NAT reversal issue

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

 



Hello,

I've got a NAT rule which NATs a TCP port from an outside IP onto a non-routable address. If this is a local IP address, it will work fine without any issues. However, if the IP is on the other end of a GRE tunnel, it never seems to reverse the DNAT rule when it throws the packet back out the external interface.

23:17:15.214869 IP 207.166.193.108.47867 > 207.166.203.130.2525: S 3159601203:3159601203(0) win 5840 <mss 1460,sackOK,timestamp 3010684420 0,nop,wscale 2>
23:17:15.277547 IP 10.2.4.5.25 > 207.166.193.108.47867: S 2921522304:2921522304(0) ack 3159601204 win 5792 <mss 1460,sackOK,timestamp 1996689776 3010684420,nop,wscale 2>
23:17:18.215536 IP 207.166.193.108.47867 > 207.166.203.130.2525: S 3159601203:3159601203(0) win 5840 <mss 1460,sackOK,timestamp 3010687420 0,nop,wscale 2>
23:17:18.278776 IP 207.166.193.237 > 207.166.193.108: icmp 48: host 207.166.203.130 unreachable
23:17:18.289971 IP 10.2.4.5.25 > 207.166.193.108.47867: S 2921522304:2921522304(0) ack 3159601204 win 5792 <mss 1460,sackOK,timestamp 1996692783 3010684420,nop,wscale 2>
23:17:19.036447 IP 10.2.4.5.25 > 207.166.193.108.47867: S 2921522304:2921522304(0) ack 3159601204 win 5792 <mss 1460,sackOK,timestamp 1996693476 3010684420,nop,wscale 2>
23:17:21.291234 IP 207.166.193.237 > 207.166.193.108: icmp 48: host 207.166.203.130 unreachable


tcp 6 29 SYN_SENT src=207.166.193.108 dst=207.166.203.130 sport=47867 dport=2525 [UNREPLIED] src=10.2.4.5 dst=207.166.193.108 sport=25 dport=47867 mark=21 use=1

I'm drawing a blank as to why the NAT rule is not reversed - There are no specific rules which limit traffic to or from the GRE interfaces, which are entirely non-routable, and I'm sure it's not a routeing issue with the tunnel as I see the SYN come in and the SYN ACK go out on the remote end of the tunnel. I can connect to port 25 on 10.2.4.5 from the firewall without any problems. The only issue may be that the MTU of the GRE tunnels is 1420, rather than 1500...

Any debugging pointers would be appreciated - I'm not sure exactly where to go from here.

Thanks,
David

--
David J. Coulson
email: david@xxxxxxxxxxxxxxxx
web: http://www.davidcoulson.net/
phone: (216) 920-3100 / (216) 258-4942


[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