我正在尝试使用虚拟机和2个虚拟路由器构建虚拟网络。
VM - ＆gt; Router1 - ＆gt; Router2 - ＆gt;外部网络
Router1执行SNAT并且运行良好。预计Router2会进行ip转发，但不能正常工作。
以下是我检查过的Router2的详细信息。 （Router2位于netns d3dcb2df-f3ca-4079-a434-491b23f84b5a内。）
网卡和地址
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
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: qr-70aabff6-c8@if60: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1450 qdisc pfifo_fast state UP qlen 1000
link/ether fa:16:3e:29:3b:ea brd ff:ff:ff:ff:ff:ff link-netnsid 0
inet 192.168.1.1/24 brd 192.168.1.255 scope global qr-70aabff6-c8
valid_lft forever preferred_lft forever
inet6 fe80::f816:3eff:fe29:3bea/64 scope link
valid_lft forever preferred_lft forever
3: qg-30c10598-27@if63: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether fa:16:3e:fc:1b:5b brd ff:ff:ff:ff:ff:ff link-netnsid 0
inet 10.10.52.82/24 brd 10.10.52.255 scope global qg-30c10598-27
valid_lft forever preferred_lft forever
inet 10.10.52.158/32 brd 10.10.52.158 scope global qg-30c10598-27
valid_lft forever preferred_lft forever
inet 10.10.52.73/32 brd 10.10.52.73 scope global qg-30c10598-27
valid_lft forever preferred_lft forever
inet 10.10.52.68/32 brd 10.10.52.68 scope global qg-30c10598-27
valid_lft forever preferred_lft forever
inet6 fe80::f816:3eff:fefc:1b5b/64 scope link
valid_lft forever preferred_lft forever
路线规则
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a ip rule
0: from all lookup local
32766: from all lookup main
32767: from all lookup default
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a ip route
default via 10.10.52.1 dev qg-30c10598-27
10.10.52.0/24 dev qg-30c10598-27 proto kernel scope link src 10.10.52.82
192.168.1.0/24 dev qr-70aabff6-c8 proto kernel scope link src 192.168.1.1
转发已启用
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a sysctl net.ipv4.ip_forward
net.ipv4.ip_forward = 1
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a sysctl net.ipv4.conf.qr-70aabff6-c8.forwarding
net.ipv4.conf.qr-70aabff6-c8.forwarding = 1
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a sysctl net.ipv4.conf.qg-30c10598-27.forwarding
net.ipv4.conf.qg-30c10598-27.forwarding = 1
清除iptables规则
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a iptables -t mangle -F
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a iptables -t nat -F
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a iptables -t filter -F
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a iptables -t mangle -L -n
Chain PREROUTING (policy ACCEPT)
target prot opt source destination
Chain INPUT (policy ACCEPT)
target prot opt source destination
Chain FORWARD (policy ACCEPT)
target prot opt source destination
Chain OUTPUT (policy ACCEPT)
target prot opt source destination
Chain POSTROUTING (policy ACCEPT)
target prot opt source destination
Chain neutron-l3-agent-FORWARD (0 references)
target prot opt source destination
Chain neutron-l3-agent-INPUT (0 references)
target prot opt source destination
Chain neutron-l3-agent-OUTPUT (0 references)
target prot opt source destination
Chain neutron-l3-agent-POSTROUTING (0 references)
target prot opt source destination
Chain neutron-l3-agent-PREROUTING (0 references)
target prot opt source destination
Chain neutron-l3-agent-float-snat (0 references)
target prot opt source destination
Chain neutron-l3-agent-floatingip (0 references)
target prot opt source destination
Chain neutron-l3-agent-mark (0 references)
target prot opt source destination
Chain neutron-l3-agent-scope (0 references)
target prot opt source destination
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a iptables -t nat -L -n
Chain PREROUTING (policy ACCEPT)
target prot opt source destination
Chain INPUT (policy ACCEPT)
target prot opt source destination
Chain OUTPUT (policy ACCEPT)
target prot opt source destination
Chain POSTROUTING (policy ACCEPT)
target prot opt source destination
Chain neutron-l3-agent-OUTPUT (0 references)
target prot opt source destination
Chain neutron-l3-agent-POSTROUTING (0 references)
target prot opt source destination
Chain neutron-l3-agent-PREROUTING (0 references)
target prot opt source destination
Chain neutron-l3-agent-float-snat (0 references)
target prot opt source destination
Chain neutron-l3-agent-snat (0 references)
target prot opt source destination
Chain neutron-postrouting-bottom (0 references)
target prot opt source destination
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a iptables -t filter -L -n
Chain INPUT (policy ACCEPT)
target prot opt source destination
Chain FORWARD (policy ACCEPT)
target prot opt source destination
Chain OUTPUT (policy ACCEPT)
target prot opt source destination
Chain neutron-filter-top (0 references)
target prot opt source destination
Chain neutron-l3-agent-FORWARD (0 references)
target prot opt source destination
Chain neutron-l3-agent-INPUT (0 references)
target prot opt source destination
Chain neutron-l3-agent-OUTPUT (0 references)
target prot opt source destination
Chain neutron-l3-agent-local (0 references)
target prot opt source destination
Chain neutron-l3-agent-scope (0 references)
target prot opt source destination
最后，当我从VM ping 8.8.8.8时，路由器只能看到收到的数据包，没有转发的数据包。
[root@controller ~]# ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a tcpdump -i any -nn icmp
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on any, link-type LINUX_SLL (Linux cooked), capture size 65535 bytes
14:00:37.138271 IP 10.10.52.140 > 8.8.8.8: ICMP echo request, id 9616, seq 10258, length 64
14:00:38.139298 IP 10.10.52.140 > 8.8.8.8: ICMP echo request, id 9616, seq 10259, length 64
14:00:39.140488 IP 10.10.52.140 > 8.8.8.8: ICMP echo request, id 9616, seq 10260, length 64
感谢您的帮助。
答案 0 :(得分：0)
感谢上帝。在深入研究内核源代码后，我终于找到了答案。有一个小开关＆＃39; rp_filter＆＃39;告诉内核掉线“坏”数据包。以下是内核doc的完整描述：
rp_filter - INTEGER
0 - No source validation.
1 - Strict mode as defined in RFC3704 Strict Reverse Path
Each incoming packet is tested against the FIB and if the interface
is not the best reverse path the packet check will fail.
By default failed packets are discarded.
2 - Loose mode as defined in RFC3704 Loose Reverse Path
Each incoming packet's source address is also tested against the FIB
and if the source address is not reachable via any interface
the packet check will fail.
Current recommended practice in RFC3704 is to enable strict mode
to prevent IP spoofing from DDos attacks. If using asymmetric routing
or other complicated routing, then loose mode is recommended.
The max value from conf/{all,interface}/rp_filter is used
when doing source validation on the {interface}.
Default value is 0. Note that some distributions enable it
in startup scripts.
在我的情况下，这样做是很好的：
ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a sysctl net.ipv4.conf.all.rp_filter=0
ip netns exec qrouter-d3dcb2df-f3ca-4079-a434-491b23f84b5a sysctl net.ipv4.conf.qr-70aabff6-c8.rp_filter=0