• Author: Chris Bailey

  • Updated: 7/15/2019

  • Duration: 1h 48m

  • Skill Level: Intermediate

  • Course URL: https://www.lynda.com/Node-js-tutorials/Cloud-Native-Development-Node-js-Docker-Kubernetes/808675-2.html

  • Description:

    While the vast majority of Node.js apps are deployed and run in the cloud, few leverage all that modern cloud computing platforms have to offer. If you’re looking to take the next step in your cloud computing journey, then this course is for you. Join Chris Bailey as he shows developers how to go from merely hosting apps in the cloud to building and deploying cloud native apps. Chris begins by providing an overview of what it means to be cloud native and create Node.js apps that exploit the cloud’s features. He then steps through how to take an existing Node.js app and package it with Docker, deploy the app to Kubernetes, and enhance it with cloud native capabilities, including support for self-healing and metrics.

    Topics include:

    • Using Node.js in the cloud
    • Creating a Node.js app
    • Building a production Dockerfile
    • Deploying an app to Kubernetes using Helm
    • Adding self-healing capabilities
    • Building custom charts and graphs
    • Adding support for metrics and request tracking

Table of Content


The power of cloud native

What you should know

1. What Is Cloud Native?

What does “cloud native” mean?

Node.js in the cloud

Cloud native Node.js

2. Creating the App

Creating your Node.js app

Add a Dockerfile

  • dockerfile template: https://github.com/CloudNativeJS/docker

  • install brew on macOS, and brew install wget

  • cat Dockerfile
    • FROM node:10 (cat /etc/os-release: debian 9, uname -r: the container shares the host kernel ), more at GitHub: nodejs/docker-node

    • [email protected]:/app$ cat /etc/os-release  
      PRETTY_NAME="Debian GNU/Linux 9 (stretch)"
      NAME="Debian GNU/Linux"
      VERSION="9 (stretch)"
      [email protected]:/app$ uname -a
      Linux d4cb68ddd497 5.2.11-1-MANJARO #1 SMP PREEMPT Thu Aug 29 07:41:24 UTC 2019 x86_64 GNU/Linux
      [email protected]:/app$ uname -r
    • COPY package*.json ./ ; RUN npm install --production; COPY . /app : copy the source code as late as possible, so that when running docker build only the last few steps(layers) are executed
  • add .dockerignore

  • docker build -t nodeserver -f Dockerfile . with tag nodeserver

  • docker run -i -p 3000:3000 -t nodeserver

    • -i, --interactive: Keep STDIN open even if not attached
    • with port forwarding
    • -t, --tty: Allocate a pseudo-TTY
  • Run the image in the background and attach it with bin/bash

  • # -d, --detach: Run container in background and print container ID
    # --rm  Automatically remove the container when it exits
    [[email protected] ~]$ docker run --rm -d nodeserver 
    [[email protected] ~]$ docker ps
    CONTAINER ID        IMAGE               COMMAND                  CREATED             STATUS              PORTS               NAMES
    d4cb68ddd497        nodeserver          "docker-entrypoint.s…"   5 seconds ago       Up 3 seconds        3000/tcp            amazing_pascal
    [[email protected] ~]$ docker exec -it amazing_pascal /bin/bash
    # note here: `npm` has PID 1 rather than `init`
    [email protected]:/app$ ps aux 
    node          1  2.5  0.6 734156 42592 ?        Ssl  08:52   0:00 npm
    node         18  0.0  0.0   4280   712 ?        S    08:52   0:00 sh -c node ./bin/www
    node         19  0.7  0.6 565016 38288 ?        Sl   08:52   0:00 node ./bin/www
    node         26  5.7  0.0  18200  3240 pts/0    Ss   08:53   0:00 /bin/bash
    node         33  0.0  0.0  36632  2764 pts/0    R+   08:53   0:00 ps aux

Build a dev and debug Dockerfile

  • Dockerfile-run
  • Dockerfile-toolssee https://github.com/CloudNativeJS/docker#using-dockerfile-tools
    • ENV NODE_HEAPDUMP_OPTIONS nosignal, see: https://github.com/RuntimeTools/appmetrics/issues/517
    • Development mode: mount source code and node_modules into the container, and start the node.js process
    • Debug Mode: start the node.js process with inspect, and also port-forwarding the debug port
  • Dockerfile: production use

Build a production Dockerfile

Tagging and version control

3. Deploying to Kubernetes

Docker vs. Kubernetes

Add a Helm chart

Deploy to Kubernetes

Deploy multiple instances

4. Adding Support for Health

Liveness and readiness endpoints

Add a liveness check

Add a readiness check

See it in Kubernetes

5. Add Support for Metrics

Introduction to Prometheus

Deploy Prometheus to Kubernetes

Deploy Grafana to Kubernetes

Build charts for your applications

6. Add Support for Request Tracking

Introduction to OpenTracing

Adding OpenTracing to the app

Deploy OpenTracing to Kubernetes


Next steps