Hi!
I've a server with two NICs each connected to a seperate LAN.
For example:
eth0: 192.168.0.0/24
eth1: 10.0.0.0/24
At the moment i've only set up ONE default getway like this:
0.0.0.0 10.0.0.1 0.0.0.0 UG 0 0 0 eth1
But when now a paket comes in on eth0 from for example 192.168.10.5 it tries to route it back through the default getway of eth1. But i want to archieve, that it routes it through another defaultgateway for eth0.
So traffic from eth0 should go to the defaultgw for eth0 and traffic from eth1 should go through the default gateway of eth1. Is this possible? Any other ideas? I don't want to setup an ip for each LAN on eth0.
Thanks!
Stefan
This Question has been solved and asker verified All Experts Exchange premium technology solutions are available to subscription members.
Experts Exchange has been collecting answers to technology questions since 1996…3 million and counting! If you have a question, chances are we already have your answer.
If you can't find the exact answer you're looking for, ask our exclusive community of 50,000 experts. You’ll get a personalized answer from a trusted professional.
Thousands of free tech tips, tricks, how-to’s and tutorials are available in our peer reviewed articles section. See for yourself how smart our experts are, no login required.
Save solutions to your questions, answers you’ve discovered through searching plus helpful articles in your personal knowledgebase for easy future access.
Access the answers to your technology questions today.
30-day free trial. Register in 60 seconds.
Members of the expert community talk about why the experience at Experts Exchange is different than what you will find anywhere else.
Try it out and discover for yourself.
30-day free trial. Register in 60 seconds.
Join the community of experts here and help other tech pros by answering question in your area of expertise. You can earn FREE access to all Experts Exchange's premium features and resources.
Found the solution myself but in german on www.ubuntuusers.de
Hi,
I guess you must be mistaken. In a system with 2 ethernet adapters the first will be in the 10.0.0.0/24 segment while the second will be in 192.16.0.0/24 segment. When you want to communicate with a system located in the 192.168.0.0 segment the communication will be performed over ethernet. Because the local interface si tattached to the 192.168.0.0 segment and it will discover the system on the same subnet via arp and will directly communicate with it. So it will never use eth1 for a response.
This is being said if you want two default gateways with the system you need to setup metric so that system knows which to use for internet access
you need to add:
route add -net 192.168.0.0 netmask 255.255.0.0 gw 192.168.0.1 metric 10
Cheers,
K.
I don't see how advanced routing works better than assigning static routes to a particular interface by destination. Using the iproute feature to have two 0.0.0.0/0 routes will make your outbound packet flow even less predictable, as the new routing table will probobally load share based on flows, and traffic will not go to a predictable interface. If you have more than 10 subnets that you are trying to route to, then you may well be better off using a routing ptotocol such as RIP or OSPF, but this is not "two default networks" by any means, and it implies that the routers on the network are exchanging routing information with the host.
Please post what your final configurations look like though, I am interested.
You should look into iproute2. Basically you need to create multiple routing tables - then bind them to an eth interface. Here is an example of a script I used to do this on a box I have long since retired....
ip route add table T1 192.168.0.0/24 dev eth0 src 192.168.0.50
ip route add table T1 default via 192.168.0.1 dev eth0
ip route add table T2 10.0.0.0/24 dev eth1 src 10.0.0.50
ip route add table T2 default via 10.0.0.1 dev eth1
ip rule add from 192.168.0.50 table T1
ip rule add from 10.0.0.50 table T2
ip route flush cache
The problem is:
Your adapters are these:
eth0: 192.168.0.0/24
eth1: 10.0.0.0/24
So since 192.168.10.x is out of the scope of both your interfaces it is obvious that your request to access 192.168.10.x will go to your default route. If you have an internal router that has an access 10 192.168.10.x network then you'll need to specify it through :
route add -net 192.168.10.0 netmask 255.255.255.0 gw 192.168.0.2
Otherwise the packets will go through you default network. As I've told earlier the Default Gateway is used whenever your host does not know how to access locally through interfaces. Though it is rather in your 192.168 side it does not mean that there's an automatic way for your sytem to discover that. So you need to specify iy manually. On the example above I am assuming that 192.168.0.2 is the router who has access to the 192.168.10.x network.
Cheers,
K.
Hi,
The problem is that: The asker expects something fancy as an answer. But the truth is he is expecting something unusual and what happening is quite the normal for TCP/IP if he needs to send 192.168 traffic to somewhere outside of the default route all he needs is to add a routing entry. I believe I've tried to tell him about this situation in my postings.
Cheers,
K.
KeremeE - I think you misinterpreted what the OP asked. He wants traffic to go to 192.168.10.0 (or any subnet for that matter) through the same interface that it came in on. You can add a static route as you mentioned: "route add -net 192.168.10.0 netmask 255.255.255.0 gw 192.168.0.2" but now traffic coming from 192.168.10.0 into the eth1 (10.0.0.0/24) will go out eth0. The OP wants it to go out eth1 in that case because it came in on eth1.
EE Admin: In fairness the author did post that he found his solution before anyone posted an answer for them so I imagine at least a partial refund would be fair. They however didn't close the question afterwards - and it would have been nice if he had posted the solution that worked for him instead of just "found something"...
chenry334: I did not misinterpret the question just thought that what the asker needs is to send intranet packet through the eth0 and the rest to the default gateway and suggested him a different approach. But unfortunately none of us experts could get futher feedback. IT seems that the asker has abondoned the question altogether.
Hi,
I would always accept if the asker finds a solution working for him. But he should not forget thath he was the one asking a question and he should have explained it clearly what was the answer for hm and how did it work for him. This way of abandoning the question is just something against the EE code.
I'll suggest you close the question with points refunded. However I'd also suggest you, the moderaotr, remind the asker about his responsibilities as an asker. If all questions had ended up like that no expert would like to answer any question thus this would kill EE.
Cheers,
K.
Business Accounts
Answer for Membership
by: lanboyoPosted on 2009-11-05 at 00:56:59ID: 25747689
This is somewhat difficult to do unless you know all of the networks that should go out thru eth0.
From the command line;
route add -net 192.168.10.0 netmask 255.255.255.0 gw 192.168.0.1
(assuming the gateway on eth0 is 192.168.0.1) will route traffic to the 192.168.10.0/24 network thru eth0.
route add -net 192.168.0.0 netmask 255.255.0.0 gw 192.168.0.1
will route all of the 192.168 networks that way... Since these routes are more specific than the default route, they will get routed out eth0.
These route commands will need to be added every time the box boots up, so you will need to add them to rc.local or another appropriate startup config file.
If you do not know what devices will connect from eth0, then you may have to do some kind of trickiness to get it to work, like using a different device to do network address translation to make incoming traffic to look like it comes from the local interface connected to eth0.