scan2deploy

Just write down your docker file , take a snap and deploy the app in Kubernetes!

Demo Link

Requirements

Install OpenCV by

$ pip install opencv-python

Do this if your Kubernetes cluster is running in aws ->

Install boto3 (calls APIs of AWS services) by

$ pip install boto3

Install aws command line tool by

$ pip install awscli

Since I am running the code on Windows machine , Docker Desktop is required to get the docker enviornment for building and pushing images Download Docker Desktop from -> link

Enviornment settings

Docker Desktop-

After opening Docker Desktop, provide your Docker Hub credentials so that pushing images happens smoothly.

AWS-

configure AWS-CLI by -

$ aws configure

Create a bucket in S3 (here the snap will be uploaded)

$ aws s3api create-bucket --bucket <bucket-name> --region <region-name> --create-bucket-configuration LocationConstraint=<region-name>

Have the Kubernetes master key (.pem file) in the working directory, so that deployment can be made using ssh

Working and Usage

The image of your page is uploaded in the S3 bucket. This image is then processed by AWS Textract service, which extracts the text from it line by line. More on AWS Textract -> link

textract = boto3.client('textract')

response = textract.detect_document_text(
    Document={
        
        'S3Object': {
            'Bucket': bucket_name,
            'Name': file_name,
            
        }
    }
)

the lines are written in the Dockerfile, which is then used to build the container image and push it to Docker Hub.

After pushing the image to Docker Hub, the code runs the command to create a deployment using the image pushed (using ssh) and expose it to create a service.

To get the port number where the service is launched, go to the master node and run ->

$ kubectl get svc

To get the IP of node on which the pod is launched run this ->

$ kubectl describe pod <name-of-pod>

The app can be accessed by http://<IP-of-node>:<port_no>