ts-7000
[Top] [All Lists]

RE: [ts-7000] RE: ts-7800 how do I make 4 (at least 2) identical network

To:
Subject: RE: [ts-7000] RE: ts-7800 how do I make 4 (at least 2) identical network stacks?
From: Paul Yanzick <>
Date: Mon, 4 Nov 2013 12:13:50 -0600


Perhaps I am not familiar with the way the 8700 works… however if it acts as a multihomed unit, you will have issues.  if you use 1.1.1.xxx and plan on using the same on all , they ARE the same network from an IP perspective.  They would all be 1.1.1.0/24. 

 

From: [ On Behalf Of Jonathan Leslie
Sent: Monday, November 4, 2013 12:11 PM
To:
Subject: Re: [ts-7000] RE: ts-7800 how do I make 4 (at least 2) identical network stacks?

 

 

What do you mean about "4 NICs… all with different IP’s on the same network. " ?   I want 4 DIFFERENT networks. All 4 networks are 1.1.1.xxx.   The TS-8700 maintains  itself as 4 completely different nodes on all 4 networks, but it has the same ip address (1.1.1.200 for example.)  I want 4 separate programs (or threads)  all thinking they are node 1.1.1.200, but all on different networks.   I never have one node 1.1.1.101 talking to another node 1.1.1.101 on a different network. each device 1.1.1.101 talks on its own separate network to a node 1.1.1.200 on its network. 

 

 

 

 




From: Paul Yanzick <m("hotmail.com","yanzickp");">>
To:
m("yahoogroups.com","ts-7000");">
Sent: Monday, November 4, 2013 11:28 AM
Subject: RE: [ts-7000] RE: ts-7800 how do I make 4 (at least 2) identical network stacks?

 

 

Here is the thing that I think will be the problem…

 

  1. 4 NICs… all with different IP’s on the same network. 4 legacy devices all with the same IP…  how is it going to route?

 

Lets say, for example, the IP’s are 10.1.1.1, .2, .3, and .4 for the 4 NICs, and we will assume subnetted to 24 bits.  Lets then also say that the IP’s of the legacy devices are all 10.1.1.10.  The legacy devices can talk to the 8700 without issue… getting the message back could be the problem.  The routing table is going to see that it is on the same network, so it doesn’t have to hit a router… but then which NIC will it send a reply back to?  By detault the route will show routes for the 10.1.1.x network either from 1 network adapter, or all of them (haven’t tried, so can’t say).  If it shows the route for all of them… how will it know which one to send it to?  It will  use the routing table to determine what NIC to send the packet out of.  It isn’t going to just send a reply back out of the NIC that it received the message from.  You can’t play tricks with more specific subnetting either since all of the legacy devices are on the same IP. 

 

There may be some magic that can be done… I haven’t ever tried it… but it seems what this would be doing would be breaking fundamental routing rules, so not sure how easily that will be done. 

 

Another option (requiring more hardware for sure, but more feasible) would be to insert something else between the legacy devices and the 8700 which performs some NATting.  Then you can have each device on a different network, allowing the 8700 to manage communication as it would for any other multihomed device.

 

I wish you the best of luck doing it though!  Interesting problem.

 

 

 

 I think the salient issue is that the four legacy boxes need to be on separate networks. Each of the ethernet interfaces can be a separate network. The four interfaces can be given four different addresses. The custom application (or any application at all; the magic will be done in the kernel with the iptables rules) talks to IP addresses which get trapped by rules which NAT/route the destination addresses to a specific interface. Any application can then talk to a specific IP address, but the kernel will route the traffic to a 10.1.1.xxx address on the correct interface. This is conceptually similar to using Linux to create a LAN-WAN router with port forwarding, except in this case, the WAN side is internal to the Linux host, and there are multiple LAN interfaces.

Totally do-able, but will take some fiddling to get the right rules.

 

You can be a pro in some areas but not others. The guy admits his experience is only at the socket level. That's not enough for what he wants to do. Its not silly to be ignorant, it is silly to stay that way. BTW he's been given a lot of bad advice on this forum.
He will either have to hire help, or do a lot of work.
--------------------------------------------

On Mon, 11/4/13, Petr Štetiar <m("...","ynezz");" target="_blank">> wrote:

Subject: Re: [ts-7000] RE: ts-7800 how do I make 4 (at least 2) identical network stacks?
To:
m("yahoogroups.com","ts-7000");" target="_blank">
Date: Monday, November 4, 2013, 2:14 AM

Jonathan Leslie <
m("...","jleslie48");" target="_blank">>
[2013-11-03 06:30:51]:

> I'm confused,  I have to send messages to  two
different devices, both have
> address 1.1.1.101. The way I've done it in the past was
to set up two
> different IP stacks, with different SOCK.  I want to
repeat this on a
> TS-7800 only this time I need 4 different IP stacks as
I have 4 different
> devices 1.1.1.101.

What's so confusing in "Buy TCP/IP book and read it" ?

I don't know what a SOCK is neither I know what you've done
in the past.

What I know is, that you're kind of a silly guy, which is
calling himself a
Pro, but don't know even basic networking stuff. And as a
bonus point, you're
going to use USB ethernet in production...

Pro would buy OpenWrt capable 4-5 port router and would do
NAT with iptables.

-- ynezz


------------------------------------

Yahoo Groups Links


   
m("yahoogroups.com","ts-7000-fullfeatured");" target="_blank">

 



__._,_.___


Your email settings: Individual Email|Traditional
Change settings via the Web (Yahoo! ID required)
Change settings via email: =Email Delivery: Digest | m("yahoogroups.com?subject","ts-7000-fullfeatured");=Change Delivery Format: Fully Featured">Switch to Fully Featured
Visit Your Group | Yahoo! Groups Terms of Use | =Unsubscribe

__,_._,___
<Prev in Thread] Current Thread [Next in Thread>
Admin

Disclaimer: Neither Andrew Taylor nor the University of NSW School of Computer and Engineering take any responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU