Create an AMI for Amazon EC2 that runs OpenTripPlanner and reads data from S3

For this project, I require the creation of an AMI (Amazon Machine Image) that runs Linux and OpenTripPlanner (also known as OTP, <[login to view URL]>) OpenTripPlanners is a Java application, so a Java server (such as Apache Tomcat) must be used. The AMI must be designed so it reads the OTP server files directly from Amazon S3 on boot. Additionally it must read the data files directly from Amazon S3 also, as I will need to update the data files frequently. There is sample data you can use to test the installation available on the download site. The installation must support multiple data files (that is, multiple cities) for a single installation. (Note: I'm 90% sure this is possible, but it may be the case that only 1 city will work at a time) Successful approval of your work will involve: * **A working installation of OpenTripPlanner.** This means once the EC2 instance has full booted there will be a running web server on port 80 that runs the web interface to OTP * **The OpenTripPlanner API must be accessible.** I believe the web frontend won't work without this component anyway. * **OpenTripPlanner server files downloaded from S3 on boot.** This is so I can upgrade OTP without having to recreate the AMI. The name of the S3 bucket must be configurable when starting the instance * **OpenTripPlanner data files downloaded from S3 on boot.** This is so I can replace the data file when required * **Multiple data files (cities) supported at once.** I need to run this for up to about 50 cities. I will scale the solution by starting more EC2 instances. You will need to use your Amazon AWS account to complete this job. All tasks should be completed your own account, and I believe can all be achieved using the "free tier" of AWS services. Once completed, we will arrange to give me access to the AMI so I can use it in my own environment. ## Deliverables While I will leave it up to you to decide exact implementation, I believe the workflow when booting an EC2 instance with the created AMI would be something like this: 1. Start EC2 instance 2. Fetch OTP installation files from specified S3 bucket 3. Fetch OTP city data files from spceified S3 bucket 4. Start OTP web <!-- --> 1. server The bucket name should be specified as an instance tag when creating the EC2 instance. I've attached a screenshot show where this occurs. I think this would be achieved like this: 1. Specify tag when creating instance 2. When EC2 instance boots, discover instance ID (e.g. using <[login to view URL]>) 3. Use a tool such as ec2-describe-tags to determine tags (I think this is in AWS or EC2 command line tools)