Skip to content

Latest commit

 

History

History
372 lines (284 loc) · 15 KB

README.md

File metadata and controls

372 lines (284 loc) · 15 KB
Raw

NodeJS 20 minimal container image

This container image includes Node.JS 20 on top of a minimal base image for your Node.JS 20 applications. This image is designed to be used with the full-sized s2i-enabled Node.JS 20 image to build the application. The image can be used as a standalone s2i-enabled image as well, but compared to the full-sized Node.JS 20 image it will be missing many build-time dependencies. Users can choose between RHEL, CentOS and Fedora based images. The RHEL images are available in the Red Hat Container Catalog, the CentOS Stream images are available in the Quay.io, and the Fedora images are available in Quay.io. The resulting image can be run using podman.

Note: while the examples in this README are calling podman, you can replace any such calls by docker with the same arguments

Description

Node.js 20 available as a minimal container is a base platform for running various Node.js 20 applications and frameworks. Node.js is a platform built on Chrome's JavaScript runtime for easily building fast, scalable network applications. Node.js uses an event-driven, non-blocking I/O model that makes it lightweight and efficient, perfect for data-intensive real-time applications that run across distributed devices.

Usage in OpenShift

In this example, we will assume that you are using the ubi8/nodejs-20 image, available via nodejs:20-ubi8 imagestream tag in Openshift to build the application, as well as the ubi8/nodejs-20-minimal image, available via nodejs:20-ubi8-minimal image stream for running the resulting application.

With these two images we can create a chained build in Openshift using two BuildConfigs:

The first BuildConfig defines and builds the builder image, using the source-to-image strategy, and pushes the result into the nodejs-builder-image imagestream.

apiVersion: build.openshift.io/v1
kind: BuildConfig
metadata:
  name: nodejs-builder-image
spec:
  output:
    to:
      kind: ImageStreamTag
      name: nodejs-builder-image:latest
  source:
    git:
      uri: https://github.com/sclorg/nodejs-ex.git
  strategy:
    sourceStrategy:
      from:
        kind: ImageStreamTag
        name: nodejs:20-ubi8
        namespace: openshift

The second BuildConfig takes the resulting image from the nodejs-builder-image imagestream, copies the application source (including build artifacts) from the image and creates a new runtime image on top of the nodejs minimal image, with the application copied in and prepared to run. The resulting runtime image is then pushed into the nodejs-runtime-image imagestream.

apiVersion: build.openshift.io/v1
kind: BuildConfig
metadata:
  name: nodejs-runtime-image
spec:
  output:
    to:
      kind: ImageStreamTag
      name: nodejs-runtime-image:latest
  source:
    dockerfile: |-
      FROM nodejs:20-ubi8-minimal
      COPY src $HOME
      CMD /usr/libexec/s2i/run
    images:
    - from:
        kind: ImageStreamTag
        name: nodejs-builder-image:latest
      paths:
      - sourcePath: /opt/app-root/src
        destinationDir: "."
  strategy:
    dockerStrategy:
      from:
        kind: ImageStreamTag
        name: nodejs:20-ubi8-minimal
  triggers:
  - imageChange: {}
    type: ImageChange

Source-to-Image framework and scripts

This image supports the Source-to-Image (S2I) strategy in OpenShift. The Source-to-Image is an OpenShift framework which makes it easy to write images that take application source code as an input, use a builder image like this Node.js container image, and produce a new image that runs the assembled application as an output.

To support the Source-to-Image framework, only the run script is included in this image.

  • The /usr/libexec/s2i/run script is set as the default command in the resulting container image (the new image with the application artifacts). It runs npm run for production, or nodemon if DEV_MODE is set to true (see the Environment variables section below).

Building an application using a Dockerfile

Compared to the Source-to-Image strategy, using a Dockerfile is a more flexible way to build a Node.js container image with an application. Use a Dockerfile when Source-to-Image is not sufficiently flexible for you or when you build the image outside of the OpenShift environment.

To use the Node.js image in a Dockerfile, follow these steps:

1. Pull the base builder and minimal runtime images

podman pull ubi8/nodejs-20
podman pull ubi8/nodejs-20-minimal

The UBI images ubi8/nodejs-20 and ubi8/nodejs-20-minimal that are used in this example are both usable and freely redistributable under the terms of the UBI End User License Agreement (EULA). See more about UBI at UBI FAQ.

2. Pull an application code

An example application available at https://github.com/sclorg/nodejs-ex.git is used here. Feel free to clone the repository for further experiments.

git clone https://github.com/sclorg/nodejs-ex.git app-src

3. Prepare an application inside a container

This step usually consists of at least these parts:

  • putting the application source into the container
  • installing the dependencies
  • setting the default command in the resulting image

For all these three parts, users can either setup all manually and use commands nodejs and npm explicitly in the Dockerfile (3.1.), or users can use the Source-to-Image scripts inside the image (3.2.; see more about these scripts in the section "Source-to-Image framework and scripts" above), that already know how to set-up and run some common Node.js applications.

3.1. To use your own setup, create a Dockerfile with this content:
# First stage builds the application
FROM ubi8/nodejs-20 as builder

# Add application sources
ADD app-src $HOME

# Install the dependencies
RUN npm install

# Second stage copies the application to the minimal image
FROM ubi8/nodejs-20-minimal

# Copy the application source and build artifacts from the builder image to this one
COPY --from=builder $HOME $HOME

# Run script uses standard ways to run the application
CMD npm run -d start
3.2. To use the Source-to-Image scripts and build an image using a Dockerfile, create a Dockerfile with this content:
# First stage builds the application
FROM ubi8/nodejs-20 as builder

# Add application sources to a directory that the assemble script expects them
# and set permissions so that the container runs without root access
USER 0
ADD app-src /tmp/src
RUN chown -R 1001:0 /tmp/src
USER 1001

# Install the dependencies
RUN /usr/libexec/s2i/assemble

# Second stage copies the application to the minimal image
FROM ubi8/nodejs-20-minimal

# Copy the application source and build artifacts from the builder image to this one
COPY --from=builder $HOME $HOME

# Set the default command for the resulting image
CMD /usr/libexec/s2i/run

4. Build a new image from a Dockerfile prepared in the previous step

podman build -t node-app .

5. Run the resulting image with the final application

podman run -d node-app

Environment variables for Source-to-Image

Application developers can use the following environment variables to configure the runtime behavior of this image in OpenShift:

Used in the minimal image

NODE_ENV
NodeJS runtime mode (default: "production")

DEV_MODE
When set to "true", nodemon will be used to automatically reload the server while you work (default: "false"). Setting DEV_MODE to "true" will change the NODE_ENV default to "development" (if not explicitly set).

NPM_BUILD
Select an alternate / custom build command, defined in your package.json file's scripts section (default: npm run "build"). These user-defined run-scripts are unavailable while DEV_MODE is in use.

NPM_RUN
Select an alternate / custom runtime mode, defined in your package.json file's scripts section (default: npm run "start"). These user-defined run-scripts are unavailable while DEV_MODE is in use.

NODE_CMD When specified (e.g.Specify NODE_CMD="node server.js") the value of NODE_CMD is used to start the application instead of npm start.

INIT_WRAPPER When set to "true", the application is started via the init-wrapper script instead of using npm start, by looking for the presence of the files server.js, index.js or main.js in the order in which they are listed. In case of NODE_CMD environemnt variale is specified, then init-wrapper script will use the value of NODE_CMD to start your application.

Additional variables used in the full-sized image

HTTP_PROXY
Use an npm proxy during assembly

HTTPS_PROXY
Use an npm proxy during assembly

NPM_MIRROR
Use a custom NPM registry mirror to download packages during the build process

One way to define a set of environment variables is to include them as key value pairs in your repo's .s2i/environment file.

Example: DATABASE_USER=sampleUser

NOTE: Define your own "DEV_MODE":

The following package.json example includes a scripts.dev entry. You can define your own custom NPM_RUN scripts in your application's package.json file.

Note: Setting logging output verbosity

To alter the level of logs output during an npm install the npm_config_loglevel environment variable can be set. See npm-config.

Development Mode

This image supports development mode. This mode can be switched on and off with the environment variable DEV_MODE. DEV_MODE can either be set to true or false. Development mode supports two features:

  • Hot Deploy
  • Debugging

The debug port can be specified with the environment variable DEBUG_PORT. DEBUG_PORT is only valid if DEV_MODE=true.

A simple example command for running the container in development mode is:

podman run --env DEV_MODE=true my-image-id

To run the container in development mode with a debug port of 5454, run:

$ podman run --env DEV_MODE=true DEBUG_PORT=5454 my-image-id

To run the container in production mode, run:

$ podman run --env DEV_MODE=false my-image-id

By default, DEV_MODE is set to false, and DEBUG_PORT is set to 5858, however the DEBUG_PORT is only relevant if DEV_MODE=true.

Hot deploy

As part of development mode, this image supports hot deploy. If development mode is enabled, any souce code that is changed in the running container will be immediately reflected in the running nodejs application.

Using Podman's exec

To change your source code in a running container, use Podman's exec command:

$ podman exec -it <CONTAINER_ID> /bin/bash

After you Podman exec into the running container, your current directory is set to /opt/app-root/src, where the source code for your application is located.

Using OpenShift's rsync

If you have deployed the container to OpenShift, you can use oc rsync to copy local files to a remote container running in an OpenShift pod.

Warning:

The default behaviour of the s2i-nodejs container image is to run the Node.js application using the command npm start. This runs the start script in the package.json file. In developer mode, the application is run using the command nodemon. The default behaviour of nodemon is to look for the main attribute in the package.json file, and execute that script. If the main attribute doesn't appear in the package.json file, it executes the start script. So, in order to achieve some sort of uniform functionality between production and development modes, the user should remove the main attribute.

Below is an example package.json file with the main attribute and start script marked appropriately:

{
    "name": "node-echo",
    "version": "0.0.1",
    "description": "node-echo",
    "main": "example.js", <--- main attribute
    "dependencies": {
    },
    "devDependencies": {
        "nodemon": "*"
    },
    "engine": {
        "node": "*",
        "npm": "*"
    },
    "scripts": {
        "dev": "nodemon --ignore node_modules/ server.js",
        "start": "node server.js" <-- start script
    },
    "keywords": [
        "Echo"
    ],
    "license": "",
}

Note:

oc rsync is only available in versions 3.1+ of OpenShift.

init-wrapper

init-wrapper script is located on /usr/libexec/s2i/init-wrapper and is used to handle:

  • Proper signal handling and propagation, as Node.js was not designed to run as PID 1.
  • Reaping zombie child processes Avoiding use of npm, there is more information on why you want to avoid that in the Node.js reference architecture. When the INIT_WRAPPER is set to true the application is started via the init script instead of using npm start.

A detailed explanation on how the init-wrapper script works is avalable in this url.

Example of using init-wrapper:

During image build

s2i -e INIT_WRAPPER=true build . buildImage  node-app
docker run node-app

During container start

s2i build . buildImage  node-app
docker run -e INIT_WRAPPER=true  node-app

init-wrapper script can be disabled by setting the INIT_WRAPPER env variable to false.

docker run -e INIT_WRAPPER=false node-app

NODE_CMD can be used during the build process or container start, in order to have more control on the command that init-wrapper script will wrap.

For example:

during container build

s2i -e INIT_WRAPPER=true -e NODE_CMD="node index.js" build . buildImage  node-app

during container start

docker run -e INIT_WRAPPER=false -e NODE_CMD="node index.js" node-app

See also

Dockerfile and other sources are available on https://github.com/sclorg/s2i-nodejs-container. In that repository you also can find another versions of Node.js environment Dockerfiles. Dockerfile for CentOS Stream 9 is called Dockerfile.c9s, for RHEL8 it's Dockerfile.rhel8 and the Fedora Dockerfile is called Dockerfile.fedora.