StatiX

ISP network statistics and monitoring dashboard

An R/Shiny webapp that shows graphics for an ISP. It includes:

1. Graphics for connected PPPoE users in the entire network
2. Graphics for connected users split by service
3. Graphics for inbound/outbound traffic in the Area Border Routers
4. Graphics for HotSpot users
5. A search tool to find issues of HotSpot users: where the user is located and the stats of the connection.

You need MongoDB on localhost:27017, MySQL access to the RADIUS server, and R with shiny, RMongo (from the CRAN archive), RMySQL, dplyr, ggplot2, scales, stringr and rjson. The Traffic and HotSpot tabs need more; Getting started lists it. The collector writes listaMKT.csv to /home/tecnico/EstadisticasWEB, where the app reads it, so clone the repo there or edit scriptServicioBBDD.R line 20. Set the RADIUS host and password in the collector. The collector loops forever, so start it in the background, then the web app:

Rscript scriptServicioBBDD.R &
Rscript -e 'shiny::runApp(".", port = 8081, host = "127.0.0.1")'

Open http://127.0.0.1:8081. The AtStartup*.sh scripts show which collector feeds which tab.

The docs are at geiserx.github.io/statix.

• Getting started: what it needs, the archived R packages, the first run, starting at boot
• Usage: the five tabs and what each chart shows
• Configuration: every setting, file by file and line by line
• How it works: the collectors, what MongoDB stores, the security model
• Troubleshooting: blank charts, empty lists, wrong traffic points, and what to put in an issue

• genieacs-container: Helm chart and container for GenieACS TR-069
• router-express: auto-configures client routers and syncs databases
• services-isp: automates common ISP operational tasks
• ScriptPoblar: adopts a whole network of devices into CRM Control in parallel

GPL-3.0-or-later