Исходное сообщение
"Не получается балансировка нагрузки на 2 ISP" Отправлено sire, 20-Фев-07 17:12
>отправ в форум >ifconfig >iptables -t nat -L ifconfig eth0      Link encap:Ethernet  HWaddr 00:0E:0C:4B:77:63           inet addr:IP_P1  Bcast:0.0.0.0  Mask:255.255.255.248           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1           RX packets:283334260 errors:0 dropped:572 overruns:0 frame:0           TX packets:282930392 errors:0 dropped:0 overruns:0 carrier:0           collisions:0 txqueuelen:100           RX bytes:925389565 (882.5 MiB)  TX bytes:1171249970 (1.0 GiB)           Base address:0xef80 Memory:dffe0000-e0000000 eth2      Link encap:Ethernet  HWaddr 00:07:E9:2A:A8:35           inet addr:IP_P2  Bcast:0.0.0.0  Mask:255.255.255.248           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1           RX packets:954469 errors:0 dropped:0 overruns:0 frame:0           TX packets:791211 errors:0 dropped:0 overruns:0 carrier:0           collisions:0 txqueuelen:100           RX bytes:741232320 (706.8 MiB)  TX bytes:196984480 (187.8 MiB)           Base address:0xdf80 Memory:dfee0000-dff00000 eth3      Link encap:Ethernet  HWaddr 00:0E:0C:4B:77:62           inet addr:IP_INT  Bcast:0.0.0.0  Mask:255.255.255.0           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1           RX packets:257455117 errors:0 dropped:453 overruns:0 frame:0           TX packets:302234325 errors:0 dropped:0 overruns:0 carrier:0           collisions:0 txqueuelen:100           RX bytes:822227681 (784.1 MiB)  TX bytes:2066265344 (1.9 GiB)           Base address:0xcf80 Memory:dfde0000-dfe00000 lo        Link encap:Local Loopback           inet addr:127.0.0.1  Mask:255.0.0.0           inet6 addr: ::1/128 Scope:Host           UP LOOPBACK RUNNING  MTU:16436  Metric:1           RX packets:626 errors:0 dropped:0 overruns:0 frame:0           TX packets:626 errors:0 dropped:0 overruns:0 carrier:0           collisions:0 txqueuelen:0           RX bytes:54788 (53.5 KiB)  TX bytes:54788 (53.5 KiB) iptables -t nat -L Chain PREROUTING (policy ACCEPT) target     prot opt source               destination Chain POSTROUTING (policy ACCEPT) target     prot opt source               destination SNAT       0    --  anywhere             anywhere            MARK match 0x1 to:IP_P1 SNAT       0    --  anywhere             anywhere            MARK match 0x2 to:IP_P2 Chain OUTPUT (policy ACCEPT) target     prot opt source               destination iptables -t mangle -L Chain PREROUTING (policy ACCEPT) target     prot opt source               destination TOS        tcp  --  anywhere             anywhere            tcp dpt:telnet TOS set Minimize-Delay TOS        tcp  --  anywhere             anywhere            tcp dpt:ssh TOS set Minimize-Delay TOS        tcp  --  anywhere             anywhere            tcp dpt:ftp TOS set Minimize-Delay TOS        tcp  --  anywhere             anywhere            tcp dpt:domain TOS set Minimize-Delay TOS        udp  --  anywhere             anywhere            udp dpt:domain TOS set Minimize-Delay TOS        tcp  --  anywhere             anywhere            tcp dpt:sftp TOS set Maximize-Throughput TOS        tcp  --  anywhere             anywhere            tcp dpt:ftp-data TOS set Maximize-Throughput TOS        tcp  --  anywhere             anywhere            tcp dpt:www TOS set Maximize-Throughput TOS        tcp  --  anywhere             anywhere            tcp dpt:https TOS set Maximize-Throughput TOS        tcp  --  anywhere             anywhere            tcp dpt:webmin TOS set Maximize-Throughput TOS        tcp  --  anywhere             anywhere            tcp dpts:netbios-ns:netbios-ssn TOS set Maximize-Throughput TOS        udp  --  anywhere             anywhere            udp dpts:netbios-ns:netbios-ssn TOS set Maximize-Throughput CONNMARK   0    --  anywhere             anywhere            CONNMARK restore Chain INPUT (policy ACCEPT) target     prot opt source               destination Chain FORWARD (policy ACCEPT) target     prot opt source               destination TCPMSS     tcp  --  anywhere             anywhere            tcp flags:SYN,RST/SYN TCPMSS set 1400 Chain OUTPUT (policy ACCEPT) target     prot opt source               destination Chain POSTROUTING (policy ACCEPT) target     prot opt source               destination MARK       0    --  anywhere             anywhere            state NEW MARK set 0x1 MARK       0    --  anywhere             anywhere            state NEW MARK set 0x2 CONNMARK   0    --  anywhere             anywhere            state NEW CONNMARK save Вот так должно быть понятнее: $IPT -t mangle -A POSTROUTING -m state --state NEW -o $IF_EXT1 -j MARK --set-mark 1 $IPT -t mangle -A POSTROUTING -m state --state NEW -o $IF_EXT2 -j MARK --set-mark 2 $IPT -t mangle -A POSTROUTING -m state --state NEW -j CONNMARK --save-mark $IPT -t mangle -A PREROUTING -j CONNMARK --restore-mark # NAT $IPT -t nat -A POSTROUTING -m mark --mark 1 -j SNAT --to-source $IP_EXT1 $IPT -t nat -A POSTROUTING -m mark --mark 2 -j SNAT --to-source $IP_EXT2 Сначала делал так: $IPT -t mangle -A PREROUTING -i $IF_EXT1 -j MARK --set-mark 1 $IPT -t mangle -A PREROUTING -i $IF_EXT2 -j MARK --set-mark 2 $IPT -t nat -A POSTROUTING -s 10.0.0.0/255.0.0.0 -o $IF_EXT1 -j SNAT --to-source $IP_EXT1 $IPT -t nat -A POSTROUTING -s 10.0.0.0/255.0.0.0 -o $IF_EXT2 -j SNAT --to-source $IP_EXT2 Потом нашёл в инете тот вариант, что указан выше. Он вроде бы лучше. Однако на мою ситуацию смена этих методов не влияет.