Cloned the appcode files in VS code https://gitlab.com/nanuchi/developing-with-docker.git
Installed node latest https://nodejs.org/en
Tried running node server.js got error “express module was missing” Run npm install express Run node server.js
In browser : not able to see UI portion for localhost:3000 ? Working now Run Docker pull mongo Run Docker pull mongo-express
Gets copies of these images for future use docker network create mongo-network docker network ls
Create MONGO & Mongo-express containers: they can connect with each other just using the container names since both of them are in the same isolated docker network
docker run -p 27017:27017 -d -e MONGO_INITDB_ROOT_USERNAME=admin -e MONGO_INITDB_ROOT_PASSWORD=secret --name mongodb --net mongo-network mongo
open port 27017 on local host and bind to mongo container port 27017 (default port of mongodb is 27017) -d run the container in detached mode – allows to use terminal while the docker container is running in background -e MONGO_INITDB_ROOT_USERNAME -e MONGO_INITDB_ROOT_PASSWORD Define root username & password for mongo at startup of the container --name – define container’s custom name -- net launch container in the network created earlier If launched successfully – we get a containerID
docker logs containerID
docker run -d -p 8081:8081 -e ME_CONFIG_MONGODB_ADMINUSERNAME=admin -e ME_CONFIG_MONGODB_ADMINPASSWORD=secret --name mongo-express --net mongo-network -e ME_CONFIG_MONGODB_SERVER=mongodb mongo-express
- Open port 8081 on localhost and bind to mongo-express default port (8081)
- Give mongo username & password so that express can connect to mongo db
- Give mongo containerID as env variable: mongoServer
In red font we can see the basic auth credentials => admin and pass Giving the same in localhost:8081 will launch mongo-express UI
Created new db from ui or can also done with userint db as env variable in startup command for docker container
Now, we need to connect to this database by giving username and password from the previous command in the code base (not a good practice) 4 places in code
Create a collection in mongo-express UI: users
Now try to add user from UI the same should be visible in the mongo-express DB
Docker logs mongo | tail to view the last logs
Docker logs mongo -f to stream the logs
So, now we have a fully functional app serving user profile in the frontend using javascript and nodejs in backend up and running
Here, we have given many docker commands to get this app running too much manual work!!
We can automate this using docker compose file If we have bunch of containers to work with, we probably want to automate the deployment and starting of applications Docker compose: allows to running multiple docker containers with all the configuration easily than with individually passing docker run commands
Running this file from terminal :
docker compose -f mongo.yaml up
Since both containers are started at once, logs are mixed for both, we can add waiting logic in yaml file when one container depends on another as here Also the previously created db from mongo-express ui is gone because we don’t have data persistence so when containers are restarted all data is lost. This can be prevented using data volumes starting the app again now
Stopping the containers:
docker compose -f mongo.yaml down
The network created is also deleted
Creating Dockerfile & building the image: To deploy this app on dev, test or prod server, we need to package this app along with the dependencies into a single artifact Docker file
Now we are going to simulate what Jenkins does locally: When we commit changes to git a CI that runs (Jenkins), that builds the app -> packages app to a docker image pushes to registry
Run: docker build -t imageName:tag location-of-docker-file
Make sure to give proper path in cmd path to find the app source file to start the app
All the files in app folder got copied in the container
Pushing image to AMAZON ECR – private registry
- Login to amazon console and head to Amazon ECR
In ECR, we have to create a repo per image, unlike docker hub where we can store multiple images in a repo In the repo, we can store different versions of the same image
We want to push this image to ECR First authenticate to your aws account using aws configure command
Retrieve an authentication token and authenticate your Docker client to your registry. Use the AWS CLI:
Build your Docker image using the following command.
For information on building a Docker file from scratch, see the instructions here . You can skip this step if your image has already been built:
docker build -t my-app:1.0
After the build is completed, tag your image so you can push the image to this repository:
docker tag my-app:1.0 703458314422.dkr.ecr.us-east-1.amazonaws.com/my-app:1.0
Run the following command to push this image to your newly created AWS repository: We have created repo with same name as image name Image pushed to private repo Image pushed successfully to ECR
Making some changes to app source code or docker file rebuild the image => push to ECR with updated version
Pushing to ECR the new version
Deploy the app on server now.. Now this server has to pull the app image from private ECR repo, we first need to provide docker login so that server can authenticate to ECR
Install aws cli and docker on the server Authenticate to aws : aws configure Now run the compose file having my-app added as container name
Run the app , and test it out
Docker Volumes: When we restart a container, all the data stored in mongodb is lost! So mounting Named volumes to containers, will mount a folder in host to the containers for persisting data when containers are restarted
Database and user info persists in UI even after restarting the containers