Hi all,
I have 2 physical locations in different cities.
Location A:
192.168.0.0
192.168.0.1 - OPNSense dedicated router with WG enabled
192.168.0.6 - test Linux VM
Location B:
192.168.2.0
192.168.2.1 - ISP router (cannot be changed)
192.168.2.10 - Proxmox host with running VMs, firewall disabled on host and dc level
192.168.2.11 - OPNsesne VM and WG enabled on it, only LAN used
Overall -> The Wireguard works fine.
Ping results:
192.168.0.6 -> 192.168.2.1 : OK
192.168.0.6 -> 192.168.2.11 : OK
BUT
when you try to ping Proxmox server itself from the remote location -> after aprox 5 tries -> it fails. Also you cannot connect via GUI. I have Prometheus monitoring setup, it also fails.
BUT when you run ping 192.168.2.1 (ISP router) in the same time -> it magically starts to work. Everything. You can Ping Proxmox, connect to GUI, Prometheus finds the IP.
When you stop pinging the 192.168.2.1 Proxmox connections will also fail after 2-3 sec.
I suspect some problem with routing on the Proxmox side but I have no idea how to confirm it and troubleshoot. (as I said connections to other hosts in the VPN network works totally fine that's why I think its Proxmox issue, for example pinging the OPNSense VM or other test Linux VMs in 192.168.2.0/24 works all the time)
Here is trace route from 192.168.0.6 (test VM) to 192.168.2.10 (Proxmox host) when there is no other ping triggered:
traceroute to 192.168.2.10 (192.168.2.10), 30 hops max, 60 byte packets
1 192.168.0.1 (192.168.0.1) 0.342 ms 0.328 ms 0.323 ms
2 10.10.20.2 (10.10.20.2) 98.365 ms 102.322 ms 102.318 ms
3 * * *
4 * * *
5 * * *
6 * * *
7 * * *
8 *
And here is same trace route when I ping 192.168.2.1 in a background from even a totally different host:
traceroute to 192.168.2.10 (192.168.2.10), 30 hops max, 60 byte packets
1 192.168.0.1 (192.168.0.1) 0.654 ms 0.623 ms 0.618 ms
2 10.10.20.2 (10.10.20.2) 29.749 ms 29.745 ms 29.741 ms
3 192.168.2.10 (192.168.2.10) 47.574 ms 49.494 ms 51.544 ms
What kind of magic is going on here? Any ideas?
FYI I have one different WG VPN running and 2 others OpenVPN connections setup in very similar way (but without Proxmox on the other side) and everything works perfectly fine.
I have 2 physical locations in different cities.
Location A:
192.168.0.0
192.168.0.1 - OPNSense dedicated router with WG enabled
192.168.0.6 - test Linux VM
Location B:
192.168.2.0
192.168.2.1 - ISP router (cannot be changed)
192.168.2.10 - Proxmox host with running VMs, firewall disabled on host and dc level
192.168.2.11 - OPNsesne VM and WG enabled on it, only LAN used
Overall -> The Wireguard works fine.
Ping results:
192.168.0.6 -> 192.168.2.1 : OK
192.168.0.6 -> 192.168.2.11 : OK
BUT
when you try to ping Proxmox server itself from the remote location -> after aprox 5 tries -> it fails. Also you cannot connect via GUI. I have Prometheus monitoring setup, it also fails.
BUT when you run ping 192.168.2.1 (ISP router) in the same time -> it magically starts to work. Everything. You can Ping Proxmox, connect to GUI, Prometheus finds the IP.
When you stop pinging the 192.168.2.1 Proxmox connections will also fail after 2-3 sec.
I suspect some problem with routing on the Proxmox side but I have no idea how to confirm it and troubleshoot. (as I said connections to other hosts in the VPN network works totally fine that's why I think its Proxmox issue, for example pinging the OPNSense VM or other test Linux VMs in 192.168.2.0/24 works all the time)
Here is trace route from 192.168.0.6 (test VM) to 192.168.2.10 (Proxmox host) when there is no other ping triggered:
traceroute to 192.168.2.10 (192.168.2.10), 30 hops max, 60 byte packets
1 192.168.0.1 (192.168.0.1) 0.342 ms 0.328 ms 0.323 ms
2 10.10.20.2 (10.10.20.2) 98.365 ms 102.322 ms 102.318 ms
3 * * *
4 * * *
5 * * *
6 * * *
7 * * *
8 *
And here is same trace route when I ping 192.168.2.1 in a background from even a totally different host:
traceroute to 192.168.2.10 (192.168.2.10), 30 hops max, 60 byte packets
1 192.168.0.1 (192.168.0.1) 0.654 ms 0.623 ms 0.618 ms
2 10.10.20.2 (10.10.20.2) 29.749 ms 29.745 ms 29.741 ms
3 192.168.2.10 (192.168.2.10) 47.574 ms 49.494 ms 51.544 ms
What kind of magic is going on here? Any ideas?
FYI I have one different WG VPN running and 2 others OpenVPN connections setup in very similar way (but without Proxmox on the other side) and everything works perfectly fine.
Last edited: