Container unresponsive on 2nd network interface

[awado]

This is something really confusing: I have a proxmox server with two networks. Each network has its own internet connection. Network interfaces of VMs and CTs can be assigned to any of them via VLAN tagging. There's one CT with Debian 11 and a nginx reverse proxy. It has both of these two network interfaces, but proxying only works on the first interface. The second one never gets any http requests handled by nginx. Why? Do CT interfaces have some kind of routing priorities? When I do the exact same setup on a physical server it works as expected.

 

[mira]

Have you compared the routing tables of both your CT and the physical server insallation?
ip r and ip -6 r (IPv6)

 

[awado]

root@ct:~# ip r
default via 192.168.0.1 dev eth0
10.0.0.0/24 dev eth1 proto kernel scope link src 10.0.0.131
192.168.0.0/24 dev eth0 proto kernel scope link src 192.168.0.66

Both show the same routing tables. IPv6 is disabled.

 

[mira]

Which IP address do you use to access the container on the interfaces?
Can you access it via SSH on both interfaces?

 

[awado]

Sorry for the late reply. I tested SSH and yes, SSH works on both interfaces. So it seems to be an nginx issue?

 

[mira]

Most likely since the network itself is working fine.

 

[awado]

I'll investigate this further. But thanks anyway for your support, mira.