I’ve got Dude monitoring a large network over 3 continents and I need to make some changes.
The way things are now, I’ve got Dude running on a ProLiant server located at the BT facility from where our satellite service is uplinked. It obviously monitors the servers and routers in the UK and the USA, but it also monitors the routers and networks on the “other end” of the satellite links in places like Mozambique, Botswana, Zimbabwe all the way up to the UAE. It also monitors servers and networks in South Africa and Kenya that are connected via undersea cable.
Result is that I am wasting quite a bit of money on bandwidth for what could basically be thought of as “internet noise”.
Both satellite and undersea bandwidth are expensive, so I’ve been thinking of setting up a distributed monitoring system.
The plan is to replace the ProLiant server in the UK with an RB600 and to dump some RB750s all over the place and then to get them to also run Dude, interrogate what’s in their immediate vicinity and then send that information back to the RB600.
Are there any wikis or tutorials on how I can set about doing this?
COOL you have what sounds to be a very interesting network.
It is pretty straight forward, add a new dude server somewhere, allow access to it through the firewall from the main server if you need to…
On the core server click on settings, agents tab, + add agent, give it a name, plug in the IP and that agent should show up on the map.
Since I have not personally done this could someone chime in on any other things to be done?
Does adding putting Agents in the middle of each subnet(or physical location) actually reduce traffic? Or does it aggregate the same amount of traffic, instead of x amount of traffic originating from 20 devices, is it 20x traffic from one device.