Skip to content

Latest commit

 

History

History
227 lines (154 loc) · 5.97 KB

02-dockerize-apps.md

File metadata and controls

227 lines (154 loc) · 5.97 KB
Raw

Dockerizing Applications

Build Container Images

For the first container, we will be creating a Dockerfile from scratch. For the other containers, the Dockerfiles are provided.

Web Container

  1. Create a Dockerfile

    • Access the jumpbox

    • In the ~/blackbelt-aks-hackfest/app/web directory, add a file called "Dockerfile"

      • If you are in an SSH session, use vi as the editor
      • In RDP or in Azure Cloud Shell, you can use code . (use Visual Studio Code)

    • Add the following lines and save:

      FROM node:9.4.0-alpine

ARG VCS_REF
ARG BUILD_DATE
ARG IMAGE_TAG_REF

ENV GIT_SHA $VCS_REF
ENV IMAGE_BUILD_DATE $BUILD_DATE
ENV IMAGE_TAG $IMAGE_TAG_REF

WORKDIR /usr/src/app
COPY package*.json ./
RUN npm install

COPY . .
RUN apk --no-cache add curl
EXPOSE 8080

CMD [ "npm", "run", "container" ]

  2. Create a container image for the node.js Web app

    From the terminal session:

    cd ~/blackbelt-aks-hackfest/app/web

docker build --build-arg BUILD_DATE=`date -u +"%Y-%m-%dT%H:%M:%SZ"` --build-arg VCS_REF=`git rev-parse --short HEAD` --build-arg IMAGE_TAG_REF=v1 -t rating-web .

  3. Validate image was created with docker images

API Container

In this step, the Dockerfile has been created for you.

  1. Create a container image for the node.js API app

    cd ~/blackbelt-aks-hackfest/app/api

docker build -t rating-api .

  2. Validate image was created with docker images

MongoDB Container

  1. Create a MongoDB image with data files

    cd ~/blackbelt-aks-hackfest/app/db

docker build -t rating-db .

  2. Validate image was created with docker images

Run Containers

Setup Docker Network

Create a docker bridge network to allow the containers to communicate internally.

docker network create my-network

MongoDB Container

  1. Run mongo container

    docker run -d --name db --net my-network -p 27017:27017 rating-db

  2. Validate by running docker ps -a

  3. Import data into database

    docker exec -it db bash

    You will have a prompt inside the mongo container. From that prompt, run the import script (./import.sh)

    root@61f9894538d0:/# ./import.sh
2018-01-10T19:26:07.746+0000	connected to: localhost
2018-01-10T19:26:07.761+0000	imported 4 documents
2018-01-10T19:26:07.776+0000	connected to: localhost
2018-01-10T19:26:07.787+0000	imported 72 documents
2018-01-10T19:26:07.746+0000	connected to: localhost
2018-01-10T19:26:07.761+0000	imported 2 documents

  4. Type exit to exit out of container

API Container

  1. Run api app container

    docker run -d --name api -e "MONGODB_URI=mongodb://db:27017/webratings" --net my-network -p 3000:3000 rating-api

    Note that environment variables are used here to direct the api app to mongo.

  2. Validate by running docker ps -a

  3. Test api app with curl

    curl http://localhost:3000/api/heroes

Web Container

  1. Run web app container

    docker run -d --name web -e "API=http://api:3000/" --net my-network -p 8080:8080 rating-web

  2. Validate by running docker ps -a

  3. Test web app

    The jumpbox has a Public IP address and port 8080 is open. You can browse your running app with a link such as: http://13.90.246.114:8080

    You can also test via curl

    curl http://localhost:8080

Azure Container Registry (ACR)

Now that we have container images for our application components, we need to store them in a secure, central location. In this lab we will use Azure Container Registry for this.

Create Azure Container Registry instance

  1. In the browser, sign in to the Azure portal at https://portal.azure.com. Your Azure login ID will look something like odl_user_9294@gbbossteamoutlook.onmicrosoft.com

  2. Click "Create a resource" and select "Container Registry"

  3. Provide a name for your registry (this must be unique)

  4. Use the existing Resource Group

  5. Enable the Admin user

  6. Use the 'Standard' SKU (default)

    The Standard registry offers the same capabilities as Basic, but with increased storage limits and image throughput. Standard registries should satisfy the needs of most production scenarios.

Login to your ACR with Docker

  1. Browse to your Container Registry in the Azure Portal

  2. Click on "Access keys"

  3. Make note of the "Login server", "Username", and "password"

  4. In the terminal session on the jumpbox, set each value to a variable as shown below

    # set these values to yours
ACR_SERVER=
ACR_USER=
ACR_PWD=

docker login --username $ACR_USER --password $ACR_PWD $ACR_SERVER

Tag images with ACR server and repository

# Be sure to replace the login server value

docker tag rating-db $ACR_SERVER/azureworkshop/rating-db:v1; 
docker tag rating-api $ACR_SERVER/azureworkshop/rating-api:v1;
docker tag rating-web $ACR_SERVER/azureworkshop/rating-web:v1;

Push images to registry

docker push $ACR_SERVER/azureworkshop/rating-db:v1;
docker push $ACR_SERVER/azureworkshop/rating-api:v1;
docker push $ACR_SERVER/azureworkshop/rating-web:v1;

Output from a successful docker push command is similar to:

The push refers to a repository [mycontainerregistry.azurecr.io/azureworkshop/rating-db]
035c23fa7393: Pushed
9c2d2977a0f4: Pushed
d7b18f71e002: Pushed
ec41608c0258: Pushed
ea882d709aca: Pushed
74bae5e77d80: Pushed
9cc75948c0bd: Pushed
fc8be3acfc2d: Pushed
f2749fe0b821: Pushed
ddad740d98a1: Pushed
eb228bcf2537: Pushed
dbc5f877c367: Pushed
cfce7a8ae632: Pushed
v1: digest: sha256:f84eba148dfe244f8f8ad0d4ea57ebf82b6ff41f27a903cbb7e3fbe377bb290a size: 3028

Validate images in Azure

  1. Return to the Azure Portal in your browser and validate that the images appear in your Container Registry under the "Repositories" area.
  2. Under tags, you will see "v1" listed.