FlowTrack2

This is my rewrite of FlowTrack in node and elastic search.

Why?

• Want to learn something new
• The time bucketing code was annoying to write, elastic search seems to do it automatically.
• scale. This will be able to handle more than sqlite.

Progress

Take a look at the DevLog to see more detail

• Main collection loop works
  • Flows are captured and written to elasticsearch db
• Setup a grunt scaffold
  • seems to be the right way to go
• Created flow storage lib
  • I think it's created cleanly, I'm sure I'll notice stuff as I learn more.
• Unit testing & Code Coverage setup added
  • Not super happy with it, want to look at something other than istanbul
• Unit tests are a little cleaner now.
• Start web UI
• Added config file system
• Add table view of flows to web UI

Next up (in no particular order)

• Performance testing
  • I want to see how fast this will actually go
  • Using a single node for ES

Install

• Have a running elastic search instance
  • Currently assumes localhost:9200
• have a working node + npm
• sends bunyan logs to stdout, no way to turn off currently.
• Webserver is running on port 3000

git clone https://github.com/skarfacegc/FlowTrack2.git
cd FlowTrack2.git
npm install
bin/flowTrack