我在网格上使用 RED HAT 5.9 操作系统，有 3 台机器：1 个头节点（称为 ilmn-qm.ilmn）和 2 个计算节点（又称为compute-00-00 和compute-00-01）。
问题是我无法从任一计算节点单元使用 SSH。
我试过：
1) SSH FROM 和 TO 头节点工作正常。
2) 从头节点到计算节点的 SSH 有效。
3) 反之亦然，从计算节点到头节点的 SSH 也可以工作。
4）头节点定义为网关：
[root@compute-00-01 ~]# route
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
172.20.22.0 * 255.255.255.0 U 0 0 0 eth1
172.20.20.0 * 255.255.255.0 U 0 0 0 eth0
169.254.0.0 * 255.255.0.0 U 0 0 0 eth0
default ilmn-qm.ilmn 0.0.0.0 UG 0 0 0 eth0
5) 我检查了头节点上是否启用了 ipv4 转发
cat /etc/sysctl.conf
# Kernel sysctl configuration file for Red Hat Linux
#
# For binary values, 0 is disabled, 1 is enabled. See sysctl(8) and
# sysctl.conf(5) for more details.
# Controls IP packet forwarding
net.ipv4.ip_forward = 1
# Controls source route verification
net.ipv4.conf.default.rp_filter = 1
# Do not accept source routing
net.ipv4.conf.default.accept_source_route = 0
# Controls the System Request debugging functionality of the kernel
kernel.sysrq = 0
# Controls whether core dumps will append the PID to the core filename
# Useful for debugging multi-threaded applications
kernel.core_uses_pid = 1
# Controls the use of TCP syncookies
net.ipv4.tcp_syncookies = 1
# Controls the maximum size of a message, in bytes
kernel.msgmnb = 65536
# Controls the default maxmimum size of a mesage queue
kernel.msgmax = 65536
# Controls the maximum shared segment size, in bytes
kernel.shmmax = 68719476736
# Controls the maximum number of shared memory segments, in pages
kernel.shmall = 4294967296
然而任何 ssh 尝试都会以：
ssh: connect to host 132.68.107.69 port 22: Connection timed out
从头节点：
root@ilmn-qm ~ # ip a show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether f0:4d:a2:0b:2d:b9 brd ff:ff:ff:ff:ff:ff
inet 132.68.106.1/28 brd 132.68.106.15 scope global eth0
inet6 fe80::f24d:a2ff:fe0b:2db9/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether f0:4d:a2:0b:2d:bb brd ff:ff:ff:ff:ff:ff
inet 172.20.20.5/24 brd 172.20.20.255 scope global eth1
inet6 fe80::f24d:a2ff:fe0b:2dbb/64 scope link
valid_lft forever preferred_lft forever
4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether f0:4d:a2:0b:2d:bd brd ff:ff:ff:ff:ff:ff
inet 172.20.21.2/24 brd 172.20.21.255 scope global eth2
inet6 fe80::f24d:a2ff:fe0b:2dbd/64 scope link
valid_lft forever preferred_lft forever
5: eth3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop qlen 1000
link/ether f0:4d:a2:0b:2d:bf brd ff:ff:ff:ff:ff:ff
6: sit0: <NOARP> mtu 1480 qdisc noop
link/sit 0.0.0.0 brd 0.0.0.0
root@ilmn-qm ~ # ip route show
132.68.106.0/28 dev eth0 proto kernel scope link src 132.68.106.1
172.20.21.0/24 dev eth2 proto kernel scope link src 172.20.21.2
172.20.20.0/24 dev eth1 proto kernel scope link src 172.20.20.5
169.254.0.0/16 dev eth2 scope link
default via 132.68.106.14 dev eth0
来自compute-00-00：
[root@compute-00-00 ~]# ip a show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether f0:4d:a2:0b:2d:c2 brd ff:ff:ff:ff:ff:ff
inet 172.20.20.6/24 brd 172.20.20.255 scope global eth0
inet6 fe80::f24d:a2ff:fe0b:2dc2/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether f0:4d:a2:0b:2d:c4 brd ff:ff:ff:ff:ff:ff
inet 172.20.22.6/24 brd 172.20.22.255 scope global eth1
4: eth2: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether f0:4d:a2:0b:2d:c6 brd ff:ff:ff:ff:ff:ff
5: eth3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether f0:4d:a2:0b:2d:c8 brd ff:ff:ff:ff:ff:ff
6: sit0: <NOARP> mtu 1480 qdisc noop
link/sit 0.0.0.0 brd 0.0.0.0
[root@compute-00-00 ~]# ip route show
172.20.22.0/24 dev eth1 proto kernel scope link src 172.20.22.6
172.20.20.0/24 dev eth0 proto kernel scope link src 172.20.20.6
169.254.0.0/16 dev eth1 scope link
default via 172.20.20.5 dev eth0
来自compute-00-01：
[root@compute-00-01 ~]# ip a show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 84:2b:2b:f9:9e:11 brd ff:ff:ff:ff:ff:ff
inet 172.20.20.7/24 brd 172.20.20.255 scope global eth0
inet6 fe80::862b:2bff:fef9:9e11/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 84:2b:2b:f9:9e:13 brd ff:ff:ff:ff:ff:ff
inet 172.20.22.7/24 brd 172.20.22.255 scope global eth1
4: eth2: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 84:2b:2b:f9:9e:15 brd ff:ff:ff:ff:ff:ff
5: eth3: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 84:2b:2b:f9:9e:17 brd ff:ff:ff:ff:ff:ff
6: sit0: <NOARP> mtu 1480 qdisc noop
link/sit 0.0.0.0 brd 0.0.0.0
[root@compute-00-01 ~]# ip route show
172.20.22.0/24 dev eth1 proto kernel scope link src 172.20.22.7
172.20.20.0/24 dev eth0 proto kernel scope link src 172.20.20.7
169.254.0.0/16 dev eth0 scope link
default via 172.20.20.5 dev eth0
答案1
您可能需要向头节点添加 SNAT 规则，以便在尝试到达 Internet 上的某些服务器时，来自计算节点的流量具有正确的源 IP 地址。发送源 IP 地址在 172.20.20.0/24 范围内的数据包将不起作用。
您可以在头节点上像这样配置 SNAT：
iptables -t nat -A POSTROUTING -o eth0 -s 172.20.20.0/24 -j SNAT --to 32.68.106.1