commands.txt

Steps to Deploy ML Models on Docker

1. Build a Docker :  sudo docker build -t gr_deploy:test .
2. Run a Docker : sudo docker run -d -p 80:5000 gr_deploy:test
   ## 80 : local
   ## 5000 : inside container
3. Check if port is free : netstat -nultp
4. Docker logs : sudo docker logs 2f22d6945655
5. Enter a Docker  : sudo docker exec -it 970987e98ece  bash
6. Delete all docker images and containers : sudo docker system prune -a
7. Deleting a Docker Container : sudo docker container rm cc3f2ff51cab cd20b396a061
8. Stop a Running Docker Container :sudo docker container stop f9142795b2ce ffda66ea92f1

Steps to Deploy on Kubernetes
1. Tag Image : gcr.io/project-id/image-name:tag
   sudo docker tag gr_deploy:v1 gcr.io/project-id/image-name:tag
2. Push the docker image to registry. Registry stores docker image
   sudo docker push gcr.io/project-id/image-name:tag
3. Update the pushed image-name in yaml file
4. Create GKE Cluster
5. Install kubectl
   Link: https://kubernetes.io/docs/tasks/tools/install-kubectl/#install-kubectl-on-linux
6. Go to cluster. Click connect. Copy command and run in local
7. sudo kubectl get nodes
   To validate connection with cluster
8. kubectl apply -f kubernetes.yaml
9. Verify deployment pods and services on cluster
   kubectl get pods, svc
10. For accessing the URL, create proxy
    Go to GKE console
    Go to Service and Ingress.
    Click on Service.
    Scroll Down. Go to Ports. Click Port Forwarding.
    Copy the command and execute in local.