Configuration and monitoring of WR networks

Dear Rabbits,

I would like to start a discussion about your current ways to configure and monitor your WR networks. I would like to know how do you do that and how can it be improved.

As today WR switches can be configured by:
–web interface
–ssh & make menuconfig
–download configuration form remote server (via tftp, ftp, http, etc)
Two first solutions does not scale at all. For the 3rd one, it would be nice to have something to manage configuration files on the remote server (at CERN we have such solution, but as today it is not exportable).

For monitoring we recommend to use SNMP. However there are some other possibilities that I can imagine like ssh+wrs tools or webinterface. IMHO: The latter two does not scale well.

At CERN for monitoring we use:
–Kibana for handling syslog messages
–icinga+Graphana for monitoring via SNMP

Adam Wujek

A plugin/script that would fetch relevant data over SNMP and store it e.g. into influxDB would be useful. InfluxDB has a ‘telegraf’ component/daemon for collecting data, and this could be one plugin for that?
We would probably use such a script if it was developed as a joint effort on e.g. ohwr.

There are ofcourse n+1 alternative databases/monitoring-systems also…

In LHAASO, we use Zabbix for monitoring via SNMP. It is open source and scalable. It’s also easy to implement new WR device and analysis/export historical data.

At GSI/FAIR for monitoring we use :

  • Icinga via SNMP and Python scripts
  • for syslog messages we use a server with no special software
  • messages from the timing clients we use Graylog

marcus zweig