---
url: https://docs.fluxify.rest/concepts/telemetry-configuration.md
description: Configure project-wide destinations for logs, traces, and metrics.
---

# Telemetry Configuration

Telemetry lets you observe routes after they are deployed. Fluxify workers can send **logs**, **traces**, and **metrics** to the destinations configured for a project.

Open **Project settings → Telemetry** to choose an observability integration for each signal.

![Project settings Telemetry page with separate logs, traces, and metrics destinations](/project-settings/telemetry-configuration.png)

## Configure each signal

| Signal | What it tells you | When it is sent |
| --- | --- | --- |
| **Logs** | Messages written by logging blocks and the JavaScript `logger` API. | When a route emits a log entry. |
| **Traces** | The work performed during a route run, useful for following latency and failures. | Only for routes where tracing is enabled. |
| **Metrics** | Route request counts and durations. | Alongside recorded traced runs. |

You can use a different destination for each signal, or leave a signal unset. When no logs destination is selected, logs fall back to the server console. When traces or metrics are unset, Fluxify does not export them.

## Tracing and cost

Tracing is enabled per route because detailed execution data has a cost. Enable it for routes you are investigating or monitoring closely, rather than treating it as a default for every high-volume endpoint.

Workers publish telemetry away from the request path, so exporting it does not make the caller wait for an observability service. The telemetry pipeline uses tail-based selection: failed runs are retained, while successful runs are sampled. This preserves the failure signals that matter most while keeping normal-traffic volume under control.

Metrics are produced from the same recorded runs, so they reflect the routes selected for tracing rather than every request.

## Logging in a route

Use **Console Log** or **Cloud Logs** blocks for a visual route, or call the JavaScript logger when a message belongs next to code:

```javascript
logger.logInfo("Customer lookup completed", { customerId: input.id });
logger.logWarn("Customer service returned a fallback result");
logger.logError("Customer lookup failed", error);
```

For the full JavaScript logging API, see the [JavaScript API Reference](../scripting/javascript-api.md#logging).

## Related pages

* [Observability integrations](../integrations/observability.md)
* [Console Log block](../blocks/console-log.md)
* [Cloud Logs block](../blocks/cloud-logs.md)