pydantic/monty · error · Error

Monty telemetry is already configured

Error message

Monty telemetry is already configured

What it means

`instrumentTelemetry` installs OpenTelemetry hooks (tracer/meter/logger) into the monty native pool, and installation is process-wide and single-shot. If telemetry was already configured — by a prior `instrumentTelemetry` call or the native side registering an owner — the function throws this Error instead of silently overwriting the active configuration.

Source

Thrown at crates/monty-js/ts/telemetry.ts:173

      return callback()
    })
  } catch (error) {
    if (invoked) {
      throw error
    }
    return callback()
  }
}

/**
 * Instrument Monty with standard OpenTelemetry components.
 *
 * Installation is process-wide and must happen before creating a pool. Each
 * signal is independently optional.
 */
export function instrumentTelemetry(value: TelemetryComponents): void {
  if (owner !== undefined) {
    throw new Error('Monty telemetry is already configured')
  }
  if (value.tracer === undefined && value.meter === undefined && value.logger === undefined) {
    throw new Error('at least one OpenTelemetry component is required')
  }
  activate(directOwner, value)
}

/** Wait until telemetry queued by the native pool has reached JavaScript. */
export async function flushTelemetry(): Promise<void> {
  await flushNativeTelemetry()
  if (!acceptingTelemetry) {
    components = undefined
    spans.clear()
    instruments.clear()
  }
}

/**

View on GitHub (pinned to adc986b362)

Solutions

1. Call `instrumentTelemetry` exactly once per process, ideally at startup before creating any pool.
2. Guard the call with an `isTelemetryConfigured()`-style check or your own module-level boolean.
3. In tests, install telemetry once in a global setup file instead of per-test; if state is corrupted, reset module state or spawn a fresh process/worker.

Example fix

// before
instrumentTelemetry({ tracer })
// ...later, same process
instrumentTelemetry({ tracer }) // throws

// after
if (!telemetryInstalled) {
  telemetryInstalled = true
  instrumentTelemetry({ tracer })
}

let telemetryInstalled = false
function setupTelemetryOnce(components: TelemetryComponents): void {
  if (!telemetryInstalled) {
    instrumentTelemetry(components)
    telemetryInstalled = true
  }
}

try {
  instrumentTelemetry({ tracer })
} catch (err) {
  if (err instanceof Error && err.message === 'Monty telemetry is already configured') {
    // another module installed telemetry; keep existing config
  } else {
    throw err
  }
}

When it happens

Trigger: Calling `instrumentTelemetry(...)` more than once in the same process; calling it after a pool or module load already activated telemetry; duplicate setup in tests where module state persists across cases.

Common situations: Test suites importing the module in multiple files with a shared global `beforeAll` that installs telemetry each time; hot-reload or double-import of the telemetry module in bundlers; library code and app code both attempting to install hooks.

Related errors

• at least one OpenTelemetry component is required
• ClassInstance expects an object instance
• notCallableMessage(method)
• ClassInstance expects an instance of a class, not a null-pro
• classType does not match the instance's class

AI-assisted analysis of pydantic/monty@adc986b362 (2026-09-13). Data as JSON: /api/errors/1929ba32bf15c44b. Report an issue: GitHub.