> For the complete documentation index, see [llms.txt](https://developer.harness.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.harness.io/feature-management-experimentation/use-fme/config-management/setup.md).

# Setup

Set up Config Management in Harness FME.

Before you can retrieve configuration values through a Configs SDK, create and configure your configs in Harness FME. A config defines the structure, variations, and targeting behavior that your application consumes at runtime.

{% hint style="info" %}
A config is not returned by the SDK until the target environment has been initialized and the config has a value configured for that environment.
{% endhint %}

You need the appropriate Config Management permissions to create and manage configs. For more information, go to [Feature Management and Experimentation permissions](https://developer.harness.io/harness-platform/use-harness-platform/platform-access-control/permissions-reference#feature-management-and-experimentation).

Setting up Config Management consists of the following steps:

1. **Create a config**: Define the config metadata, including its name, identifier, traffic type, owners, and initial variations.
2. **Define the schema and variations**: Create a JSON schema and define one or more variations that your application can receive.
3. **Configure environment values**: Initialize each environment and define the JSON value served by each variation.
4. **Configure targeting**: Define targeting rules or individual targets to determine which variation is returned for each SDK evaluation.
5. **Promote changes across environments**: After validating changes, promote configuration values and targeting rules from one environment to another.
6. **Monitor evaluations**: Use Live Tail to verify configuration evaluations and troubleshoot runtime behavior in real time.
7. **Retrieve config values from your application**: Integrate a Configs SDK to evaluate and consume configuration values at runtime.

## Create a config

Harness Configs provide general-purpose runtime configuration for application settings, operational parameters, business rules, and other structured configuration.

To create a config:

1. In Harness FME, navigate to **Configs**.
2. Select **Create config**.

   ![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-71958551b873e73ce55baa5bdd42b0e6ddf3acc8%2Fconfigs-3.png?alt=media)
3. In the **Summary** section, enter the following details:

   | Field            | Description                                                                                      |
   | ---------------- | ------------------------------------------------------------------------------------------------ |
   | **Name**         | The display name for your config.                                                                |
   | **Identifier**   | The unique identifier used to reference the config. This value cannot be changed after creation. |
   | **Description**  | Optional information that describes the purpose of the config.                                   |
   | **Traffic Type** | The traffic type used for targeting and evaluation.                                              |
   | **Tags**         | Optional metadata used to organize and filter configs.                                           |
   | **Owners**       | The users or teams responsible for managing the config.                                          |
4. In the **Initial variations** section, create one or more variations that your application can receive at runtime. Enter a name for each variation and click **+ Add new variation** to create additional variations.

   ![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-13c6718e96c45f55f2ea028eb6890e282d4bc8c9%2Fconfigs-1.png?alt=media)
5. When you're finished, click **Create**. Each config represents a reusable runtime behavior that can be independently versioned, targeted, and promoted across environments.

### Manage config details

After you create a config, you are redirected to the **Config Details** page.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-6a5f1b7abf6bffde0b3d96d824eb184271e8763b%2Fconfigs-4.png?alt=media)

A config is composed of four components that define how runtime behavior is managed:

```mermaid
flowchart LR

    CONFIG["Config"]:::root --> META["Metadata"]:::section
    META --> SCHEMA["Schema &<br/>Variations"]:::section
    SCHEMA --> ENV["Environments"]:::section
    ENV --> VERSION["Version<br/>History"]:::section

    classDef root fill:#8110B5,stroke:#8110B5,color:#fff;
    classDef section fill:#9b5de5,stroke:#9b5de5,color:#fff;
```

On the **Config Details** page, you can:

* View the config metadata, including its identifier, traffic type, tags, and owners.
* Edit the config name, description, tags, and owners.
* Manage the JSON schema and variations.
* Configure environment-specific values and targeting.
* Promote configuration changes between environments.

### Edit schema and variations

The JSON schema validates the structure of every variation. A variation is a JSON value, and each variation must conform to the schema before you can save it. This helps ensure that configuration values delivered through the [Configs SDK](https://developer.harness.io/feature-management-experimentation/new-to-fme/sdks-and-customer-deployed-components/server-side-configs-sdks.md) conform to the expected format.

```json
// Example JSON Schema
{
  "type": "object",
  "properties": {
    "bannerMessage": {
      "type": "string"
    },
    "freeShippingThreshold": {
      "type": "number"
    },
    "discountPercent": {
      "type": "number"
    },
    "showCountdownTimer": {
      "type": "boolean"
    },
    "maxCartItems": {
      "type": "integer"
    }
  },
  "required": [
    "bannerMessage",
    "freeShippingThreshold",
    "discountPercent",
    "showCountdownTimer",
    "maxCartItems"
  ]
}
```

Click **Edit schema & variations** to update the JSON schema or modify the variations associated with the config.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-6dc4bdb946f3b8916a1d3939a74809fd4ca799cb%2Fconfigs-2.png?alt=media)

You can also add, rename, or remove variations that your application can receive at runtime.

After making your changes, click **Save** to update the config.

## Configure environments

Each [Harness FME environment](https://developer.harness.io/feature-management-experimentation/management-and-administration/environments.md) can be configured independently using the **Values**, **Targeting**, **Live Tail**, and **Audit logs** tabs.

* **Values**: Configure JSON values served for each variation.
* **Targeting**: Configure targeting rules and individual targets.
* **Live Tail**: Monitor runtime configuration evaluations.
* **Audit Logs**: Track configuration changes and user activity.

### Configure environment values

Each environment maintains its own runtime configuration. This allows teams to validate configuration changes in lower environments before promoting them to production.

For example, a config can serve one set of values in a development environment while serving different values in production, all from the same config definition. Before a config can be evaluated by the SDK in an environment, you must initialize that environment.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-6a5f1b7abf6bffde0b3d96d824eb184271e8763b%2Fconfigs-4.png?alt=media)

### Review changes

Changes to environment values and targeting rules are reviewed before they are saved. After updating the value or targeting rule, click **Review Changes**.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-e42ed61b98166665c166b28c871f98c7560ab355%2Fconfigs-6.png?alt=media)

Review the pending changes and click **Save** to apply them to the selected environment.

Changes to both environment values and targeting rules are recorded in the audit logs, providing a history of every configuration update. Changes can be submitted for approval before they are published.

{% tabs %}
{% tab title="Values" %}
To configure environment values:

1. In the **Environments** section, select an environment.
2. On the **Values** tab, click **Initiate environment**.

Clicking **Initiate environment** creates an environment-specific configuration layer where you can define the values served by each variation. You can start from scratch or initiate the environment with a copy of the values from another environment.

For each variation, update the JSON value that should be returned when the variation is evaluated in this environment.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-5e2662a274e214937ebc49a29e5e636f4a3e552d%2Fconfigs-5.png?alt=media)

Then click **Review Changes** to review the pending updates before saving them.
{% endtab %}

{% tab title="Targeting" %}
The **Targeting** tab determines which variation is returned for each SDK evaluation. Targeting must be initialized separately for each environment before you can define individual targets or targeting rules.

To configure targeting:

1. Select an environment.
2. Open the **Targeting** tab.
3. Click **Initiate environment**.

   ![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-d7f703cf507aa2405752eba3d25dc232b417bfdc%2Fconfigs-7.png?alt=media)

After initialization, you can configure both individual targets and targeting rules.

* Use **Individual targets** to serve a specific variation to individual users or segments. Individual targets take precedence over targeting rules, allowing you to create explicit overrides for specific users or segments.
* **Targeting rules** allow you to serve different variations based on user attributes or segment membership.

To create a target, click **New individual target**.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-48e2f5bf15d9ce0095f8fbb8656c0895f7ae7b53%2Fconfigs-8.png?alt=media)

Each individual target lets you specify:

* The variation to serve.
* One or more users.
* One or more segments.

You can create multiple individual targets to override the default targeting behavior for specific audiences.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-a196f0d04c4bf0d1ac92cfe051f53030b13f39ca%2Fconfigs-9.png?alt=media)

Click **Add attribute based targeting rules** and select **Add condition**.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-9661b024c2cc3f00cb5f623a56e8a767a0749fd9%2Fconfigs-10.png?alt=media)

Each targeting rule follows an `IF`/`THEN` structure:

* **IF**: Define one or more conditions that determine when the rule matches.
* **THEN**: Select the variation to serve and optionally distribute traffic across multiple variations.

Conditions can target users, segments, lists, and other supported attributes, allowing you to create progressively more granular rollout strategies.

Within a targeting rule, you can optionally limit the percentage of traffic evaluated against that rule. Click **Limit exposure** to add the **Limit exposure** section, then specify the percentage of traffic that should be evaluated using the targeting rule.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-58adb29bb7af80ff5cfd655d718aeff30e8a800c%2Fconfigs-11.png?alt=media)

Requests that fall outside the configured percentage receive the default variation. This allows you to gradually introduce new targeting logic before exposing it to all eligible traffic.
{% endtab %}

{% tab title="Evaluations" %}
The **Live Tail** tab provides real-time visibility into configuration evaluations performed by your applications. You can use Live Tail to verify targeting behavior, troubleshoot SDK evaluations, and confirm that the expected variation is being served.

To start querying evaluation data in Live Tail, first select the type of runtime data you want to view from the **Data type** menu: **Impressions** or **Events**.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-68b4778af15066066fce023fc99ed55778d7c836%2Fconfigs-14.png?alt=media)

Live Tail streams configuration evaluations as they occur, allowing you to verify targeting behavior and troubleshoot SDK evaluations in real time. To narrow your results, click **Add query criteria** to apply additional filters.

When you're ready, click **Query** to retrieve matching evaluations. You can also pause incoming data, clear the current results, and download the displayed data.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-b61ca4728c6c3ad2593eba40ed5fec5081587d9d%2Fconfigs-15.png?alt=media)

Use the **Columns** menu to show or hide individual fields.

![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-7a72eb58e4be74309bb5f6da58ee6009e4113165%2Fconfigs-13.png?alt=media)

The **Impressions** view includes the following columns:

| Column              | Description                                                |
| ------------------- | ---------------------------------------------------------- |
| Config              | The config evaluated by the SDK.                           |
| Key                 | The identifier used for the evaluation.                    |
| Treatment           | The variation or treatment returned during evaluation.     |
| Targeting rule      | The targeting rule that determined the returned variation. |
| Timestamp           | The time the evaluation occurred.                          |
| SDK                 | The SDK used to evaluate the configuration.                |
| SDK Version         | The version of the SDK performing the evaluation.          |
| Bucketing Key       | The key used to determine the evaluation bucket.           |
| Machine IP          | The IP address of the machine performing the evaluation.   |
| Traffic Type Name   | The traffic type associated with the evaluation.           |
| Version             | The configuration version evaluated.                       |
| Reception Timestamp | The time the evaluation data was received.                 |
| Environment Name    | The environment where the evaluation occurred.             |

The **Events** view includes the following columns:

| Column            | Description                                            |
| ----------------- | ------------------------------------------------------ |
| Key               | The identifier associated with the event.              |
| Traffic Type Name | The traffic type associated with the event.            |
| Event Type ID     | The event type that was tracked.                       |
| Source            | The origin of the event.                               |
| Value             | The event value, if provided.                          |
| Timestamp         | The time the event occurred.                           |
| SDK               | The SDK that recorded the event.                       |
| SDK Version       | The version of the SDK.                                |
| Machine IP        | The IP address of the client that generated the event. |
| Environment       | The environment where the event occurred.              |
| {% endtab %}      |                                                        |

{% tab title="Audit logs" %}
The **Audit logs** tab shows the history of changes made to a config within the selected environment, including who made each change and whether it was published, is awaiting action, or did not go through.

On the left, use the search bar to **Search by editors**, and click **+ Add filter** to narrow the list further by:

* **Change type**: `Created`, `Updated rules`, `Updated tags`, `Updated owners`.
* **Change status**: `Published`, `Requested`, `Withdrawn`, `Rejected`, `Failed`.

On the right, use the **Columns** menu to show or hide fields in the Audit logs table:

| Column        | Description                                                                                                                  |
| ------------- | ---------------------------------------------------------------------------------------------------------------------------- |
| Title         | A short description of the change that was made.                                                                             |
| Change type   | The kind of change made to the config in this environment (`Created`, `Updated rules`, `Updated tags`, or `Updated owners`). |
| Edited by     | The user who made the change.                                                                                                |
| Change status | The current status of the change (`Published`, `Requested`, `Withdrawn`, `Rejected`, or `Failed`).                           |
| Time          | The date and time the change was made.                                                                                       |
| {% endtab %}  |                                                                                                                              |
| {% endtabs %} |                                                                                                                              |

## Promote environments

After validating configuration changes in one environment, you can promote them to another environment.

Promotion allows teams to move runtime configuration through development, staging, and production without manually recreating values or targeting rules.

To promote a configuration:

1. Click **Promote environment**.
2. Select the source environment.
3. Select the target environment.
4. Optionally, select **Also copy targeting rules** to include targeting configuration during promotion.
5. Review the environment differences in the **Diff preview**.
6. Optionally, provide a title and comments describing the promotion.
7. Click **Promote**.

   ![](https://1473625385-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FrT2RGHAjtBTaObzchfpb%2Fuploads%2Fgit-blob-78e7d0b7612bf9b3d061ad411fb962e66c5bbc96%2Fconfigs-12.png?alt=media)

## Next steps

After creating and configuring a config, integrate your application using a [server-side Configs SDK](https://developer.harness.io/feature-management-experimentation/new-to-fme/sdks-and-customer-deployed-components/server-side-configs-sdks.md) or the [client-side JavaScript Configs SDK](https://developer.harness.io/feature-management-experimentation/new-to-fme/sdks-and-customer-deployed-components/client-side-configs-sdks.md) to retrieve configuration values at runtime.

For information about the permissions required to manage Configs, go to [Feature Management and Experimentation permissions](https://developer.harness.io/harness-platform/use-harness-platform/platform-access-control/permissions-reference#feature-management-and-experimentation).
