Discussion in 'Tomato Firmware' started by jsmiddleton4, Nov 14, 2011.
What does this setting do?
Not really, although google turns this up. It would be useful to me to be able to issue a second gateway to clients. Or maybe this isn't what it does.
Enabling slows my WDS clients throughput down. Adds a noticeable lag. If we don't know what it does and don't have any information on how and when to use it, why is it there?
I've looked as well and I have found nothing that gives instructions on when, what kind of network setups, etc., to use it for.
Teddy Bear added it, the code has been tidied up several times so one assumes it is stable by now. But we can't ask him how to use it at the moment, unfortunately. Maybe someone else will pop up and tell us? I have a second gateway on most of my buildings, it would be nice to use this to notify clients of that, if that's what it's for, instead of assigning individually by Static DHCP. But without any instructions as to how to use it, I can't make it do anything noticeable, but will experiment a bit later. Maybe something will happen in logs ... whatever.
The source code does refer to static classless routes, so this is on the right track. I expect a good session with google would reveal how to use it, if only life wasn't so short
But where do you configure the second gateway information and which router gets this selected? Just the gateway router that is attached to the internet?
Need more input.
I described it (what I use presently) here:
Well when having multiple VLANs in different ports, I had to disable DHCP Routes to make it work properly for Tomato Shibby's MultiWAN, too many options not much documentation...
So does anyone know the answer to the OP?
bump for more info
It seems that RIPv1 and RIPv2 are a type of overhead. It plans where the data packet is now, where it is going, how big is the domain place is of where it is going, and how many "trips/stops" is it going to take along the way. The way i see it, RIPv1 or RIPv2 will ultimately overrride the (R1) internal IP/external IP to the other (R2) internal IP/external IP. What this may mean will be it won't matter what you call wan IP or lan IP, the packer is "smart enough" to know where it is going. This is kinda big for gateway to gateway connection of R1 and R2. No more using tomato routers as a bunch of cheap switches. One question remains though, why is it that RIPv1 and RIPv2 only available for lan1 or wan1. Seems this is moving towards multiwan in some way shape or form, but keep in mind this big, big, big HUGE problem. It's one thing to make a "smart" network packet of information, it's another to make a "multi-router server ready" to handle it.L ast note: for every wan (or ISP), you'll need one tomato router to handle its ROUTING IP TABLES. That means, vpn, bit torrent, adblocking, and all the SECs. This is one f-bomb of a system we are talking about. Which router or "multi-router server ready" networking machine is going to be ready for this shiet?
*Note: suck it, koitsu!
**Double-note: probably want to look at how multi-wan combines packets? Maybe even TTL will have to be longer and then, what happens when network packets have longer lifespans? Now you got "gangs" of network packets. How would to combine them? How would you separate them?
***Triple-note: will there be a zero tolerance rule?