Hi List,
I am running into a problem where I have 2 interfaces bridged with and ip
address assigned.
I have another interface in which traffic has ingress traffic that needs to go
out the bridged interface.
I am trying unsuccessfully to SNAT the traffic leaving the bridge interface to
its assigned address.
# brctl show xbrdg0
bridge name bridge id STP enabled interfaces
xbrdg0 8000.000c297aa55f no eth0
eth1
# ip a s xbrdg0
11: xbrdg0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state
UNKNOWN
link/ether 00:0c:29:7a:a5:5f brd ff:ff:ff:ff:ff:ff
inet 192.168.100.3/24 scope global xbrdg0
# ip a s eth5
7: eth5: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen
1000
link/ether 00:0c:29:7a:a5:7d brd ff:ff:ff:ff:ff:ff
inet 10.10.0.1/29 scope global eth5
default via 192.168.100.1 dev xbrdg0
So I want traffic coming in eth5 with 10.10.0.x addresses to be source natted to
192.168.100.3.
But my iptables nat statement never gets hit.
Chain POSTROUTING (policy ACCEPT 172 packets, 31384 bytes)
pkts bytes target prot opt in out source destination
0 0 SNAT all -- * xbrdg0 0.0.0.0/0 0.0.0.0
to:192.168.100.3
29 1933 MASQUERADE all -- * tun+ 0.0.0.0/0 0.0.0.0/0
# ping -I 10.10.0.1 8.8.8.8
# tcpdump -nli xbrdg0 icmp or arp
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on xbrdg0, link-type EN10MB (Ethernet), capture size 65535 bytes
12:52:06.914295 IP 10.10.0.1 > 8.8.8.8: ICMP echo request, id 38932, seq 1,
length 64
12:52:07.914592 IP 10.10.0.1 > 8.8.8.8: ICMP echo request, id 38932, seq 2,
length 64
12:52:08.914579 IP 10.10.0.1 > 8.8.8.8: ICMP echo request, id 38932, seq 3,
length 64
Any ideas?
Thanks,
Steve
--
Stephen Clark
On 01/20/2016 09:55 AM, Steve Clark wrote:> Any ideas?IP forwarding needs to be enabled, and you also need rules in your FORWARD chain to allow the packets.
On 01/20/2016 04:21 PM, Gordon Messmer wrote:> On 01/20/2016 09:55 AM, Steve Clark wrote: >> Any ideas? > IP forwarding needs to be enabled, and you also need rules in your > FORWARD chain to allow the packets. >Thanks, but forwarding is turned on and my FW rules are empty. Chain INPUT (policy ACCEPT 359K packets, 136M bytes) pkts bytes target prot opt in out source destination Chain FORWARD (policy ACCEPT 55801 packets, 4736K bytes) pkts bytes target prot opt in out source destination Chain OUTPUT (policy ACCEPT 319K packets, 141M bytes) pkts bytes target prot opt in out source destination -- Stephen Clark
On 20/01/2016 19:55, Steve Clark wrote:> > So I want traffic coming in eth5 with 10.10.0.x addresses to be source > natted to 192.168.100.3. > But my iptables nat statement never gets hit. > > Chain POSTROUTING (policy ACCEPT 172 packets, 31384 bytes) > pkts bytes target prot opt in out source > destination > 0 0 SNAT all -- * xbrdg0 0.0.0.0/0 > 0.0.0.0 to:192.168.100.3 > 29 1933 MASQUERADE all -- * tun+ 0.0.0.0/0 > 0.0.0.0/0 > > # ping -I 10.10.0.1 8.8.8.8First you should try to match without SNAT at all with a simple log target and see if it matches. I would start with: iptables -t nat -I POSTROUTING -s 10.0.0.1 -o xbrdg0 -j LOG --log-prefix "Should-SNAT: " --log-level 4 And then: iptables -t nat -I POSTROUTING -s 10.0.0.1 -o xbrdg0 -j SNAT --to-source 192.168.100.3 And see what happens. Also there might be something about this bridge settings and it maybe needs the "-o eth1" but it would be a bit weird. Eliezer