Skip to content

Latest commit

 

History

History
126 lines (90 loc) · 3.62 KB

README.md

File metadata and controls

126 lines (90 loc) · 3.62 KB
Raw

Set GCS credentials

A valid GCS key file must be placed in the following directory in the repository:

/creds/WebCrawler-feb11a08e450.json

In order to generate this json file, see the instructions

Run tests locally

Unit tests can be run locally with the following command:

./run_unit_tests.sh

Local Dev Environment with Docker

The application requires the following containers:

  • Main Application
  • Crawler Manger
  • Crawler
  • MySQL
  • Redis

In production, MySQL and Redis will not be ran on containers.
Crawler Manager container also has an internal Redis service.

Dependencies

  • Docker Desktop (2.x)

Steps

  1. Build and start the containers from the root of the repository:

    touch main/kubeconfig; touch crawler-manager/kubeconfig
docker-compose -f docker-compose.yml up --build --no-start
docker-compose -f docker-compose.yml start

  2. All three containers will be started, but only the main container will have a permanent webserver running, this is the access route:

    Main: `http://localhost:8001/`

    If the main application is not loading, then likely the database isn't created properly. Enter the shell for the mysql container and create the test database

    docker exec -it webcrawler_crawler-mysql5_1 bash
$ mysql -u root -proot

mysql> CREATE DATABASE test;

  3. For running the crawler-manager (No Longer Required. Crawler manager is also getting started in step 1):

    This does not need to be built, it's done already in the docker-compose command, but to simulate a new crawler manager, run the following:

    docker exec -i crawler-manager /bin/bash -c "export JOB_ID=123; export ENVIRONMENT=prod && service redis-server start && python app.py"

    That will start the webserver for 60s, like is done in kubernetes

  4. For running the crawler, is like the manager (No Longer Required. Crawler is also getting started in step 1)

    Sames as above, does not need to be built.

    For processing urls

    docker exec -i crawler /bin/bash -c "export URL='http://google.com,https://cnn.com'; export ENVIRONMENT=prod && python app.py"

    To clear Redis cache

    docker exec -it webcrawler_crawler-redis_1 redis-cli FLUSHALL

    Crawler runs for 15s

    Note: Change the value in the export command, for the URLS you want to process

  5. In order to reload code changes into the containers, run these comands:

    Main:

    docker cp main/. main-app:/srv/www/web-crawler/
docker-compose -f docker-compose.yml restart main

    Crawler Manager:

    docker cp crawler-manager/. crawler-manager:/srv/www/web-crawler/    
docker-compose -f docker-compose.yml restart crawler-manager

    Crawler:

    docker cp crawler/. crawler:/srv/www/web-crawler/
docker-compose -f docker-compose.yml restart crawler

  6. After all containers are running, you need to run the initialize-django.sh script, to initialize the DB and set up a super user, so you can use the admin UI to create more users.

    docker exec -it main-app bash -c "./initialize-django.sh"

Database Migration In case that a new field is added to the model, the easiest way to update the database is by droping the database, creating it again and then run the initialize script:

drop database test;
create database test;

then run again the initialize script:

docker exec -it main-app bash -c "./initialize-django.sh"