express-prom-bundle

What is express-prom-bundle?

The express-prom-bundle package is a middleware for Express.js that integrates Prometheus metrics collection and reporting. It allows you to easily monitor the performance and health of your Express applications by exposing various metrics that Prometheus can scrape.

What are express-prom-bundle's main functionalities?

Basic Setup

This code demonstrates the basic setup of express-prom-bundle in an Express application. It initializes the middleware and includes it in the app, which will start collecting and exposing metrics.

const express = require('express');
const promBundle = require('express-prom-bundle');

const app = express();
const metricsMiddleware = promBundle({ includeMethod: true });

app.use(metricsMiddleware);

app.get('/', (req, res) => {
  res.send('Hello World!');
});

app.listen(3000, () => {
  console.log('Server is running on port 3000');
});

Custom Metrics

This code demonstrates how to create and use custom metrics with express-prom-bundle. It shows how to define a custom counter and increment it in a route handler.

const express = require('express');
const promBundle = require('express-prom-bundle');
const promClient = require('prom-client');

const app = express();
const metricsMiddleware = promBundle({ includeMethod: true });

app.use(metricsMiddleware);

const customCounter = new promClient.Counter({
  name: 'custom_counter',
  help: 'Example of a custom counter',
});

app.get('/increment', (req, res) => {
  customCounter.inc();
  res.send('Counter incremented');
});

app.listen(3000, () => {
  console.log('Server is running on port 3000');
});

Custom Metrics Endpoint

This code demonstrates how to customize the endpoint where Prometheus metrics are exposed. By setting the `metricsPath` option, you can change the default `/metrics` endpoint to a custom path.

const express = require('express');
const promBundle = require('express-prom-bundle');

const app = express();
const metricsMiddleware = promBundle({ includeMethod: true, metricsPath: '/custom-metrics' });

app.use(metricsMiddleware);

app.get('/', (req, res) => {
  res.send('Hello World!');
});

app.listen(3000, () => {
  console.log('Server is running on port 3000');
});

Other packages similar to express-prom-bundle

prom-client

prom-client is a Prometheus client for Node.js that allows you to create and expose custom metrics. Unlike express-prom-bundle, it does not provide middleware for Express, so you need to manually integrate it into your application.

express-status-monitor

express-status-monitor is a simple, self-hosted module based on Socket.io and Chart.js to report realtime server metrics for Express-based applications. It provides a web-based dashboard for monitoring but does not integrate with Prometheus.

README

express prometheus bundle

express middleware with popular prometheus metrics in one bundle.

Internally it uses prom-client. See: https://github.com/siimon/prom-client

Included metrics:

• up: normally is just 1
• nodejs_memory_heap_total_bytes and nodejs_memory_heap_used_bytes
• http_request_seconds: http latency histogram labeled with status_code

Install

npm install express-prom-bundle

Usage

You MUST call app.use(metricsMiddleware) before the use-ing your middleware, otherwise those won't count in http_request_seconds histogram

const promBundle = require("express-prom-bundle"),
const metricsMiddleware = promBundle({/* options */ });

app.use(metricsMiddleware);
app.use(/* your middleware */);
app.listen(3000);

• call your endpoints
• see your metrics here: http://localhost:3000/metrics

Options

• prefix: prefix added to every metric name
• whitelist, blacklist: array of strings or regexp specifying which metrics to include/exclude
• buckets: buckets used for http_request_seconds histogram

Example

setup std. metrics but exclude up-metric:

"use strict";

const express = require("express"),
    app = express(),
    promBundle = require("express-prom-bundle");

app.use(promBundle({
    prefix: "demo_app:something"
}));

app.get("/hello", (req, res) => res.send("ok"));

app.listen(3000);

See an advanced example on github

License

MIT