A project to extract, persist and serve data from a router's state.
This program uses a router user interface to read status and statistics data from the router, saving updates and serving it as a JSON API.
A long-running background process maintains the session and watches variables for changes. Updates are saved in JSONL files separated by source, type, name, and hour (./data/{source}-{type}/{name/{yyyy-mm-dd-hh}.json).
The extraction requires a authorized connection (session) that is loaded from the ./data/session-id.txt file or the --session argument.
The program will also perform authentication if credentials are provided: They can be specified by the --user and --pass program arguments or loaded from the ROUTER_USERNAME and ROUTER_PASSWORD environment variables. A sample file at .env.sample can be filled and renamed to .env too.
Download with git and start it with node.js, the project does not depend on any project to work:
git clone https://github.com/GuilhermeRossato/router-vis.git
cd router-vis
node index.js
You can execute the main script with either npm run start, yarn start or node index.js.
The initial setup will initialize the extraction server and begin streaming the network speed information. The extraction is done in a separated process so you can stop the script (with Ctrl+C or Ctrl+D) and it will continue persiting variable updates.
The background process can be disabled by including the --standalone flag argument or by configuring it as a default at ./settings.js.
--speed / --kbps / --mbps Stream the differences of total usage data.
--usage / --kb / --mb / --gb Stream the total usage data as kilobytes, megabytes or gigabytes.
--shutdown / --stop Stop the background extraction process (if it is running).
--restart / --start / --init Restart the background extraction server (stopping it first if needed).
--status / --info / -s Request the current background server status and display it.
--api / --url / --open Display and open the browser at the api url.
--standalone / --direct Prevents the execution of the background process and start extraction.
--logs Stream the extraction server logs continuously.
--debug / -d Print extra execution logs.
--user <user> Specifies the username to use on authentication.
--pass <pass> Specifies the password to use.
--session <code> Specifies a session id to use when extracting.
--server Executes the script as the background server.
--help / -h Display a help text with a link to this repository.
Default arguments can be configured at the root file ./settings.js.
After executing the program once you can access the API by default at http://127.0.0.1:49736/.
You can use the --api / --open argument to launch your browser to it. The host and port of the internal server can be replaced by setting the INTERNAL_DATA_SERVER_HOST and INTERNAL_DATA_SERVER_PORT environment variables.
The API server routes can be listed at the index endpoint. To get server status you can use the status endpoint.
To read data from variables you can use the data endpoint. If you define the type and the src variables it will list the variable names available. The response object will contain the acceptable (and alternatives) values at the options property.
For example, to read the latest usage data you can specify the type, src and name as object, statistics, eth-intf-sts at the data endpoint. You can also specify the section parameter to get updates of different interval. The section property at the response (inside options property) will contains the alternative values.
To read the latest extraction server logs you can use the logs endpoint.
The primary objective of this project is to experiment with project organization while learning some data-related skills.
Modern user-customer routers provides useful but instantaneous information such as fiber optical signal stregth, port routing, wifi status, connected hosts, wifi usage, uploaded bytes, etc. I wanted to analyze this data over time which is not available because customer-grade routers are hardware limited.
I reverse-engineered the HTTP interface of my router and implemented a process to authenticate, extract and save its data and its updates. The data can then be queried from a internal JSON API maintained by a background process.
This project is executed with Node.js without any external or third-party dependencies so executing npm install is not necessary.
The project is intended to be used with a specific router software named Vivo Box BR_SV_g000_R3505VWN1001_s42. The router model code is RTF3505VW-N2.
The router's interface dashboard looks like this:
