我有一个Linux机器处理路由为我的networking。
这里有一些相关的configuration（公共IP地址掩码到192.0.0.x）：
IPconfiguration：
[root@gw ~]# ip addr 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000 link/ether 00:0d:b9:32:fa:d0 brd ff:ff:ff:ff:ff:ff inet 192.168.1.1/24 brd 192.168.1.255 scope global eth0 valid_lft forever preferred_lft forever inet6 fe80::20d:b9ff:fe32:fad0/64 scope link valid_lft forever preferred_lft forever 3: eth1: <NO-CARRIER,BROADCAST,MULTICAST,PROMISC,UP> mtu 1500 qdisc fq_codel master br0 state DOWN group default qlen 1000 link/ether 00:0d:b9:32:fa:d1 brd ff:ff:ff:ff:ff:ff 4: eth2: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc fq_codel master br0 state UP group default qlen 1000 link/ether 00:0d:b9:32:fa:d2 brd ff:ff:ff:ff:ff:ff inet6 fe80::20d:b9ff:fe32:fad2/64 scope link valid_lft forever preferred_lft forever 5: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default link/ether 00:0d:b9:32:fa:d1 brd ff:ff:ff:ff:ff:ff inet 192.0.0.2/30 brd 192.0.0.3 scope global br0 valid_lft forever preferred_lft forever inet6 fe80::20d:b9ff:fe32:fad1/64 scope link valid_lft forever preferred_lft forever （ br0是eth1和eth2之间的桥梁，目前没有连接到eth1 。
路由表：
[root@gw ~]# ip route default via 192.0.0.1 dev br0 192.0.0.0/30 dev br0 proto kernel scope link src 192.0.0.2 192.168.1.0/24 dev eth0 proto kernel scope link src 192.168.1.1
testing路线：
[root@gw ~]# ip route get 192.168.1.2 192.168.1.2 dev eth0 src 192.168.1.1 cache
数据包捕获的PING：
(as seen from PING) [root@gw ~]# ping 192.168.1.2 PING 192.168.1.2 (192.168.1.2) 56(84) bytes of data. 64 bytes from 192.168.1.2: icmp_seq=1 ttl=64 time=0.476 ms 64 bytes from 192.168.1.2: icmp_seq=2 ttl=64 time=0.701 ms 64 bytes from 192.168.1.2: icmp_seq=3 ttl=64 time=0.466 ms 64 bytes from 192.168.1.2: icmp_seq=4 ttl=64 time=0.701 ms ^C --- 192.168.1.2 ping statistics --- 4 packets transmitted, 4 received, 0% packet loss, time 3000ms rtt min/avg/max/mdev = 0.466/0.586/0.701/0.115 ms (as seen from tcpdump on same machine) [root@gw ~]# tcpdump -i eth0 -n host 192.168.1.2 tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth0, link-type EN10MB (Ethernet), capture size 262144 bytes 12:48:58.187862 IP 192.0.0.2 > 192.168.1.2: ICMP echo request, id 1451, seq 1, length 64 12:48:58.188242 IP 192.168.1.2 > 192.0.0.2: ICMP echo reply, id 1451, seq 1, length 64 12:48:59.187086 IP 192.0.0.2 > 192.168.1.2: ICMP echo request, id 1451, seq 2, length 64 12:48:59.187696 IP 192.168.1.2 > 192.0.0.2: ICMP echo reply, id 1451, seq 2, length 64 12:49:00.188837 IP 192.0.0.2 > 192.168.1.2: ICMP echo request, id 1451, seq 3, length 64 12:49:00.189214 IP 192.168.1.2 > 192.0.0.2: ICMP echo reply, id 1451, seq 3, length 64 12:49:01.187834 IP 192.0.0.2 > 192.168.1.2: ICMP echo request, id 1451, seq 4, length 64 12:49:01.188450 IP 192.168.1.2 > 192.0.0.2: ICMP echo reply, id 1451, seq 4, length 64 ^C 8 packets captured 8 packets received by filter 0 packets dropped by kernel
机器上的PING捕获192.168.1.2：
[admin@raspberry ~]$ sudo tcpdump -i eth0 -n icmp tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth0, link-type EN10MB (Ethernet), capture size 262144 bytes 12:51:32.586016 IP 192.0.0.2 > 192.168.1.2: ICMP echo request, id 1451, seq 1, length 64 12:51:32.586822 IP 192.168.1.2 > 192.0.0.2: ICMP echo reply, id 1451, seq 1, length 64 12:51:33.587715 IP 192.0.0.2 > 192.168.1.2: ICMP echo request, id 1451, seq 2, length 64 12:51:33.588446 IP 192.168.1.2 > 192.0.0.2: ICMP echo reply, id 1451, seq 2, length 64 12:51:34.589160 IP 192.0.0.2 > 192.168.1.2: ICMP echo request, id 1451, seq 3, length 64 12:51:34.589885 IP 192.168.1.2 > 192.0.0.2: ICMP echo reply, id 1451, seq 3, length 64 12:51:35.590523 IP 192.0.0.2 > 192.168.1.2: ICMP echo request, id 1451, seq 4, length 64 12:51:35.591254 IP 192.168.1.2 > 192.0.0.2: ICMP echo reply, id 1451, seq 4, length 64 ^C 8 packets captured 8 packets received by filter 0 packets dropped by kernel [admin@raspberry ~]$
好，所以PING工作，但我的问题依然存在：为什么Linux将面向公众的接口的源IP分配给发往内部接口的ping？
当我在内部局域网上设置了一个故意没有默认网关的testing机时，问题就变得很明显了。 因此，PING正在从192.0.0.2获得机器，因此机器的答复从未到达服务器（因为它没有configuration默认网关）。
进入网关盒的PING可以正常回来，很可能是因为它们的源IP设置正确，所以来自网关的PING回复正在使用正确的源IP进行回复。
鉴于路由表显示正确的来源，任何人都可以解释为什么源IP是错误的，以及如何解决它？
BTW：
[root@gw ~]# uname -a Linux gw 4.0.4-2-ARCH #1 SMP PREEMPT Fri May 22 03:19:32 UTC 2015 i686 GNU/Linux