When it comes to monitoring, keeping track of our CPU, memory, and other infrastructure metrics often isn’t enough. Beyond the common infrastructure-based metrics we expose via tools such as the Node Exporter, we also want to instrument our application to send out metrics related to the app itself. This is done by using the Prometheus client library available to our application’s language.
Once we have the client library set up in our application, we can then define any metrics we wish to collect and write the code to support them. We can also have the Prometheus client record a number of default metrics already included in the library and available to us. Once finished, we’ll have added metrics for our to-do application that track added and removed tasks, as well as response times across our app.
Learning Objectives
Successfully complete this lab by achieving the following learning objectives:
- Add the Prometheus client and track any default metrics.
Move into the
forethoughtdirectory:$ cd forethought
Install the
prom-clientvianpm, Node.js’s package manager:$ npm install prom-client –save
Open the
index.jsfile, where we’ll be adding all of our metrics code:$ $EDITOR index.js
Require the use of the
prom-clientby adding it to our variable list:var express = require(‘express’);
var bodyParser = require(‘body-parser’);
var app = express();
const prom = require(‘prom-client’);With
prombeing the name we’ll use when calling the client library.Enable default metrics scraping:
const collectDefaultMetrics = prom.collectDefaultMetrics;
collectDefaultMetrics({ prefix: ‘forethought_’ });Use Express to create the
/metricsendpoint and call in the Prometheus data:app.get(‘/metrics’, function (req, res) {
res.set(‘Content-Type’, prom.register.contentType);
res.end(prom.register.metrics());
});
- Add the `forethought_tasks_added` metric.
Define the metric:
const tasksadded = new prom.Counter({ name: 'forethought_tasks_added', help: 'The number of items added to the to-do list, total' });Call the new metric in the
addtaskpost function so it increases by one every time the function is called while adding a task:// add a task app.post("/addtask", function(req, res) { var newTask = req.body.newtask; task.push(newTask); res.redirect("/"); tasksadded.inc(); });
- Add the `forethought_tasks_complete` metric.
Define the metric:
const tasksdone = new prom.Counter({ name: 'forethought_tasks_complete', help: 'The number of items completed' });Call the new metric in the
/removetaskpost function so it increases by one every time the function is called:// remove a task app.post("/removetask", function(req, res) { var completeTask = req.body.check; if (typeof completeTask === "string") { complete.push(completeTask); task.splice(task.indexOf(completeTask), 1); } else if (typeof completeTask === "object") { for (var i = 0; i < completeTask.length; i++) { complete.push(completeTask[i]); task.splice(task.indexOf(completeTask[i]), 1); tasksdone.inc() } } res.redirect("/"); });
- Add the `forethought_current_tasks` metric.
Define the metric:
const taskgauge = new prom.Gauge({ name: 'forethought_current_tasks', help: 'Amount of incomplete tasks' });Add an increase to the
/addtaskmethod:// add a task app.post("/addtask", function(req, res) { var newTask = req.body.newtask; task.push(newTask); res.redirect("/"); tasksadded.inc(); taskgauge.inc(); });Add a decrease to the
/removetaskmethod:// remove a task app.post("/removetask", function(req, res) { var completeTask = req.body.check; if (typeof completeTask === "string") { complete.push(completeTask); task.splice(task.indexOf(completeTask), 1); } else if (typeof completeTask === "object") { for (var i = 0; i < completeTask.length; i++) { complete.push(completeTask[i]); task.splice(task.indexOf(completeTask[i]), 1); tasksdone.inc(); taskgauge.dec(); } } res.redirect("/"); });Save and exit.
- Add the `forethought_response_time_summary` metric.
Add the
response-timemodule:$ npm install response-time --saveReopen the
index.jsfile.Define the metric:
const responsetimesumm = new prom.Summary ({ name: 'forethought_response_time_summary', help: 'Latency in percentiles', });Add
response-time:var responseTime = require('response-time');Write the code to retrieve the response time:
app.use(responseTime(function (req, res, time) { responsetimesumm.observe(time); }));
- Add the `forethought_response_time_histogram` metric.
Define the metric:
const responsetimehist = new prom.Histogram ({ name: 'forethought_response_time_histogram', help: 'Latency in history form', buckets: [0.1, 0.25, 0.5, 1, 2.5, 5, 10] });Call the histogram in the existing
response-timecode:app.use(responseTime(function (req, res, time) { responsetimesumm.observe(time); responsetimehist.observe(time); }));
- Reimage the application and redeploy the container; add to Prometheus.
Stop the current Docker container for our application:
$ docker stop ft-appRemove the container:
$ docker rm ft-appRemove the image:
$ docker image rm forethoughtRebuild the image:
$ docker build -t forethought .Deploy the new container:
$ docker run --name ft-app -p 80:8080 -d forethoughtSwitch to the
monitoringserver.Add our application endpoint to Prometheus (replacing
PRIVATE_IPwith the private IP for the Application server listed on the lab page):$ sudo $EDITOR /etc/prometheus/prometheus.yml - job_name: 'forethought' static_configs: - targets: ['PRIVATE_IP:80']Save and exit.
Restart Prometheus:
$ sudo systemctl restart prometheus
