ClassHelper

Development

• Web App
  • Fork the repo.
  • Clone from your fork
  • npm install
  • bower install
• Database
  • The schema is in /server/config/schema.sql
  • Install mysql (or brew install mysql) and start the mysql service with mysql.server start
  • Import the schema with the command mysql -u root -p < server/config/schema.sql
  • By default, all users initialized with the schema file have password 'test'.
• Running the app
  • Rename start.sh.example to start.sh and edit it to use the username and password for your mysql database.
  • Run chmod +x start.sh to make the start.sh file executable.
  • Run npm start to start the server with nodemon.
• Documentation
  • Documentation is built with YUIDoc
  • Install yuidoc with npm install -g yuidocjs
  • Generate the documentation with yuidoc .
  • The documentation will reside in the './out' folder and can be served as static html files.

Directory Layout and Description of Files

• client/
  • app/
    • attendance
    • auth/ -- Login and Signup views
    • grades/
    • landing_page
    • services
  • lib/
  • style/ index.html
• node_modules/
• server/
  • config/
    • config.js -- Configuration for variables that aren't handled by Environment Variables. Environment variables should be used first. If they don't exist, the app falls back to config.js.
    • db.js -- Configuration for knex/mysql/bookshelf. Uses environment variables defined from the start.sh which is run with the npm start command.
    • express.js -- Required from server/server.js and sets up our express app and imports our routes.
    • schema.sql -- Seed the initial database with mysql -u username -p password < server/config/schema.sql.
  • controllers/
    • ... -- Each controller has its own model and route in their respective folders.
  • models/
    • ... -- Each model has its own controller and route in their respective folders.
  • routes/
    • ... -- Each route has its own controller and model in their respective folders.
  • views/ server.js
• tests/
  • server
    • controllers
      • ... -- Each controller has its own test file.
    • models
      • ... -- Each model has its own test file.
    • routes
      • .. -- Each route has its own test file.
• .bowerrc -- Configure bower install to install to the client/lib directory.
• .dockerignore -- Ignore the node_modules folder when a docker file runs the ADD or COPY command. The primary reason for this is that we need to do a fresh npm install when building docker images. Some modules install differently between Mac and Linux (bcrypt specifically). It can be the source of bugs causing great frustration and fury.
• .gitignore -- Ignore config, tmp, and lib files.
• bower.json -- Installs to client/lib due to .bowerrc config.
• CONTRIBUTING.md
• docker_build.sh -- Builds docker images from scratch. This should be run the first time you start using docker to deploy/develop the app and again each time you want to push your local changes to the docker containers. Develop locally. Run docker_build.sh when you are ready to test your changes on docker and push to deployment.
• docker_start.sh -- Restarts the containers. Use this to start the containers again after a reboot or any time they stop/crash. No changes you make to your local filesystem will be reflected by running this command. To reflect changes, run docker_build.sh.
• mysql_build.sh -- docker_build.sh calls mysql_build.sh to build the MySQL image. You can build the MySQL image in isolation by running ./mysql_build.sh.
• mysql_Dockerfile -- Copies the schema file from server/config/schema.sql.
• msyql_start.sh -- This runs the mysql container as a background process with the -d flag. It maps port 3306 on the container to 3306 on the host machine. It gives the container a name of classroom-db. It uses the classroom/mysql:v1 image which was created with the docker_build.sh command.
• node_build.sh -- docker_build.sh calls node_build.sh to build the Node image. You can run ./node_build.sh to build the Node image in isolation by running ./node_build.sh.
• node_Dockerfile -- Copies the entire app directory to the /greenfield folder in the docker image. It also sets environment variables, the working directory, and installs nodemon, bower, and the required packages.
• node_start. - Runs the node container as a background process, forwards necessary ports, and links to the database container. No changes made to the local filesystem will show when you run this command. To reflect changes, re-run docker_build.sh.
• package.json
• PRESS-RELEASE.md
• README.md
• start.sh -- This is the script run by npm start. It sources from config.js to import environment variables, then it runs nodemon server/server.js.
• STYLE-GUIDE.md

Deployment

Docker

• If you're using Mac, you'll have to install boot2docker and docker: https://docs.docker.com/installation/mac/

• Download the Boot2Docker-x.x.x.pkg file.

• Install Boot2Docker by double-clicking the package.

  • The installer places Boot2Docker and VirtualBox in your "Applications" folder.
  • The docker and boot2docker binaries will be in your /usr/local/bin directory.

• To run docker from the command line

  1. Create a new Boot2Docker virtual machine
  • boot2docker init
  2. Start the boot2docker VM.
  • boot2docker start
  3. Display the environment variables for the Docker client.
  • boot2docker shellinit
  4. Set the environment variables in your shell with the following:
  • $(boot2docker shellinit)
  5. Run the hello-world container to verify your setup.
  • docker run hello-world

• Every time you open a new terminal window, you will need to set the environment variables of the new window by running the $(boot2docker shellinit) command.

• Scripts have been created to handle docker instance setup.

  • docker_build.sh (Necessary on the first build and every time you want to push changes you make locally to the docker containers).
  • docker_start.sh (Necessary every time you want to spin up a container of the image you made with docker_build).
  • node_build.sh (Used by docker_build.sh. Creates a new node docker image).
  • node_start.sh (Only necessary if you want to spin up the node container by itself for testing purposes or if it crashed).
  • mysql_build.sh (Used by docker_build.sh. Creates a new mysql docker image).
  • mysql_start.sh (Only necessary if you want to spin up the mysql container by itself for testing purposes or if it crashed).

• Once the containers are up and running, you should be able to access them at 192.168.59.103:3000 on your local machine.

Problems you might run into during deployment/development

• [ERROR] InnoDB: Cannot allocate memory for the buffer pool
  • You might receive this error while trying to start the MySQL server on a VPS with limited resourcs (such as a $5 DigitalOcean droplet.
  • This is caused by not having enough memory. The solution https://www.digitalocean.com/community/tutorials/how-to-add-swap-on-ubuntu-12-04
• MySQL ERROR! The server quit without updating PID file...
  • Seems to be fixed with a reboot.
  • This could be because of an edit to your my.cnf file that MySQL is unhappy with.
  • It could also be for some other reason I haven't been able to narrow down.
• ...dial unix /var/run/docker.sock: no such file or directory. Are you trying to connect to a TLS-enabled daemon without TLS?...
  • Started getting this after a boot2docker upgrade.
  • After running $(boot2docker shellinit) I got a different error: An error occurred trying to connect: Get https://192.168.59.103:2376/v1.19/containers/json: x509: certificate is valid for 127.0.0.1, 10.0.2.15, not 192.168.59.103
  • Fixed with restarting the docker service inside boot2docker -- ``boot2docker ssh 'sudo /etc/init.d/docker restart'

Team

Product Owner: Richard Stanley Scrum Lord: Eric Ihli Product Team: Jake Lee, Devon Harvey