大网数据广告活动
服务器运维问答

Configuration to use choose the hardware interface for the request?

· 52CE 我爱测

I have a linux machine (Debian 10 based OS), with 3 hardware interfaces connected to the internet, 2 being USB modems: ifconfig gives -> eth0: flags=4099<UP,BROADCAST,MULTICAST> mtu 1500 ether b8:27:eb:95:a0:2c txqueuelen 1000 (E…

问题原文

原帖链接 · Ernest Dujo · CC BY-SA 4.0

I have a linux machine (Debian 10 based OS), with 3 hardware interfaces connected to the internet, 2 being USB modems: ifconfig gives ->

eth0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500
        ether b8:27:eb:95:a0:2c  txqueuelen 1000  (Ethernet)
        RX packets 0  bytes 0 (0.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 0  bytes 0 (0.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 14258  bytes 1613046 (1.5 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 14258  bytes 1613046 (1.5 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

ppp0: flags=4305<UP,POINTOPOINT,RUNNING,NOARP,MULTICAST>  mtu 1500
        inet 10.199.25.78  netmask 255.255.255.255  destination 10.64.64.64
        ppp  txqueuelen 3  (Point-to-Point Protocol)
        RX packets 7  bytes 130 (130.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 8  bytes 181 (181.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

ppp1: flags=4305<UP,POINTOPOINT,RUNNING,NOARP,MULTICAST>  mtu 1500
        inet 10.227.136.222  netmask 255.255.255.255  destination 10.64.64.65
        ppp  txqueuelen 3  (Point-to-Point Protocol)
        RX packets 7  bytes 130 (130.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 8  bytes 181 (181.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

wlan0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.2.105  netmask 255.255.255.0  broadcast 192.168.2.255
        inet6 fe80::e1b9:e62c:3140:bfc5  prefixlen 64  scopeid 0x20<link>
        ether b8:27:eb:c0:f5:79  txqueuelen 1000  (Ethernet)
        RX packets 26548  bytes 5187998 (4.9 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 20025  bytes 5171235 (4.9 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

wlan0 is the default interface here, when I try: curl --interface ppp1 ifconfig.me or curl --interface ppp0 ifconfig.me the request times out, with sudo, sudo curl --interface ppp1 ifconfig.me gives response but the equivalent for ppp0 times out, what are the right routing rules to add to be able to select the hardware interface to route through?

Actual routing rules:

default via 192.168.2.1 dev wlan0 proto dhcp src 192.168.2.105 metric 303 
10.64.64.64 dev ppp0 proto kernel scope link src 10.233.6.240 
10.64.64.65 dev ppp1 proto kernel scope link src 10.149.182.92 
192.168.2.0/24 dev wlan0 proto dhcp scope link src 192.168.2.105 metric 303 

ip rule gives:

0:  from all lookup local 
32766:  from all lookup main 
32767:  from all lookup default 

ip -4 route show table all :

default via 192.168.2.1 dev wlan0 
10.64.64.64 dev ppp0 proto kernel scope link src 10.235.137.107 
10.64.64.65 dev ppp1 proto kernel scope link src 10.121.55.130 
192.168.2.0/24 dev wlan0 proto kernel scope link src 192.168.2.101 
local 10.121.55.130 dev ppp1 table local proto kernel scope host src 10.121.55.130 
local 10.235.137.107 dev ppp0 table local proto kernel scope host src 10.235.137.107 
broadcast 127.0.0.0 dev lo table local proto kernel scope link src 127.0.0.1 
local 127.0.0.0/8 dev lo table local proto kernel scope host src 127.0.0.1 
local 127.0.0.1 dev lo table local proto kernel scope host src 127.0.0.1 
broadcast 127.255.255.255 dev lo table local proto kernel scope link src 127.0.0.1 
broadcast 192.168.2.0 dev wlan0 table local proto kernel scope link src 192.168.2.101 
local 192.168.2.101 dev wlan0 table local proto kernel scope host src 192.168.2.101 
broadcast 192.168.2.255 dev wlan0 table local proto kernel scope link src 192.168.2.101 

PS. after having rebooted, all the routing I tried was gone(as expected)

已采纳解答

原帖链接 · Piotr P. Karwasz · CC BY-SA 4.0

What curl --interface <iface> http://example.com does is to connect to example.com, using interface's <iface> IP address as source. Otherwise routing proceeds as usual.

If you wish to route based on the packets source address, you need to add two routing tables (one for packets with source 10.199.25.78 and one for packets with source 10.227.136.222). Let's give the routing tables names, by adding:

200 ppp0
201 ppp1

to /etc/iproute2/rt_tables and fill them with default routes for packets from those interfaces:

ip route add default dev ppp0 via 10.64.64.64 table ppp0
ip route add default dev ppp1 via 10.64.64.65 table ppp1

Now, all you need to do is to add two rules which will select table ppp0 or ppp1 according to the source address:

ip rule add from 10.199.25.78 table ppp0
ip rule add from 10.227.136.222 table ppp1

Edit: I dug deeper into curl's --interface ppp0 option. It can work in two ways (cf. connect.c):

  • if the process has the CAP_NET_RAW capability (like when you are root) it binds to the given interface (through SO_BINDTODEVICE) and all routes that don't use the ppp0 interface are ignored.
  • if the process is unprivileged, curl binds to the IP address of the ppp0 interface (10.199.25.78) and routing proceeds as usual.

摘自公开归档,保留原文语言,版式经整理,外部媒体与邮箱地址已移除。

资料来源与适用范围

资料来源
Server Fault
适用范围
以原帖系统、软件版本与发布日期为准(2020-02-14)
内容语言
中文;命令、路径和错误标识保留技术原文
本地收录日期
2026-11-15
授权说明
CC BY-SA 4.0

正文与所需资源由本站本地存储。查看来源及更新记录 ↗

请按适用版本和实际日志确认原因;本站没有对每一种系统或设备逐项执行文中的操作。