Logging

Lyn logs with pino. It writes one line per request, and the stack frames of unexpected errors. The error message itself is left out: a database error's message contains the SQL and the values that were bound to it.

Settings

Logging is configured with two environment variables:

Variable	Effect
NODE_ENV	production: one JSON object per line, for log collectors. Anything else: readable, colored output.
LOG_LEVEL	fatal, error, warn, info (default), debug, trace or silent

An invalid LOG_LEVEL stops the app with a LynEnvError.

Request logs

Each request produces one line with its requestId, method, path, status and durationMs:

{"level":30,"requestId":"43297dd3-…","method":"GET","path":"/users","status":200,"durationMs":0.95,"msg":"[Lyn] GET /users 200 0.95ms"}

The level follows the status: info below 400, warn for 4xx, error for 5xx.

Lyn's own messages start with [Lyn].

To also log each SQL query, see Query logs.

Request ids

Each request gets an id, sent back in the X-Request-Id response header and added to every log line of that request. When a client reports a problem, the id leads you to its logs and stack trace.

If the request already has an X-Request-Id header (set by a proxy, for example), Lyn reuses it, as long as it is 1 to 128 characters among A-Z a-z 0-9 . _ -. Otherwise it generates a UUID.

Your own logs

Use the exported logger. It shares Lyn's level and output format:

import { logger } from "@rouret/lyn";

.post("/orders", ({ body }) => {
  logger.info({ orderId: body.id }, "order received");
  return { ok: true };
})

Not supported yet

• Sending logs to a custom destination.