PHP project using Mezzio downloads usage logs from S3, parse them, and add them to a MySQL database. Following addition to the database, stats pages show usage data and charts with results graphically using Google Charts.
NOTE: This application currently is untested, but works. I intend to return and do that, but it's not done yet. Pull requests welcomed.
Clone the Beachcasts/S3-Log-Analyzer repo from Github. In your terminal, navigate to the new directory.
Install dependencies using Composer.
composer install
Create a new MySQL database by using statements in /data/schema.sql
to create the database and table.
Rename .env.dist
to become .env
, and update as needed with your AWS account (IAM) settings.
NOTE: The bucket should be the location of the stats, not necessarily the bucket you want stats on. In AWS you will need to activate the stats for a given bucket, and inform where to store them. In my case, I store the stats in a bucket named 'stats-bucket'.
Rename /config/autoload/db.local.php.dist
to become /config/autoload/db.local.php
, and update as needed for your database credentials and location.
At this point you should have a "working" application ready to serve. There are many ways to accomplish this. Here are some examples:
- Using the built-in PHP server
php -S localhost:8080 -t public/
-
Upload the application to a webserver, or running locally, as with any PHP web-based application.
-
Deploy to AWS Lambda as an HTTP function
This option can can be easily accomplished by installing Serverless, and issuing the following command to deploy to Lambda.
Update the contents of
serverless.yml
by adding the bucket ARN from AWS. Replace<stats-bucket>
with your bucket name. You may also need to update the AWS region.Also, some of the variables in
.env
may not be needed if deployed this way, because they are already handled byserverless.yml
.
serverless deploy
NOTE: If deploying to Lambda, you will need to update the menu in
/src/App/templates/layout/default.html
according to the URL provided by Lambda.