Skip to content

Latest commit

 

History

History
134 lines (106 loc) · 7.36 KB

README.md

File metadata and controls

134 lines (106 loc) · 7.36 KB
Raw

Lighthouse Auditing Bot

Summary

This project creates a containerized auditing environment for Google's Lighthouse tool. Bringing all the benefits of running your tests in a stable environment without the overhead of updating CI/CD pipelines and code

Features

  • Ad-hoc Auditing - Quickly run an audit of a website with the /lighthouse {url} command, or simply type /lighthouse to launch a dialog with all available options
  • Job Scheduling - With the /lighthouse jobs command, you can schedule an auditing job to be run whenever necessary
  • Customizeable HTML Reports - Always be able to view the full detailed report from Lighthouse as an HTML file, which is provided by a template in this project, and customize parts of the template (in src/static/reportTemplate.html) to your heart's content!
  • Trend Charts - Track changes in each of the audit categories over time for a given URL by running the /lighthouse stats {url} command and clicking the link to an intuitive dashboard (also provided as an HTML template that can be customized in src/static/statsTemplate.html)

Development

Pre-requisites & Notes

  • Node.js v12+

Regardless of the method you are deploying with, this application relies on a variety of environment variables to be able to function properly. Either use the export method, or inject your docker container with env variables depending on what method you are deploying this chatbot with.

Variable Name Example Value Explanation
PORT 3001 The port being used by this chatbot
TOKEN xoxb-921212312-125361390560-xxxxxxxxxx The OAuth token value received after installing the Slack App
MONGO_USERNAME root Auth username for a mongodb server
MONGO_PASSWORD test_passwd Auth password for a mongodb server
MONGO_SERVER 192.168.1.10:27017 The endpoint for a mongodb server
CHATBOT_SERVER http://192.168.1.10:3001 IP to be used by this chatbot (needed to set URL endpoints in Message Attachments)
TZ (optional) Asia/Seoul The timezone value that will be used on server (important for job scheduling)

Developing with Docker

0. Build local Lighthouse bot image

docker build -t wanteddev/lighthouse-bot .

1. Run mongodb (as a separate container)

docker run -d -p 27017:27017 -e MONGO_INITDB_ROOT_USERNAME="root" -e MONGO_INITDB_ROOT_PASSWORD="test_passwd" --name lighthouse-mongo mongo:latest

2. Run Lighthouse bot container
Note: Ensure you have the environment variables set when running the container

Example run command:

docker run -d -p 3001:3001 -v $PWD/src:/home/app/src -e TZ="Asia/Seoul" -e PORT=3001 -e TOKEN="xoxb-921212312-125361390560-xxxxxxxxxx" -e MONGO_USERNAME="root" -e MONGO_PASSWORD="test_passwd" -e MONGO_SERVER="192.168.1.129:27017" -e CHATBOT_SERVER="http://192.168.1.129:3001" --name lighthouse-bot wanteddev/lighthouse-bot

Developing with PM2

0. Follow the installation guide for MongoDB to set up your MongoDB instance

1. Set values for all required environment variables

export PORT=3001
export TOKEN=xoxb-921212312-125361390560-xxxxxxxxxx
export MONGO_USERNAME=root
export MONGO_PASSWORD=test_passwd
export MONGO_SERVER=192.168.1.10:27017
export CHATBOT_SERVER=http://192.168.1.10:3001
export TZ=Asia/Seoul

2. Globally install PM2

yarn global add pm2

3. Install dependencies

yarn

4. Run chatbot with pm2

pm2 start ecosystem.config.js

5. Register a slash command in Mattermost that sends a GET request to the /lighthouse endpoint

Deployment

Deploying this chatbot is done in the same way as the development environment setup, with the exception that you would set the NODE_ENV variable to production, as well as not do any volume binding to the host when running the chatbot with Docker.

Slack App registration

As we're open-sourcing this project, the responsibility to deploy and manage each application and data stored in it falls upon whoever ultimately deploys it. We're sharing the configurations used in our private Slack App so that others may reference it and apply it to their own auditing environments!

Interactive Components

In order to use the inputs and options inside of the custom auditing functionality, you need to activate the "Interactive Components" functionality, and set the Request URL value as below: Request URL: https://${chatbotUrl}/receive_submission

Slash Commands

As the main method of accessing the chatbot server, we have registered a Slash command as follows:

  • Command: /lighthouse
  • Request URL: https://${chatbotUrl}/lighthouse
  • Short Description: Use Lighthouse Audit Commands
  • Usage Hint: help

Bots

Some of the settings we've activated for our chatbot include:

  • Always Show My Bot as Online
  • Show Tabs > Messages Tab (So that you can run audits in a DM)

Permissions

In order to post messages, this chatbot requires the following 3 permissions:

  • chat:write: Basic messaging permissions
  • chat:write.public: Ability to post messages in channels the bot is not a member of
  • commands: Ability to add slash commands that people can use

Troubleshooting

  • The audit command fails with an error while loading shared libraries: libX11-xcb.so.1: cannot open shared object file: No such file or directory error on Ubuntu

    • Run the command below to install dependencies needed to launch Puppeteer from your host
    sudo apt install -y gconf-service libasound2 libatk1.0-0 libc6 libcairo2 libcups2 libdbus-1-3 libexpat1 libfontconfig1 libgcc1 libgconf-2-4 libgdk-pixbuf2.0-0 libglib2.0-0 libgtk-3-0 libnspr4 libpango-1.0-0 libpangocairo-1.0-0 libstdc++6 libx11-6 libx11-xcb1 libxcb1 libxcomposite1 libxcursor1 libxdamage1 libxext6 libxfixes3 libxi6 libxrandr2 libxrender1 libxss1 libxtst6 ca-certificates fonts-liberation libappindicator1 libnss3 lsb-release xdg-utils wget libgbm-dev

Current Tasks

  • Implementing trend charts for audit results to be accessed with /lighthouse stats {url}
  • Styling audit trends dashboard
  • Add usernames to schedule schemas so that they can be easily viewed through /lighthouse schedule list
  • Implementing /lighthouse schedule info {id} to get full details of a given job
  • Add unit testing with Jest or AVA
  • Write documentation to make command usage clearer
  • Add more comprehensive logging
  • Investigate the possibility of using workers to run audits so that multiple audits can run simultaneously
  • Add a configurable number of past audits to be fetched from stats command
    • e.g. /lighthouse stats https://google.com limit 15

Blog