Slower Network on LXC/Linux VM's than Windows VM's

[Jebula999]

Hi All,

I recently set up a speed tracker, but was confused by the results.

I have a 750/500 Fibre line at home.
These are results from various devices/VM's

Home PC (The one I'm using now)


On my PVE box, I have 2 VM's and an LXC running tests:

LXC - HomeAssistant (HAOS)​

​

​

VM - SpeedTest Tracker (Ubuntu)​

​

​

VM - Windows Server (...Windows)​

​

​

​

As you can see, the Linux VM and LXC are performing far worse than the Windows VM.
The Windows VM is less than my home PC due to not having a CAT6 cable I imagine.

But all 3 tests above are running on the same PVE machine.
Using the same network connection.



Any ideas as to what could be causing this?
Or what/where would I need to look to find out?

 

[Jebula999]

Just a slight bump to this.

I went ahead and purchased CAT6 cable for the server.
I am still getting the same results.

Definitely related the the individual VM's.

 

[_gabriel]

remote speedtest server seems different from Windows and Linux.
try same remote server.

 

[sw-omit]

Also, what are the (Virtual) hardware settings of the Windows and Linux devices?
Either Screenshot the hardware-tabs, or post the results of

qm config 101
qm config 200

 

[Jebula999]

Also, what are the (Virtual) hardware settings of the Windows and Linux devices?
Either Screenshot the hardware-tabs, or post the results of

qm config 101
qm config 200

qm config 101

agent: 1
balloon: 0
bios: ovmf
boot: order=ide2;scsi0
cores: 6
cpu: host
efidisk0: local-lvm:vm-101-disk-0,efitype=4m,pre-enrolled-keys=1,size=4M
hostpci0: 0000:07:00,pcie=1,x-vga=1
ide0: local:iso/virtio-win-0.1.240.iso,media=cdrom,size=612812K
ide2: none,media=cdrom
machine: pc-q35-8.1
memory: 6144
meta: creation-qemu=8.1.5,ctime=1713375115
name: Plex
net0: e1000=BC:24:11:DE:D8:4B,bridge=vmbr0,firewall=1
numa: 0
onboot: 1
ostype: win10
scsi0: local-lvm:vm-101-disk-1,cache=writeback,discard=on,size=64G
scsihw: virtio-scsi-pci
smbios1: uuid=c546c3d5-5536-4b97-a46d-1725a5fab6bb
sockets: 1
startup: order=3,up=30
vga: none
vmgenid: 16e933b3-7026-4cb7-a5ec-7e46d040e3ca

qm config 200

agent: 1
balloon: 0
bios: ovmf
boot: order=scsi0
cores: 3
cpu: host
description: Home Assistant OS
efidisk0: local-lvm:vm-200-disk-0,efitype=4m,size=4M
localtime: 1
memory: 4096
meta: creation-qemu=8.1.5,ctime=1714254350
name: homeassistant
net0: virtio=02:55:02:F3:1E:F4,bridge=vmbr0
numa: 0
onboot: 1
ostype: l26
scsi0: local-lvm:vm-200-disk-1,cache=writethrough,discard=on,size=32G,ssd=1
scsihw: virtio-scsi-pci
smbios1: uuid=a680ca6d-e69a-4523-b4c9-38e3833b0da8
sockets: 1
startup: order=2,up=30
tablet: 0
tags: proxmox-helper-scripts
usb0: host=10d7:b012
usb1: host=1a86:55d4
vmgenid: 3fda3e36-d6ab-4f3c-9b81-f5ac3aa5bf87

 

[sw-omit]

Thanks.
Looking at these, I don't see something jumping out at me as being "wrong" (in fact, the only thing I would have done differently would be on the Windows's side, which doesn't have the issue, with selecting the virtio network card instead of the e1000, but that's more personal preference + experience from ESXi)

As _Gabriel mentioned, it could also be a difference in server. If you can't find how to select the same server, one option you could go for (excluding your router from interfering too), would be to run Ntttcp or iperf on your (windows) PC in server-mode and then running it in client mode on both windows and one of the linux boxes, and see what the difference is then.

Other then that I'm probably out of idea's / suggestions for now.