Skip to content

Milestone 2 : Project Setup Instructions

Jump to bottom
Adithya Selvaprithiviraj edited this page Mar 31, 2020 · 24 revisions

Milestone 2 - Project Setup Instructions

Download or pull the repository in to your local machine.

git clone https://github.com/airavata-courses/CAPtivate.git

Run the application with Docker locally

Software Requirements

Steps

  1. Go to the folder containing the code in the command prompt.
  2. Run the application using the following command.
docker-compose up

To check if images are built in your local machine in Docker using the following command.

docker image ls

You can check the service running at using browser of your choice

http://localhost:3000

Run the application with Docker + Kubernetes on local machine

Software Requirements

Steps

  1. Go to the deployments folder in your directory containing the code.
  2. Run the following commands in a command prompt.
kubectl apply -f zookeeper.yaml
kubectl apply -f kafka.yaml
kubectl apply -f db.yaml
kubectl apply -f services.yaml
kubectl apply -f api-gateway.yaml
kubectl apply -f user-management.yaml
kubectl apply -f session-management.yaml
kubectl apply -f data-retrieval.yaml
kubectl apply -f model-execution.yaml
kubectl apply -f post-process.yaml
kubectl apply -f ui.yaml

You will be able to check the deployed application by using the url from ui deployment of kubectl get services

Run the application on VM Kubernetes cluster

Current master node IP - 149.165.168.90

Requirements

  • Kubernetes cluster set up on the VMs in jetstream

Steps to set up VM's and Kubernetes Cluster

  1. Login into https://iu.jetstream-cloud.org/auth/login/ and ensure you are under TG-CCR*****43 project allocation and download OpenRC File v3 from the dropdown (top right corner).

  2. Open your Git Bash terminal and navigate to the path where you placed the OpenRC v3 file and run the following command.

source <OpenRC file>

You can follow the below steps to use our network and security groups to create VM's (recommended) or you create your own VM network and security groups by following the openstack CLI commands

  1. Create SSH key pair to access VMs in the OpenStack environment. It will ask you for the password, enter your IU jetstream password.

If you already have an SSH key then follow

cd ~/.ssh

openstack keypair create --public-key id_rsa.pub tg865486-api-key

If you dont have SSH key then follow:

ssh-keygen -b 2048 -t rsa -f tg865486-api-key -P ""
openstack keypair create --public-key tg865486-api-key.pub tg865486-api-key
  1. Now we will create VM's to set up the architecture. You can replace the VM names (VM-master, VM-worker1, VM-worker2) with appropriate names as you like.
openstack server create <VM-master> \
--flavor m1.small \
--image JS-API-Featured-Ubuntu18-Feb-14-2020 \
--key-name tg865486-api-key \
--security-group CAPtivate_security_group \
--nic net-id=tg865486_Captivate_net \
--wait

openstack server create <VM-worker1> \
--flavor m1.small \
--image JS-API-Featured-Ubuntu18-Feb-14-2020 \
--key-name tg865486-api-key \
--security-group CAPtivate_security_group \
--nic net-id=tg865486_Captivate_net \
--wait

openstack server create <VM-worker2> \
--flavor m1.small \
--image JS-API-Featured-Ubuntu18-Feb-14-2020 \
--key-name tg865486-api-key \
--security-group CAPtivate_security_group \
--nic net-id=tg865486_Captivate_net \
--wait
  1. Now we will assign IPs to all 3 VMs.
openstack floating ip create public
openstack server add floating ip <VM-master> <your.ip.number.here>

openstack floating ip create public
openstack server add floating ip <VM-worker1> <your.ip.number.here>

openstack floating ip create public
openstack server add floating ip <VM-worker2> <your.ip.number.here>

Now we have the VMs ready. To set up the Kubernetes system we will run some Ansible playbooks.

Prerequisites:

  • Docker

  • Ubuntu server/machine : This server acts as an Ansible control node which we will use to connect to and control Ansible hosts (Ansible hosts are our created VM's) over SSH. Your Ansible control node can either be your local machine or a server dedicated to running Ansible.

Make sure you are a non root user and have an SSH key setup. You can follow the link for further clarifications

Steps to install Ansible:

  1. Setup Ansible
sudo apt-add-repository ppa:ansible/ansible

Press ENTER when prompted to accept the PPA addition.

sudo apt update
sudo apt install ansible
  1. Setup Inventory File.

Now you have to replace the master_ip and worker_ip with the IPs of the master and worker VMs you created above.

You have to change the hosts file for that. You open the hosts file by this command.

sudo nano /etc/ansible/hosts

Add the following to the hosts file

[servers]
master ansible_host=<ubuntu user>@<master IP created>
worker1 ansible_host=<ubuntu user>@<worker1 IP created>
worker2 ansible_host=<ubuntu user>@<worker2 IP created>

[servers:vars]
ansible_python_interpreter=/usr/bin/python3

Save and exit from the hosts file.

  1. Test Connection Next, you can test Ansible playbooks with the help of the ping command.
ansible all -m ping

Once the ping is success copy the kube-cluster folder to your Ubuntu server/machine where you installed ansible.

Download the kube-cluster folder from https://github.com/airavata-courses/CAPtivate/tree/master/docs/kube-cluster

Now just replace the IP adresses of the master and worker VM's with the IP adresses of the VM's you created in the hosts file (~/kube-cluster/hosts).

Now you are all set to run the ansible playbooks. Say yes to adding the fingerprint 3 times. (once for each of the VMs)

ansible-playbook -i hosts ~/kube-cluster/initials.yml
ansible-playbook -i hosts ~/kube-cluster/kube-dependencies.yml
ansible-playbook -i hosts ~/kube-cluster/master.yml
ansible-playbook -i hosts ~/kube-cluster/workers.yml

Once all the ansible playbooks run successfully you can login to your Kubernetes master using the command:

ssh ubuntu@<VM-master-IP>

To verify your Kubernetes installation, run the following command:

kubectl get nodes

You should see the master and worker nodes working now.

Steps to run the application

  1. Clone the repository
   git clone https://github.com/airavata-courses/CAPtivate.git

  1. cd CAPtivate/deployments

  2. Run the following commands in the kubectl prompt, in the given order.

   kubectl apply -f zookeeper.yaml
   kubectl apply -f kafka.yaml
   kubectl apply -f db.yaml
   kubectl apply -f services.yaml
   kubectl apply -f api-gateway.yaml
   kubectl apply -f user-management.yaml
   kubectl apply -f session-management.yaml
   kubectl apply -f data-retrieval.yaml
   kubectl apply -f model-execution.yaml
   kubectl apply -f post-process.yaml
   kubectl apply -f ui.yaml

Current Deployed application

You can find the current deployment here - http://149.165.168.90:31703

Continuous Integration and Continuous deployment

CI/CD is implemented through travis and a remote jenkins server. Each service have their own deployment branch as follows and the respective travis scripts can be found in .travis.yml in root of respective branches.

Jenkins build status and pipeline can be verified from http://149.165.169.77:8080/ Username: rakoduru Password : rasmitha

Deployment branches
  • deploy/api-gateway
  • deploy/user-management
  • deploy/session-management
  • deploy/post-process
  • deploy/user-interface
  • deploy/model-execution
  • deploy/data-retrieval
Clone this wiki locally