[SOLVED] VMs/Containers receiving "Destination Host Unreachable" despite having an IP

[egg82]

I have a network 10.0.0.0/24 with a proxmox on 10.0.0.8

I created an Ubuntu server 20.04 VM 10.0.0.9 and an Ubuntu server 20.04 container 10.0.0.7
The proxmox machine is hooked up to a Unifi 16 XG L2 10Gbit switch using a single Gbit line for testing
That switch is hooked into a Unifi UDMP firewall/router/etc that goes to the ISP

/etc/network/interfaces looks like this:

auto lo
iface lo inet loopback

iface enp4s0 inet manual

auto vmbr0
iface vmbr0 inet static
        address 10.0.0.8/24
        gateway 10.0.0.1
        bridge-ports enp4s0
        bridge-stp off
        bridge-fd 0

The container looks like this: https://i.imgur.com/wHzB3U0.png
And the VM: https://i.imgur.com/6z4CR24.png https://i.imgur.com/vCnghlK.png

The controller shows clients are available, and the VM even has an IP.
https://i.imgur.com/5qb7k3S.png

Pinging the VM from the proxmox works, and so does pinging the proxmox from the VM. Pinging the gateway or the outside world or any other device on the local network is a bust, though. Pinging the VM from the gateway or from another device results in timeouts.

From the VM: https://i.imgur.com/cffkLhs.png
From the proxmox: https://i.imgur.com/s8UvBq3.png

Telling the VM to get a DHCP address results in no IP being received.

Anyone have any ideas? I'm at a bit of a loss. It acts like the proxmox created its own isolated network that so happens to also be 10.0.0.0/24 but that shouldn't be the case with the current configuration.

 

[egg82]

At this point I've re-installed proxmox with default config, just in case I meddled with something earlier. Still the same problem.

I also moved the IP range from 10.0xxx to 10.60xxx which is my network's DMZ. The plan was to move to this network eventually anyway, and it seems like I have exactly the same issues on both.

I spun up a live Kali ISO which has all the tools needed to diagnose this pre-installed. I gave to the IP 10.60.0.115 (new gateway IP is 10.60.0.1 and new proxmox IP is 10.60.0.3)

Here's the result of ip, traceroute, and route commands from the Kali machine: https://i.imgur.com/CUFJ27N.png

The hardware list for this proxmox:
AMD Ryzen 9 3900X
Asus Tuf Gaming X570-Pro
NZXT C850 80+ Gold PSU
2x Samsung 870 EVO 1TB
Samsung 980 PRO 500GB
G.Skill 3600 CL16 2x16GB
2x Seagate Ironwolf 2TB

The mobo comes with a 2.5Gbit port but even attached to the 10Gbit port on the switch with a CAT6 cable it reads as 1Gbit. That's fine, I was planning on using a dual-10Gbit NIC for this anyway, but I'm thinking perhaps that's a cause of the issue?

 

[egg82]

Out of curiosity I went out and bought a USB3 to Ethernet adapter (1Gbit, "j5Create") and hooked it up to one of the USB3 ports in the back. I then re-installed Proxmox just to be extra sure, and used the USB dongle in the installation step instead of the onboard NIC. Created a test container, and suddenly everything pings out and I can SSH into the containers and VMs with no other network changes in the Unifi controller.

So.. Solved, I suppose? Having support for 2.5Gbit out of the box would be nice. I don't particularly want to build and re-build kernel modules every update. Otherwise, yay!