Friday, November 30, 2018

Breaking RFC 1519

I have a problem and I'm hoping someone has some advice for how to best deal with a vendor. Their situation bothers me.

So a vendor installed a network for a new initiative in what is a more or less shared space. This system required networking, as you would expect, they needed a couple of things, provided a couple of things, etc. So, long story short: their gear: gateway/router/firewall ... The gear we provided: switches.

We chose stacked multilayer switches because of the requirements they set forward. We gave them admin access, to all the things, so they could round out the config for final implementation.

So yesterday, I went in to see how they're doing with it, and get some minor training on supporting their equipment (I'll help with onsite work since the vendors location is pretty far away).

I found out that they're setting up Windows on a /24 network, x.x.101.y (where x is always the same across all network subnets) and they're setting the default gateway to x.x.100.1

I don't even understand how that works, at all, or why Windows would allow any communication to happen. They're not using vlans, so their x.x.100.y gear is on the same L2 domain as their x.x.101.y, but as far as I can tell, everything is set to /24.

This hurts me a little bit, but for some reason, it works. Communication happens.

I have not ever been witness to a network breaking CIDR boundaries like this. What is happening? Why does this work? What are the pitfalls here?

Obviously I just want to claw my eyes out and re-arrange the network into vlans and set up the L3 switch to route everything correctly, however, I have no access to set up routes in their gateways, so I'm pretty stuck.

Is this worth pursuing? I mean, for the purposes of shielding my client from a bad network design? Or will the problems be minimal and I shouldn't worry about it.

Thanks.



No comments:

Post a Comment