All -
I have what I think is a very simple problem but the solution has eluded me.
I have three Token Ring interfaces in one box and I want to load balance
outgoing traffic across the NICs. The system is a RH 6.2 base with 2.2.18
kernel and current RH errata. Based on the Adv-Routing HOWTO I have the
proper stuff configured into the kernel already.
I'm using DNS to do cheap and easy round robin style connections to the box.
[scottrus@xxxxx scottrus]$ host ftp3.linux.ibm.com
ftp3.linux.ibm.com has address 9.37.145.124
ftp3.linux.ibm.com has address 9.37.145.120
ftp3.linux.ibm.com has address 9.37.145.127
ip address show
1: lo: <LOOPBACK,UP> mtu 3924 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
2: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 100
link/ether 00:60:94:3f:15:63 brd ff:ff:ff:ff:ff:ff
inet 10.200.1.33/24 brd 10.200.1.255 scope global eth0
3: tr0: <BROADCAST,MULTICAST,UP> mtu 2000 qdisc pfifo_fast qlen 100
link/tr 00:20:35:ff:09:66 brd ff:ff:ff:ff:ff:ff
inet 9.37.145.124/20 brd 9.37.159.255 scope global tr0
4: tr1: <BROADCAST,MULTICAST,UP> mtu 2000 qdisc pfifo_fast qlen 100
link/tr 00:60:94:b2:19:dd brd ff:ff:ff:ff:ff:ff
inet 9.37.145.120/20 brd 9.37.159.255 scope global tr1
5: tr2: <BROADCAST,MULTICAST,UP> mtu 2000 qdisc pfifo_fast qlen 100
link/tr 00:60:94:b2:19:6d brd ff:ff:ff:ff:ff:ff
inet 9.37.145.127/20 brd 9.37.159.255 scope global tr2
All trX interfaces are on the same subnet and use the same default gateway
of 9.37.144.1. The eth0 interface is a private network. Traffic from the
public Token Ring network should never cross onto the private eth0 network
or visversa.
I'm looking for one of two possible solutions:
1) Incoming requests are responded to on the same interface they came in on.
For example an ftp connection coming in on tr1 has all TX / RX packets stick
to tr1.
2) All output packets are load balanced across the tr0, tr1, tr2
interfaces regardless of which tr interface they came in on. (This is my
preferred solution.)
What's currently happening is requests come in on tr0, tr1, tr2 but outgoing
traffic is being sent back only via tr0, the default route.
[scottrus@xxxxx scottrus]$ ip route show
10.200.1.33 dev eth0 scope link
9.37.145.120 dev tr1 scope link
9.37.145.127 dev tr2 scope link
9.37.145.124 dev tr0 scope link
10.200.1.0/24 dev eth0 proto kernel scope link src 10.200.1.33
9.37.144.0/20 dev tr0 proto kernel scope link src 9.37.145.124
9.37.144.0/20 dev tr1 proto kernel scope link src 9.37.145.120
9.37.144.0/20 dev tr2 proto kernel scope link src 9.37.145.127
10.200.0.0/16 via 10.200.1.1 dev eth0
127.0.0.0/8 dev lo scope link
default via 9.37.144.1 dev tr2
default via 9.37.144.1 dev tr1
default via 9.37.144.1 dev tr0
As you can see I've tried simply adding default routes for each token ring
interface and I think this is a step in the right direction but I'm missing
something.
*sigh* I thought I knew it all, how wrong I was. I have to say all this is
rather cool and beyond this initial problem I know I'm going to have play
with it more just for fun. The very thought of being able to twist these
packets into my own demented idea of how they should be handled is just to
tempting to ignore. Baby steps, must remember to take baby steps.
1) Understand what the heck I've been reading all day.
2) Get the routes on ftp3 working so users stop yelling at me :)
3) Read more, play with iproute2, watch mailing list, help out. :) :)
Any help would be great. Examples would be better of course. Thanks much!
--
Regards,
Scott Russell (scottrus@xxxxxxxxxxxxxxx)
Linux Technology Center, System Admin, RHCE.
