> For the complete documentation index, see [llms.txt](https://arpit-1.gitbook.io/cost-visibility/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://arpit-1.gitbook.io/cost-visibility/cost-visibilty/configurations.md).

# configurations

The **Configurations** page lets you manage configurations for Cost Visibility in Devtron. You can:

* Set the default currency for all cost-related data.
* Enable or disable cost tracking for your connected clusters.

***

## Currency (Default)

You can select your preferred currency as default.

***

## Enable Cost Tracking

To enable cost visibility for a cluster, follow the steps below:

{% stepper %}
{% step %}

### Choose Cluster and Open Edit Modal

Choose your preferred cluster, and click on **Off/Enabled**. An **Edit Cluster** modal window will open.
{% endstep %}

{% step %}

### Enable Cost Tracking Toggle

Enable the toggle next to **Enable cost tracking**.
{% endstep %}

{% step %}

### Select Cloud Provider

Select the cloud provider in which you have created your cluster.
{% endstep %}

{% step %}

### Provider-specific Configuration

Based on the cloud provider you need to do the configurations described in the tabs below.
{% endstep %}

{% step %}

### Enter Prometheus Endpoint

Enter Prometheus endpoint of your cluster. Refer to the "Fetching Prometheus Endpoint" section below to learn more.
{% endstep %}

{% step %}

### Save

Click **Save**.
{% endstep %}
{% endstepper %}

***

{% tabs %}
{% tab title="GCP" %}
To enable cost visibility for Google Cloud in Devtron, you need to generate an API key and use it to connect Devtron with your GCP account.

{% stepper %}
{% step %}
Generate the API key using standard [Google Cloud API key documentation](https://cloud.google.com/docs/authentication/api-keys#gcloud).
{% endstep %}

{% step %}
Now go back to the **Edit cluster** modal window, and enter the API key in the **Cloud Provider API Key** field.
{% endstep %}
{% endstepper %}
{% endtab %}

{% tab title="Azure" %}
To enable cost visibility for your Azure clusters in Devtron, you need to allow Devtron to access your billing data securely. This requires two steps:

* Create a custom role in Azure with billing access.
* Create a service principal (an identity) that Devtron can use to fetch cost details.

#### Step 1: Creating Custom Role

{% stepper %}
{% step %}
Open a text editor and copy the following JSON:

{% code title="myrole.json" %}

```json
{
   "Name": "OpenCostRole",
   "IsCustom": true,
   "Description": "Rate Card query role",
   "Actions": [
       "Microsoft.Compute/virtualMachines/vmSizes/read",
       "Microsoft.Resources/subscriptions/locations/read",
       "Microsoft.Resources/providers/read",
       "Microsoft.ContainerService/containerServices/read",
       "Microsoft.Commerce/RateCard/read"
   ],
   "AssignableScopes": [
       "/subscriptions/YOUR_SUBSCRIPTION_ID"
   ]
}
```

{% endcode %}
{% endstep %}

{% step %}
Replace YOUR\_SUBSCRIPTION\_ID with your actual subscription ID. Save the file as `myrole.json`.
{% endstep %}

{% step %}
Run the following command in your terminal:

```bash
az role definition create --verbose --role-definition @myrole.json
```

This creates a role called `OpenCostRole` with just enough access to read pricing information.
{% endstep %}
{% endstepper %}

#### Step 2: Create a Service Principal

{% stepper %}
{% step %}
Run the command below, replacing `YOUR_SUBSCRIPTION_ID` with your subscription ID:

```bash
az ad sp create-for-rbac \
  --name "OpenCostAccess" \
  --role "OpenCostRole" \
  --scope "/subscriptions/YOUR_SUBSCRIPTION_ID" \
  --output json
```

{% endstep %}

{% step %}
You’ll get an output like this:

```json
{
   "appId": "1d9b1532-abe4-4e08-b172-adfa5384da1",
   "displayName": "OpenCostAccess",
   "password": "3DZ8Q~KOE.WgdmmRkBHg3dF1rfrpFhSPk.Hnb4-",
   "tenant": "aee9b2ed-7ecc-4cb2-bfed-60d71c0e957"
}
```

Note this information as you need to enter it in Devtron.
{% endstep %}
{% endstepper %}

#### Step 3: Enter Details in Devtron

Now go back to the **Edit cluster** modal window, and fill the following fields:

| **Field**       | **Value to Provide**                   |
| --------------- | -------------------------------------- |
| Subscription ID | Your Azure subscription ID             |
| App ID          | Value of `appId` from the output       |
| Display Name    | Value of `displayName` from the output |
| Password        | Value of `password` from the output    |
| Tenant          | Value of `tenant` from the output      |
| Billing Account | Optional (fill if available)           |
| Offer ID        | Optional (fill if available)           |
| {% endtab %}    |                                        |

{% tab title="AWS" %}
To enable cost visibility for AWS in Devtron, follow the steps below.

#### Step 1: Set up a Spot Instance Data Feed

{% stepper %}
{% step %}
Create an S3 bucket.
{% endstep %}

{% step %}
Assign full access permissions to the AWS Spot Data Feed service using a bucket policy like:

```json
{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Principal": {
        "Service": "spot.amazonaws.com"
      },
      "Action": "*",
      "Resource": "arn:aws:s3:::devtron-spot-feed/*"
    },
    {
      "Effect": "Allow",
      "Principal": {
        "Service": "spot.amazonaws.com"
      },
      "Action": "*",
      "Resource": "arn:aws:s3:::devtron-spot-feed"
    }
  ]
}
```

{% endstep %}

{% step %}
Run the following command to subscribe to the data feed using the AWS CLI:

```bash
aws ec2 create-spot-datafeed-subscription \
  --bucket devtron-spot-feed --prefix cost
```

Note: You can subscribe to the Spot Data Feed for only one S3 bucket at a time. Running the command again updates the feed to the latest bucket.
{% endstep %}
{% endstepper %}

#### Step 2: Create an IAM Role or Use Access Keys

You can connect Devtron to AWS cost data using either of the following methods:

**IAM Role (Recommended)**

Create a Web Identity IAM role for your EKS/EC2 cluster with permissions to access the Spot Data Feed bucket. Attach the following policy (replace CHANGE-ME with your bucket name):

```json
{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Action": [
        "s3:ListAllMyBuckets",
        "s3:ListBucket",
        "s3:Get*"
      ],
      "Resource": [
        "arn:aws:s3:::CHANGE-ME",
        "arn:aws:s3:::CHANGE-ME/*"
      ],
      "Effect": "Allow"
    }
  ]
}
```

**Access Keys (Alternative)**

Provide an Access Key and Secret Key with permissions `s3:ListBucket` and `s3:GetObject` for the Spot Data Feed bucket.

#### Step 3: Configure in Devtron

Now go back to the **Edit cluster** modal window, and fill the following fields:

| **Field**         | **Description**                                                              |
| ----------------- | ---------------------------------------------------------------------------- |
| Spot Data Bucket  | Name of the S3 bucket storing Spot Instance Data Feed                        |
| Spot Data Region  | AWS region of the Spot Data Feed                                             |
| Spot Data Prefix  | Prefix (if any) used for the Spot Data Feed                                  |
| Project ID        | Your AWS Account ID                                                          |
| Access Key        | (Optional) AWS Access Key with S3 read permissions                           |
| Secret Access Key | (Optional) AWS Secret Access Key with S3 read permissions                    |
| IAM Role          | (Recommended) IAM role ARN assigned to the EKS/EC2 cluster for bucket access |
| {% endtab %}      |                                                                              |
| {% endtabs %}     |                                                                              |

***

## Fetching Prometheus Endpoint

{% hint style="warning" %}
Ensure [GitOps](broken://pages/625dbeae3c08f2bc7de41b6847d88624671a2038) is configured before deploying Prometheus. If not, Prometheus will default to being deployed via Helm.
{% endhint %}

#### Installing Prometheus

{% stepper %}
{% step %}
Go to the **Chart Store** and search for `prometheus`. Use the Prometheus community's `kube-prometheus-stack` chart to deploy Prometheus.

![Figure : Chart Store](https://content.gitbook.com/content/ppUBZub41jtajGM9gUeo/blobs/qwPbXQFX6FPLfjFtH7v6/app2.jpg)
{% endstep %}

{% step %}
After selecting the chart, configure these values as needed before deployment.

```yaml
kube-state-metrics: 
    metricLabelsAllowlist:   
    - pods=[*]
```

```yaml
serviceMonitorSelectorNilUsesHelmValues: false
podMonitorSelectorNilUsesHelmValues: false
```

Search for the above parameters, and update them as shown (or customize as needed).

![Figure a: Prometheus Chart Configuration](https://content.gitbook.com/content/ppUBZub41jtajGM9gUeo/blobs/O2sBkus5gNt2SoEEfwtU/app3.jpg)

![Figure b: Prometheus Chart Configuration (cont.)](https://content.gitbook.com/content/ppUBZub41jtajGM9gUeo/blobs/n2EWWJpwXbMJnskEvc77/app%20metrics%20config.jpg)
{% endstep %}

{% step %}
Enable `upgradeJob` parameter to install CRDs:

Since Helm does not automatically apply CRDs, you need to enable the `upgradeJob` parameter in the Helm chart to ensure CRDs are applied before deploying Prometheus.

In the Prometheus Helm chart settings, locate the `upgradeJob` parameter and set it to `true` if it is `false`.

![Figure : upgradeJob Parameter](https://content.gitbook.com/content/ppUBZub41jtajGM9gUeo/blobs/e5wgOmptiJcl51Irs6aH/app%20new2.jpg)
{% endstep %}

{% step %}
After enabling the parameter, click **Deploy Chart**.
{% endstep %}
{% endstepper %}

{% hint style="warning" %}
Common Pitfall: Prometheus Deployment Timeout due to Failed CRDs

While deploying `kube-prometheus-stack` chart, the deployment status may show as **Timed out**, and some CustomResourceDefinitions (CRDs) may appear as **Failed**.

To solve it, refer to the Troubleshoot Issues section.
{% endhint %}

#### Set Up Prometheus Endpoint

{% hint style="warning" %}
Who Can Perform This Action? Only super admin users can set up Prometheus endpoint in a cluster.
{% endhint %}

{% stepper %}
{% step %}
Once Prometheus is installed, go to its **App Details** and navigate to **Networking** → **Service** in the K8s resources. Expand the Prometheus server service to see the endpoints.
{% endstep %}

{% step %}
Copy the URL of the `kube-prometheus` service as shown in the image below.

![Figure : Prometheus Service](https://content.gitbook.com/content/ppUBZub41jtajGM9gUeo/blobs/8nQTxkN6dZj0kl1mJ3xw/app4.jpg)
{% endstep %}

{% step %}
Now enter the Prometheus endpoint:

* Paste the copied URL into the Prometheus endpoint field, ensuring it includes `http://`
* Click **Update Cluster** to save the changes.
  {% endstep %}
  {% endstepper %}