Introduction

This is a flask application that leverages RabbitMQ and Celery to asynchronously create a Cheaha user account. Currently the project is being developed on an Openstack cluster.

Project Setup

To clone this repo use the command:

$ git clone https://gitlab.rc.uab.edu/mmoo97/flask_user_reg.git
$ cd flask_user_reg

Prerequisites

Setup a Virtual Environment

• Ensure you have created a virtual environment called venv setup running python3.
  • Note, this project requires a virtual environment running python3 (3.6.8 in this case).
    • Create this by navigating to you home directory via typing $ cd and entering the following commands:
      

    $ python3 -m venv ~/venv
    $ source ~/venv

    • Upon Activation, you should see the prompt update accordingly:

    [centos@ood ~]$              <------Old Prompt
    (venv) [centos@ood ~]$       <------New Prompt

    In this case, the env name is displayed as venv but would change to reflect whatever name you initialized it with in the previous step. Additionally, this example is running on the ood node provisioned via OpenStack.
• Ensure pip is installed.

  • Mac/Linux

    • Check if installed by typing $ pip
    • Install pip using $ python -m pip install --user --upgrade pip.

  • Windows

    • Check if installed using $ py
    • Install pip using $ py -m pip install --upgrade pip
      
      
• Ensure Flask and other dependencies are installed to you virtual environment using the following commands:

$ cd ~/your/repo/path/flask_user_reg
$ pip install -r requirements.txt

• Note, to install flask in your own $HOME use pip install --user Flask.

Install RabbitMQ

(Reference: here)

• Install RabbitMQ server on the host machine. (Installation Guide)

  • Mac/Linux

    • $ brew update
    • $ brew install rabbitmq
    • it is recommended that you add the following line to your .bash_profile: export PATH=$PATH:/usr/local/opt/rabbitmq/sbin.
    • Start server using the command $ rabbitmq-server. (Note, this implementation assumes RabbitMQ is running on localhost on standard port 5672)

  • Windows

    • Download the installer from here and run.
    • Post install, the server should be running. To check, run $ rabbitmqctl.bat status.

  • CentOS 7

    • First, import signing key using $ rpm --import https://github.com/rabbitmq/signing-keys/releases/download/2.0/rabbitmq-release-signing-key.asc
    • Next, install Erlang using $ yum install https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm
    • Finally, install RabbitMq using $ yum install rabbitmq-server-3.8.2-1.el8.noarch.rpm
    • The server is not started as a daemon by default when the RabbitMQ server package is installed. To start the daemon by default when the system boots, as an administrator run: $ chkconfig rabbitmq-server on.
    • As an administrator, start and stop the server as usual:
      /sbin/service rabbitmq-server start
/sbin/service rabbitmq-server stop
• Add the user "reggie" using the command $ rabbitmqctl add_user reggie reggie.
  • The current configuration assumes the following username password combination. To change the password, type $ rabbitmqctl change_password reggie <new_password>.
  • Note that rabbitmqctl may require sudo permissions and that changing the password will require a password change in the credentials within test_producer.py and base_consumer.py as well.

Setup pre commit hook

• via Homebrew(Mac)

brew install pre-commit

• via pip

pip install pre-commit

After pre-commit is installed, run install in the repo directory:

$ pre-commit install
> pre-commit installed at .git/hooks/pre-commit

Run Project

1. Connect to OHPC/OOD

1. Network Setup:
   Assuming proper steps have been run to provision your openstack cluster, your network topology should look like the following:
   

2. Locate Floating Ip for each instance:
   
   According to this diagram of the network infrastructure, the IP Addresses that you will need to connect to in order to ssh into each machine will be 164.111.161.xxx where .xxx is the last octet of your assigned Floating IP Addresses shown below.
   
   Note that these numbers will typically vary depending on factors in the provisioning process.

3. SSH into each machine:
   Now that you have the IP Addresses of the OHPC and OOD nodes, you can connect to them via the following command in two separate shell windows/tabs:
   

   $ ssh centos@164.111.161.xxx

4. Drop Firewall:
   Currently, the firewall prevents us from connecting to our soon to be running flask app. To check if the firewall is active, type:
   $ sudo systemctl status firewalld
   In the case the firewall is still active, type:
   $ sudo systemctl stop firewalld
   In the case you need to turn the firewall back on, type:
   sudo systemctl start firewalld

5. Edit Security Rules:
   Lastly, we want to make sure that our network it open to run on our flask application which runs on localhost:5000 by default. Modify your Default security group to reflect the following rules:
   
   The main takeaway/modification in these rules is that traffic is allowed in and out of port 5000.

2. Initialize RabbitMQ

Make sure that you are running rabbitmq via the command sudo rabbitmq server.

3. Test RabbitMQ

For a simple example on the functionality of RabbitMQ, do the following:

• Open up a new ssh terminal and ssh into your ohpc node and, in a separate window, ssh into your ood node.
• Additionally, ensure you have a rabbitmq user configured with the username and password as reggie.

# Run consumer on ohpc node
$ python base_consumer.py ohpc

You will notice that the script is awaiting a message to display. To produce a message, run the following on the ood node terminal session:

$ python test_producer.py ohpc

You should now see that the message has been sent and displayed on the ohpc node.

4. Celery Worker

In order to execute our tasks asychronously, we create a celery worker using the fllowing:
$ celery -A tasks worker --loglevel=info --concurrency=4 1> ~/celery.out 2> ~/celery.err &

In this case, the celery -A tasks worker --loglevel=info --concurrency=4 portion of the command is what is initiating the worker and the remainder serves to write the stdout and stderr to two separate files located in the home directory and to run the process in the background.

5. Initialize the Flask App

Simply type $ python run.py
Alternatively, type $ python run.py > ~/flask.out 2> ~/flask.err & to run the process in the background and output to files in the home directory.

6. Check Job status

If all goes well, you should have your processes running in the background. Check the status of these by typing $ jobs. You should now see the something similar to the following:

(venv) [centos@ood flask_user_reg]$ jobs
[1]-  Running                 celery -A tasks worker --loglevel=info --concurrency=4 > ~/celery.out 2> ~/celery.err &
[2]+  Running                 python run.py > ~/server.out 2> ~/server.err &
(venv) [centos@ood flask_user_reg]$

7. Connect to Server

Open a new browser window and connect to the OOD node by typing http://164.111.161.xxx:5000 in the address bar replacing .xxx with the final octet of the OOD node. You should see something like this:

8. Key Info/Observe

• Fill out the form and hit "Submit"
• You should see an overlay signifying the account is being created. After 5 seconds it should disappear and signify the account creation has been successful.
• You can now enter the command kill %1 %2 to terminate the celery worker and the flask server.
• You can view the output/errors of either the worker and flask server by opening flask.out, flask.err, celery.out, or celery.err.