quagga networking suite

kent_dorfman766

kent_dorfman766

Anybody on here have any experience with the quagga routing software suite? It implements a whole slew of internet routing protocols, but looks like it has been innactive for many years (most links are broken). Anyone know of recent/authoritative resources related to it?

I have a customer who is looking at delay-disruption tolerant networking, and routing (supporting non-terrestrial networks) is a primary problem in that scenario. Looks like the NASA implementation has done (some) of the heavy lifting by implementing CGR (call graph routing) but to meet client needs dynamic route configuraiton will be a necessity, thus needing to investigate guagga...

now that everyone is all glassy eyed and confused...sorry.
 
I think it was originally released as zebra. I did use it mainly for OSPF, then project changed its name to quagga (or, maybe, it was a fork, I am not sure).

Last time I needed something similar in FreeBSD, I found there is new port - net/frr(8|9|10), and in pkg-descr, you see...

FRRouting (FRR) is an IP routing protocol suite for Linux and Unix platforms
which includes protocol daemons for BGP, IS-IS, OSPF and RIP. FRR has its roots
in the Quagga project.

So I think you could find something newer here.
 
what protocols from the quagga suite are you looking for?

quagga is dead, just as the extinct zebra it is named after. But you might want to take a look at the already mentioned FRR, which is the last surviving fork of zebra/quagga, or net/bird3, which is an implementation with a very different approach (i.e. a proper config file...). IMHO quagga/bird/frr is just absolutely weird to configure/manage as you have to do everything through its (IOS-like) interactive shell. I'm a cisco-guy when it comes to networking, but on a real server this approach just feels wrong on so many levels...

For only BGP there's also net/openbgpd8. I've been using it for years and love it for its great configuration syntax (comparable do PF syntax). I simply use BGP for internal and external routing, so I don't need a full suite of routing protocols, but from the description of your use case I'm not sure if BGP would fit - at least not without making a lot of adjustments and compromises...
 
So for better disclosure, I've been asked to research a delay/disruption tolerant networking toolset based on RFC9171 (bundle protocol) to allow packets/bundles to move through heterogenous networks using store-and-forward, but opportunistically and only when availability of nexthop is present. The fly in the ointment is routing, since it has to cover heterogenous network topologies and has to be somewhat dynamic and self-managing. So far I'm impressed with the work that has gone into BP, but routing is a black hole right now since BP endpoints are different than the terrestrial IP scheme, and it is not REQUIRED that the underlying convergence layers be IP based.

If BP actually catches on then I'd envision the BP stack existing as OS kernel layer drivers along side the IP stack and using the BSD sockets interface for access. Right now everthing is done in userspace daemons.
 
sko said:
IMHO quagga/bird/frr is just absolutely weird to configure/manage as you have to do everything through its (IOS-like) interactive shell. I'm a cisco-guy when it comes to networking, but on a real server this approach just feels wrong on so many levels...
Click to expand...
That's incorrect. I generate bgpd.conf for approximately 250 endpoints. It's just a text file.
(And personally I like the shell because it's exactly like Cisco, so nothing new to learn & break.)
 
You must log in or register to reply here.
Back
Top