Search results

Looks good.

Well that tutorial is for hyper-v so a little bit different. What are the specs of your proxmox host? Obviously, many of us are running opnsense VMs and pfsense vms , I have been for a couple of years and just did a reinstall for 23.1 to confirm working before I updated my main opnsense router...

Does it boot with CPU host and removing all the other flags ... flags=+pcid;+spec-ctrl;+ssbd And without passthrough.

@tmbiv56 think of each Linux bridge as virtual managed switch with unlimited ports. The IP address and gateway directly assigned to it are for the Proxmox host connection to that virtual switch not the vm/LXC VM/LXC can all attach to that same virtual switch and comunicate with the Proxmox...

Yeah removing the linux bridges from proxmox altogether was not the right direction. try the below for your /etc/network/interfaces: auto lo iface lo inet loopback auto enp1s0 iface enp1s0 inet manual auto vmbr0 iface enp1s0 inet static address 192.168.999.20/24 gateway 192.168.999.1...

as long as the proxmox host default CIDR and gateway have been moved off of vmbr0 then yes it can be removed without causing issues. if not then proxmox internet access and your webgui access will likely be impacted.

So gigabit should be no problem and would not be worth any of the passthrough hassle for me. Even 10g I use virtual without issue and good throughput, your CPU starts to come into play there if marginal. Always use paravirtualized Virtio as the network device and not e1000.

Yes that is a good way to think about a dedicated linux bridge/Vlan. A separate virtual switch with basically unlimited ports. If you don't attach an egress (to either WAN or some other external physical/virtual network) or give the Proxmox host an IP on the bridge/vlan then the guest attached...

Unasked for advice, use paravirtualized VirtIO for the network devices instead of e1000. It’s faster and takes up less host cpu resources. Only use e1000 if some unusual requirement forces it. so you want to bridge OPNsense WAN (net1) to OPNsense OPT1(net2)? Seems unusual that you would not...

Latest zfs-2.1.8 list 6.1 compatibility.

Sometimes the shift from ifupdown to ifupdown2 can cause issues, quick check below if the interface naming did not change. https://forum.proxmox.com/threads/upgrade-to-7-no-network-after-reboot.101342/post-437163

Virtio uses less CPU and has a higher throughput ceiling than e1000, so the rule of thumb would be not to use e1000 ever unless some specific reason or incompatibility forces you. Just so things are clear, WAN plugs into which interface [eno1, enp1s0f(0,1,2,3)] on the proxmox node? Is WAN...

Easy stuff first. Use paravirtualized VirtIO and not e1000 for those opnsense network devices in Proxmox And turn off the Proxmox firewall for now on each as well. first picture from OPNsense webgui - why are your lan and wan in the same 192.168.0.0 subnet? 192.168.0.100 and 192.168.0.111

Should not need anything special on the proxmox side, once iommu passthrough is working properly. When setting up Vlans on a new OPNsense you are required to enable the parent interface (new requirement since the 22 and above series opnsense) even if it is not used otherwise.

By memory @spirit may have worked on ifupdown2. Wondering if he has anything to add here?

Did you check from the truenas side if the card shows up there? Suspect the network card is caught up with the passthrough.

Can you show the output of cat /etc/network/interfaces? Vmbr0, enp2s0f0, and enp2s0f1 all have ip addresses in the 192.168.1.0/24 subnet.

If you had a backup before changes, yes. Otherwise no.

Is your gateway from your ISP? you could try Routed Configuration or Masquerading (NAT) with iptables instead of the default Configuration using a bridge.

Well looks like the non-working server is now different than the working version. Does someone else have access and passwords to your server? An update to ifupdown2 will not add a bond0 or ifb1 and ifb2 interfaces.