployment chart, choose a chart version with which you wish to deploy your application from the Version drop-down box. By default, the latest version of the helm chart is selected.

Configure the Deployment Template

You can edit a deployment template using the following two ways:

• Using GUI

• Using YAML

Using GUI

If you prefer to use a simple way to configure your chosen deployment chart, select GUI.

By default, the following fields are commonly available for you to modify in the GUI section of most charts:

If you wish to perform additional configurations, click the Switch to Advanced button or YAML button. Or perform a dry run before saving your configuration.

Customize the GUI

By default, the GUI section comes with multiple predefined fields as seen earlier in the table. However, if you wish to display a different set of fields to your team, you can modify the whole section as per your requirement.

This is useful in scenarios where:

• Your team members find it difficult to understand and edit the YAML section.

• You frequently edit certain fields in YAML, which you expect to remain easily accessible in GUI section.

• You don't require some fields in GUI section.

• You need the autonomy to keep the GUI unique for applications/clusters/environments/charts, or display the same GUI everywhere.

There are two ways you can customize the GUI, use any one of the following:

1. From Deployment Charts section

2. Using APIs (explained below)

You can pass a custom JSON (deployment schema) of your choice through the following API. You may need to run the API with the POST method if you are doing it for the first time.

PUT {{DEVTRON_BASEURL}}/orchestrator/deployment/template/schema

{
  "name": "schema-1",
  "type": "JSON",
  "schema": "{\"type\":\"object\",\"properties\":{\"args\":{\"type\":\"object\",\"title\":\"Arguments\",\"properties\":{\"value\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"title\":\"Value\"},\"enabled\":{\"type\":\"boolean\",\"title\":\"Enabled\"}}},\"command\":{\"type\":\"object\",\"title\":\"Command\",\"properties\":{\"value\":{\"type\":\"array\",\"items\":{\"type\":\"string\"},\"title\":\"Value\"},\"enabled\":{\"type\":\"boolean\",\"title\":\"Enabled\"}}},\"resources\":{\"type\":\"object\",\"title\":\"Resources(CPU&RAM)\",\"properties\":{\"limits\":{\"type\":\"object\",\"required\":[\"cpu\",\"memory\"],\"properties\":{\"cpu\":{\"type\":\"string\"},\"memory\":{\"type\":\"string\"}}},\"requests\":{\"type\":\"object\",\"properties\":{\"cpu\":{\"type\":\"string\"},\"memory\":{\"type\":\"string\"}}}}},\"autoscaling\":{\"type\":\"object\",\"title\":\"Autoscaling\",\"properties\":{\"MaxReplicas\":{\"type\":[\"integer\",\"string\"],\"title\":\"MaximumReplicas\",\"pattern\":\"^[a-zA-Z0-9-+\\\\/*%_\\\\\\\\s]+$\"},\"MinReplicas\":{\"type\":[\"integer\",\"string\"],\"title\":\"MinimumReplicas\",\"pattern\":\"^[a-zA-Z0-9-+\\\\/*%_\\\\\\\\s]+$\"},\"TargetCPUUtilizationPercentage\":{\"type\":[\"integer\",\"string\"],\"title\":\"TargetCPUUtilizationPercentage\",\"pattern\":\"^[a-zA-Z0-9-+\\\\/*%_\\\\\\\\s]+$\"},\"TargetMemoryUtilizationPercentage\":{\"type\":[\"integer\",\"string\"],\"title\":\"TargetMemoryUtilizationPercentage\",\"pattern\":\"^[a-zA-Z0-9-+\\\\/*%_\\\\\\\\s]+$\"}}},\"EnvVariables\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{\"key\":{\"type\":\"string\"},\"value\":{\"type\":\"string\"}}},\"title\":\"EnvironmentVariables\"},\"ContainerPort\":{\"type\":\"array\",\"items\":{\"type\":\"object\",\"properties\":{\"port\":{\"type\":\"integer\"}}},\"title\":\"ContainerPort\"}}}",
  "selectors": [
    {
      "attributeSelector": {
        "category": "APP",
        "appNames": ["my-demo-app"]
      }
    },
    {
      "attributeSelector": {
        "category": "ENV",
        "envNames": ["env1", "env2", "env3"]
      }
    },
    {
      "attributeSelector": {
        "category": "CLUSTER",
        "clusterNames": ["cluster1", "cluster2", "cluster3"]
      }
    },
    {
      "attributeSelector": {
        "category": "CHART_REF",
        "chartVersions": [
          {
            "type": "Deployment",
            "version": "1.0.0"
          }
        ]
      }
    },
    {
      "attributeSelector": {
        "category": "APP_ENV",
        "appEnvNames": [
          {
            "appName": "my-demo-app",
            "envName": "devtron"
          }
        ]
      }
    }
  ]
}

1. In the name field, give a name to your schema, e.g., schema-1

2. Enter the type as JSON.

3. The schema field is for entering your custom JSON representing the deployment template's GUI. You can customize the fields that are displayed in the GUI section while configuring the deployment template. Perform the following steps:

• To create a custom JSON for your deployment, you may use RJSF JSON Schema Tool.

• Copy the final JSON and stringify it using any free online tool.

• Paste the stringified JSON in the schema field of the API request body.

• Send the API request. If your schema already exists, use the PUT method instead of POST in the API call.

1. The attributeSelector object helps you choose the scope at which your custom JSON will take effect.

   Priority

   Category Scope

   Description

   1 (High)

   APP_ENV

   Specific to an application and its environment

   2

   APP

   Applies at the application level if no specific environment is defined

   3

   ENV

   Applies to specific deployment environment

   4

   CHART_REF

   Applies to all applications using a specific chart type and version

   5

   CLUSTER

   Applies across all applications and environments within a specific cluster

   6

   GLOBAL

   Universally applies if no other more specific schemas are defined

Using YAML

If you prefer to perform additional configurations in your chosen deployment template, select YAML.

Every chart version has its own YAML file that provides specifications for your application. To make it easy to use, we have created templates for the YAML file and added some variables inside the YAML. You can provide or change the values of these variables as per your requirement.

Refer the respective templates to view the YAML details.

• Deployment

• Rollout Deployment

• Job & CronJob

• StatefulSet

Before saving your configuration in YAML, make sure to perform a dry run.

Enable Application Metrics

The availability of application metrics depends on the selected chart type and version. If supported, you can view key performance metrics such as:

• Status codes (2xx, 3xx, 5xx)

• Throughput

• Latency

• And more

To enable this, turn on the Show application metrics toggle.

Once enabled, you can view the application metrics on the App Details page.

Perform a Dry Run

Before saving your configured deployment template, you can use the Dry Run option (as shown below) to preview the final Kubernetes manifests.

This feature helps you verify your configurations, detect issues, and ensure correctness before actual deployment.

Your configurations will appear in the left pane, while the right pane will display a section named Manifest generated from merged showing the computed Kubernetes manifests, each representing a separate resource after merging all your changes.

Edit a Protected Deployment Template

Any changes made to the deployment template will require approval if an approval policy is enforced. To check if your deployment template is protected, check the stamp/approve symbol as shown below.

Request Approval for Changes

Let's assume you are the application admin and your deployment template in Base Configurations is protected from edits.

1. In the YAML editor of the deployment template, modify the values.

   
2. You can change the value of a key to a desired value as shown below. Once done, click the Save Changes button.

   

1. Since the deployment configuration is protected, your changes won't be published right away. You can do either of the following:

   • Save as draft : Selecting this option will save your file as a draft. You and other users can view and edit the saved draft and propose it further for approval.

   • Save & Propose Changes : Selecting this option will allow you to choose and notify the configuration approver(s) and propose your changes for a review. Moreover, it also has the option to notify all the users having access to the application on Devtron.

   Since we are proposing the changes immediately, click Propose Changes.

   
2. You can also view the approval status if you wish.

   

Only one draft can exist at time and you cannot create multiple drafts. In the top-right corner, you have the option to discard the draft if you don't wish to proceed with the edits you made.

Grant Approval for Changes

Go to the edited configuration file to review and approve the changes as shown below.