# Grafana Forwarder (via OTel)

To build a forwarder from Apica to Grafana Labs using OpenTelemetry (OTel), you are essentially configuring Apica's Flow or Pipeline engine to act as a routing and transformation layer that exports data via the OTLP protocol. Since Grafana Cloud provides a native OTLP intake, this is a clean, standard-based integration.

### 1. Prerequisites from Grafana Labs

Before configuring Apica, you must obtain your OTLP connection details from your Grafana Cloud portal:

1. Log in to your **Grafana Cloud Portal**.
2. Navigate to **OpenTelemetry** or look for the OTLP Endpoint tile.
3. Note the following values:
   1. OTLP Endpoint: (e.g., <https://otlp-gateway-prod-us-central-0.grafana.net/otlp>)
   2. Username/Instance ID: Found on the "OpenTelemetry" or "Details" page.
   3. Cloud Access API Token/Password: Your generated Cloud Access Policy token with metrics:write, logs:write, and traces:write permissions.

### 2. Apica Configuration Strategy: The OTLP Forwarder

In the Apica UI (Ascent/Flow), you will create a Forwarder (or Target Destination) that uses the OpenTelemetry (OTLP) exporter.&#x20;

#### Step A: Define the Target

In the Apica Pipeline configuration, you define the Grafana endpoint. Because Grafana Cloud requires Basic Authentication, your OTLP headers must be encoded.

* **Endpoint**: https\://\<otlp-endpoint>
* **Protocol**: http/protobuf (recommended over gRPC for standard web-hook style forwarding)
* **Headers**: Your encoded cloud access policy token. Example:

<pre data-overflow="wrap"><code><strong>Authorization=Basic MTIzNDU6Z2xjX0FCQ0RFRkFHRkk= 
</strong>Note: You can encode using this command on linux: echo -n "12345:glc_ABCDEFGHIJ1234567890" | base64
# produces e.g. MTIzNDU6Z2xjX0FCQ0RFRkFHRkk=
</code></pre>

#### Step B: Pipeline Setup (The "Forwarder")

In the Apica "Flow" or "Pipeline" section, you will create a rule to route specific filtered data to this new destination.

1. **Select Source**: Choose the incoming OTel data or system logs you want to forward.
2. **Apply Filters**: (Optional) Use Apica's filtering to drop noisy metrics or sensitive log data to save on Grafana Cloud costs.
3. **Select Destination**: Choose the OTLP/Grafana Labs target created in **Step A**.

<table data-header-hidden><thead><tr><th width="229.87109375"></th><th></th></tr></thead><tbody><tr><td><strong>Field</strong></td><td><strong>Value / Setting</strong></td></tr><tr><td>Exporter Type</td><td><code>OpenTelemetry (OTLP)</code></td></tr><tr><td>Protocol</td><td><code>http/protobuf</code></td></tr><tr><td>Endpoint URL</td><td><code>https://&#x3C;your-otlp-endpoint-url></code> (e.g., .../otlp)</td></tr><tr><td>Custom Header Key</td><td><code>Authorization</code></td></tr><tr><td>Custom Header Value</td><td><code>Basic &#x3C;Your-Base64-String></code></td></tr><tr><td>Encoding</td><td><code>JSON</code> (for logs/traces) or <code>Protobuf</code> (for metrics)</td></tr><tr><td>Compression</td><td><code>gzip</code> (recommended for cost/bandwidth reduction)</td></tr></tbody></table>

#### Otel Forwarder:

<figure><img src="https://2948796384-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-LmzGprckLqwd5v6bs6m%2Fuploads%2FZ3B54RwcDdtjOABDQ3iS%2Fimage.png?alt=media&#x26;token=d4ff48da-db2c-4da7-a081-74fd01e51880" alt=""><figcaption></figcaption></figure>

#### Advanced: Mapping Labels and Jobs

Grafana Cloud (via Loki and Prometheus) relies heavily on labels like `job` and `instance`. You can use Apica's Transformation Layer to ensure these are set correctly before the data leaves:

SQL

```
# Map Apica attributes to Grafana-friendly labels
set(resource.attributes["service.name"], "apica-flow")
set(resource.attributes["service.namespace"], "production")
set(resource.attributes["deployment.environment"], "us-east-1")

# Ensure the job label is promoted (useful for Loki logs)
set(attributes["job"], "apica-flow-forwarder")
```

#### Critical Ingestion Limits

When forwarding to Grafana Cloud, stay within these OTLP constraints to avoid "429 Too Many Requests" or dropped spans:

* Metric Name Length: Max 2048 bytes.
* Resource Attributes: Max 40 per signal.
* Log Line Size: Max 256 KB.
* Trace Volume: Max 5 MB per individual trace.