Re: Iptables v1.4.4 + kernel 2.6.31 mangle marking changed?

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

 



Patrick McHardy schreef:
Ralph de Boom wrote:
Patrick McHardy schreef:
Ralph de Boom wrote:
Hi there,

Excuse me if this email might go wrong, it's my first message to a
mailing list.

But here's my problem: (And I hope you guys could shed light for me...)

I originally ran Debian Lenny on kernel 2.6.18.
Since today I reinstalled it to Ubuntu Server 9.10 with kernel  2.6.31.

Now I used to do this in lenny:

iptables -t mangle -A PREROUTING -s 192.168.1.0/24 -d 81.4.97.0/24 -j
MARK --set-mark 0x1

This would cause relevant packets to be marked 0x1, which in return I
had a 'ip rule':

my rules look like this:

ip rule show
0:      from all lookup local
32760:  from all fwmark 0x2 lookup upc
32761:  from all fwmark 0x1 lookup xs4all
32762:  from 192.168.1.XX lookup xs4all
32763:  from 192.168.1.XX lookup upc
32764:  from 24.132.104.XXX lookup upc
32765:  from 192.168.2.XX lookup xs4all
32766:  from all lookup main
32767:  from all lookup default

And my 'xs4all' table looks like:

ip route show table xs4all
192.168.2.0/24 dev eth0  proto kernel  scope link  src 192.168.2.XX
default via 192.168.2.X dev eth0


I know the rule matches packets i make:

iptables -t mangle -v -L
Chain PREROUTING (policy ACCEPT 3111K packets, 1861M bytes)
pkts bytes target prot opt in out source destination 16 1100 MARK all -- any any 192.168.1.0/24 ip-space.by.proserve.nl/24 MARK xset 0x1/0xffffffff

But somehow the connection is never relayed over the xs4all table...

The changes I've noticed compared to lenny:

iptables now likes to mark my --set-mark 0x1 as a --set-xmark
0x1/0xffffffff
whereas in lenny it would stay a --set-mark 0x1

Would be very pleased if someone could help me in this matter.
Please try adding a LOG rule directly after the marking rule and
see what it prints out for the MARK= value.

At first, thanks for helping me out!

Here's the info:

iptables -t mangle -v -L
Chain PREROUTING (policy ACCEPT 42M packets, 25G bytes)
pkts bytes target prot opt in out source destination 362 84150 MARK all -- any any 192.168.1.0/24 ip-space.by.proserve.nl/24 MARK xset 0x1/0xffffffff 362 84150 LOG all -- any any 192.168.1.0/24 ip-space.by.proserve.nl/24 LOG level debug prefix `fwmark 0x1: '

kern.log:
Nov  4 14:12:58 sakura kernel: [52836.368503] fwmark 0x1: IN=eth1 OUT=
MAC=00:1b:21:2d:a9:fa:00:1b:21:32:99:5f:08:00 SRC=192.168.1.30

This looks fine, it also works properly for me. Perhaps the
packets are already delivered locally through the "local"
table. The TRACE target should be able to tell you more.
Right, at this point you've lost me, how I will manage to do that, and where does the infomation get stored?


--
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