Re: CONNMARK udp comprehension question

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



This might help you without the need to mess with marks.

For each uplink interface with an incremental $x

ip route add default via $gateway_ip dev $uplink_interface table $x
ip rule add from $ip_of_interface dev $uplink_interface table $x

alternatively use -j CONNMARK --restore-mark in -t mangle PREROUTING so that ip rules can match the mark or as a 3rd option use the ROUTE target.

- Jörg

Peter Warasin schrieb:
Hi List

Have a question in order to understand if i am tilting at windmills or
if there's an issue with my setup.

Here a short description of my scenario to help me to explain myself:

I have a box with multiple uplinks running openvpn in udp mode. One of
the uplinks has the default gateway (no multipath route). Now, whenever
i connect to an uplink which actually hasn't the default gateway,
packets go in through that device and exit through another (due to the
default gateway), which will break the udp-"connection".

Now, the question:
I mark all connections entering a specific uplink with a number using
CONNMARK, in order to be able to distinguish them and make them leave
the correct interface using fwmark based ip rules.

Shouldn't now the CONNMARK target, mark the connection in such a manner
that the udp packets coming in have the same mark as the udp packets
going out whenever they belong to the same "connection"?
Or is this just not possible due to the connection-less nature of udp?

Same setup works for me with tcp and it also works with udp when the
connections going through the box (FORWARD).

Kernel is RH'L5.1's kernel-2.6.18-53 with all relevant options enabled
and iptables 1.3.8.

Any help would be greatly appreciated!

peter


-
To unsubscribe from this list: send the line "unsubscribe netfilter" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[Index of Archives]     [Linux Netfilter Development]     [Linux Kernel Networking Development]     [Netem]     [Berkeley Packet Filter]     [Linux Kernel Development]     [Advanced Routing & Traffice Control]     [Bugtraq]

  Powered by Linux