Hi,
I seem to be having a silly problem with routing. Cannot connect to
192.168.1.1 from test machine (details below). I will be grateful if
someone could suggest a fix.
---------------- 192.168.1.0
| |
___ ___ ___
| | | | | |
| 1 | | 2 |-- 192.168.2.0 ----| 1 |-- 129.x.y.z
| | | | | | |
^^^ ^^^ | ^^^
| | |
---------------- 192.168.3.0 |________
| |
| 2 |
|test |
| |
^^^^^
192.168.1.1 (192.168.3.1) is a DNS/DHCP/TFTP machine
192.168.1.2 (192.168.3.2, 192.168.2.2) is a router
192.168.2.1 (129.x.y.z) is a (diskless) firewall
192.168.2.2 is a test machine
192.168.2.0 is on a different switch
129.x.y.z is on yet another switch
I need the double set of network 192.168.1.0 and 192.168.3.0 since there
are a lot of other machines connected to them and I need network
loadbalaning (hpc cluster and bonding or equal cost multipath are not
working out well). I'm using kernel 2.4.22. All options of netfilter
code are built into the kernel but I am not using iptables and there are
no rules (INPUT OUTPUT FORWARD are empty).
All connections from directly wired machines work fine. I have
ip_forward enabled on the router. It shows the network routes
192.168.1.0 * U
192.168.3.0 * U
192.168.2.0 * U
default 192.168.2.1 UG
with a route
192.168.1.0 192.168.2.2 UG
on my test machine, I can connect (ssh) to 192.168.1.2 but not to
192.168.1.1. Also, I have dhcp relay on 192.168.1.2 and 192.168.2.1 gets
it's IP correctly (hardware ethernet), but does not load the kernel. I
tested this using atftp (client) from my test machine and it timed out.
tftpd also timed out on 192.168.1.1 (from the logs).
If you need log or tcpdump logs, please ask me.
TIA,
Indraneel
--
http://www.smartpatch.net
-
: send the line "unsubscribe linux-net" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
