Debian 12桥接网卡IPv4连接丢失，如何无需频繁配置connmand解决？

安全阿岳

2026-7-17

问题：Debian 12 + Proxmox 下无需频繁修改Connman配置实现双栈网络正常运行

我在Debian 12机器上配置了静态IPv4地址192.168.1.3/23，网关为192.168.1.1，同时通过SLAAC获取IPv6地址。为适配Proxmox，我将/etc/network/interfaces配置如下：

source /etc/network/interfaces.d/*

auto lo
iface lo inet loopback

iface eno1 inet manual

auto vmbr0
iface vmbr0 inet6 auto
iface vmbr0 inet static
    address 192.168.1.3/23
    gateway 192.168.1.1
    bridge-ports eno1
    bridge-stp off
    bridge-fd 0

这个配置对Proxmox虚拟机正常生效：虚拟机可以从本地网关获取IPv4，通过SLAAC获取IPv6，但主机仅IPv6可用。

调试后发现问题根源：avahi-daemon先以169.254.153.191加入mDNS多播组，随后connmand将主机IPv4地址替换为该值，导致IPv4连接丢失。以下是/var/log/syslog的相关片段：

2023-06-24T23:36:40.625360+03:00 cathouse avahi-daemon[622]: Joining mDNS multicast group on interface eno1.IPv4 with address 169.254.153.191.
2023-06-24T23:36:40.625549+03:00 cathouse avahi-daemon[622]: New relevant interface eno1.IPv4 for mDNS.
2023-06-24T23:36:40.625584+03:00 cathouse avahi-daemon[622]: Registering new address record for 169.254.153.191 on eno1.IPv4.
2023-06-24T23:36:40.641482+03:00 cathouse connmand[659]: eno1 {add} address 169.254.153.191/16 label eno1 family 2
2023-06-24T23:36:40.641562+03:00 cathouse connmand[659]: eno1 {add} route 169.254.0.0 gw 0.0.0.0 scope 253 <LINK>
2023-06-24T23:36:40.641595+03:00 cathouse connmand[659]: eno1 {add} route 0.0.0.0 gw 0.0.0.0 scope 253 <LINK>
2023-06-24T23:36:40.641622+03:00 cathouse connmand[659]: eno1 {add} route 0.0.0.0 gw 0.0.0.0 scope 253 <LINK>

我尝试在/etc/connman/main.conf中添加黑名单禁用connmand对eno1接口的管理：

# NetworkInterfaceBlacklist = vmnet,vboxnet,virbr,ifb,ve-,vb-
NetworkInterfaceBlacklist = eno,vmbr

此配置基本解决了主机IPv4的问题，但开机时connman-wait-online.service会持续等待网络上线，执行sudo systemctl disable connman-wait-online.service后解决了开机卡顿的问题。

主机恢复双栈连接后，创建并启动Proxmox虚拟机时，出现了tapXXXX、fwXXXX等接口，connmand再次接管这些接口并添加错误的默认路由，导致IPv4路由失效（IPv6仍正常）。此时的路由表如下：

kalle@cathouse:/etc/network$ sudo route
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
0.0.0.0         0.0.0.0         255.255.255.255 UH    0      0        0 tap201i0
0.0.0.0         0.0.0.0         255.255.255.255 UH    0      0        0 fwln102i0
0.0.0.0         0.0.0.0         255.255.255.255 UH    0      0        0 tap102i0
0.0.0.0         0.0.0.0         255.255.255.255 UH    0      0        0 fwpr102p0
0.0.0.0         0.0.0.0         255.255.255.255 UH    0      0        0 tap103i0
default         0.0.0.0         0.0.0.0         U     0      0        0 tap201i0
default         0.0.0.0         0.0.0.0         U     0      0        0 tap103i0
default         OpenWrt.home    0.0.0.0         UG    0      0        0 vmbr0
link-local      0.0.0.0         255.255.0.0     U     0      0        0 tap103i0
link-local      0.0.0.0         255.255.0.0     U     0      0        0 fwpr102p0
link-local      0.0.0.0         255.255.0.0     U     0      0        0 tap102i0
link-local      0.0.0.0         255.255.0.0     U     0      0        0 fwln102i0
link-local      0.0.0.0         255.255.0.0     U     0      0        0 tap201i0
192.168.0.0     0.0.0.0         255.255.254.0   U     0      0        0 vmbr0

再次修改/etc/connman/main.conf，将fw和tap接口加入黑名单后解决了该问题：

# NetworkInterfaceBlacklist = vmnet,vboxnet,virbr,ifb,ve-,vb-
NetworkInterfaceBlacklist = eno,vmbr,fw,tap

现在想请教：如何在无需频繁修改connmand配置文件的前提下，让主机同时正常使用IPv4和IPv6？

内容的提问来源于stack exchange，提问作者Kalle