Its a ESXI host, do you not have access to the vmkern.. throw up some VM on the lan side network... Console that VM.. Or sure console to pfsense and disable the firewall..

@provels I think that the setting to have the Time Synchronization enabled in Integration Services fixes this. Since I enabled this setting, I have only seen the clock unsynchronized error at reboot.

@johnpoz I answered 'no' to that on my previous comment in the first line. I don't have too much idea about networking, sorry. I am just giving you as much information as I can.

hi Ingenium, I know it might be a bit late for an answer, but I think one solution for your VLAN on VF problem might be to bind a VLAN on a VF on the host. You can use "ip link set PF vf X vlan Y" on the host to bind a VLAN on the VF. Replace PF with your PF-interface name, X is the VF interface and Y is the VLAN. You would end up in having a VF for each additional VLAN. The VLAN header is stripped/inserted by the VF and the interface can be used just like a standard interface in pfSense. No need to configure VLAN in pfSense. check out https://doc.dpdk.org/dts/test_plans/vf_vlan_test_plan.html for reference.

@ldiciolla as xcp-ng version 7.x you must add 4 extra in the mtu 1500 cause the nic in xen has no 802.1q vlan driver due that the max vlan 4094 is done like 1024 2048 4096 4096 is how memory is allocated but the mtu 1500 is the tcp mac window max trans (m)utex just add 4 in the mtu and there the vlan number will be stored (in v8)this is covered , pfsense runs in a vm so don't allocate vlan(s in vlan(s on the gen1 hypervisor , buy a switch vlan802.1q built there your nic vlan in and deliver pfsense simple lan interfaces reason is the nic must also be vlan capavle, by using a switch this is covered and on one realtek i ran wan lan1,2,3,4,5,6 always tag the port the 802.1q has nothing to do with the ARP MAC thats in 802.1ad vlan 4096 tag means all vlans and start by id 4 due 1-3 is are used for LAG not like LACP or LACP.802.a3d ,

It seems I might have made a mistake in my virtual network configuration. I tried ssh'ing to the gateway 10.0.0.1, and lo and behold, an ssh server running. Turns out it was my virtualization host listening on that address. A reboot of the router must have made it also on that address (is that possible?) temporarily. > virsh net-dumpxml lan_priv <network connections='2'> <name>lan_priv</name> <uuid>567ca017-512e-4211-87c7-ae0193806d20</uuid> <bridge name='virbr1' stp='on' delay='0'/> <mac address='52:54:00:2e:3d:0f'/> **<ip address='10.0.0.1' netmask='255.255.240.0'>** ^^^^^ (oops) </ip> </network> I believe it should be **"10.0.0.0"** for the network ip address. I'll clean everything up and report the results.

Here are the iperf3 tests results: Test#1 ubuntu:~$ iperf3 -4 -c bouygues.iperf.fr Connecting to host bouygues.iperf.fr, port 5201 [ 4] local 192.168.5.102 port 45186 connected to 89.84.1.222 port 5201 [ ID] Interval Transfer Bandwidth Retr Cwnd [ 4] 0.00-1.00 sec 3.33 MBytes 27.9 Mbits/sec 55 260 KBytes [ 4] 1.00-2.00 sec 2.72 MBytes 22.8 Mbits/sec 0 305 KBytes [ 4] 2.00-3.00 sec 3.09 MBytes 25.9 Mbits/sec 0 335 KBytes [ 4] 3.00-4.00 sec 3.65 MBytes 30.6 Mbits/sec 0 352 KBytes [ 4] 4.00-5.00 sec 3.52 MBytes 29.5 Mbits/sec 0 360 KBytes [ 4] 5.00-6.00 sec 3.52 MBytes 29.5 Mbits/sec 0 361 KBytes [ 4] 6.00-7.00 sec 3.65 MBytes 30.6 Mbits/sec 0 361 KBytes [ 4] 7.00-8.00 sec 3.71 MBytes 31.1 Mbits/sec 0 363 KBytes [ 4] 8.00-9.00 sec 3.58 MBytes 30.1 Mbits/sec 0 368 KBytes [ 4] 9.00-10.00 sec 3.83 MBytes 32.2 Mbits/sec 0 381 KBytes [ ID] Interval Transfer Bandwidth Retr [ 4] 0.00-10.00 sec 34.6 MBytes 29.0 Mbits/sec 55 sender [ 4] 0.00-10.00 sec 32.9 MBytes 27.6 Mbits/sec receiver iperf Done. Test#2 ubuntu:~$ iperf3 -4 -c bouygues.iperf.fr Connecting to host bouygues.iperf.fr, port 5201 [ 4] local 192.168.5.102 port 45206 connected to 89.84.1.222 port 5201 [ ID] Interval Transfer Bandwidth Retr Cwnd [ 4] 0.00-1.00 sec 4.95 MBytes 41.5 Mbits/sec 154 415 KBytes [ 4] 1.00-2.00 sec 4.45 MBytes 37.3 Mbits/sec 0 481 KBytes [ 4] 2.00-3.00 sec 4.88 MBytes 41.0 Mbits/sec 0 526 KBytes [ 4] 3.00-4.00 sec 5.25 MBytes 44.1 Mbits/sec 0 557 KBytes [ 4] 4.00-5.00 sec 5.50 MBytes 46.1 Mbits/sec 0 574 KBytes [ 4] 5.00-6.00 sec 5.62 MBytes 47.2 Mbits/sec 0 584 KBytes [ 4] 6.00-7.00 sec 5.87 MBytes 49.3 Mbits/sec 0 585 KBytes [ 4] 7.00-8.00 sec 5.69 MBytes 47.7 Mbits/sec 0 585 KBytes [ 4] 8.00-9.00 sec 5.81 MBytes 48.7 Mbits/sec 0 585 KBytes [ 4] 9.00-10.00 sec 5.99 MBytes 50.3 Mbits/sec 0 592 KBytes [ ID] Interval Transfer Bandwidth Retr [ 4] 0.00-10.00 sec 54.0 MBytes 45.3 Mbits/sec 154 sender [ 4] 0.00-10.00 sec 51.2 MBytes 42.9 Mbits/sec receiver iperf Done. Thanks N

@heper Thank you. My home network is pretty small and my use cases generally don't put much of a load on anything. Also, I only have a 200 Mbs WAN connection so guessing that pfSense performance on the hardware I have ordered won't be an issue. That said, it's good to know that some folks may consider EXSi to offer better performance than Proxmox and I will do some research in this area too. Thank you again.

Sounds like you should probably move to a proxmox forum.

I see no problem here. One nic connected to a v switch of type external (for example called INET). Connect the wan if of the pfsense vm to that. Make sure the host os cannot us it. create a new v switch of type external (called Internal) and connect one physical nic to it. last, connect the host os, lan nic of pfsense and any other internal vm to that v switch

@keitetran The reason you would like different lans could be policies. Binding them all together at the esxi level (in software) could work, but again why?

Hi awebster, you're right, i have solved my issue. the origin of my problem is in igbn driver. with the vmlinux driver "igb" all is fine. you can find all details here : https://communities.vmware.com/message/2905942#2905942 thanks for your reply

Ah, OK. That is certainly possible. I was confused by the discussion of 'teaming' which is not what you want at all. You have two choices here: Pass though 3 three NIC to pfSense and bridge them. That will give your the behaviour you want but as (I now see) you initially said it's a very inefficient way of creating a switch. Bridging can be awkward in pfSense. Create a vswitch in hyper-v and connect all 3 to it. It's still doing it in software which is not as good as a real switch but at least it's trying to be a switch rather than a bridge. That also simplifies the pfSense config significantly that would then only have a single LAN interface internally connected to the vswitch. Steve

It was working for two days. After i woke up, no internet... I had to restart pfSense and had a new public IP and it was working again. I217 is always unidentified... Nobody here has pfSense running in Hyper-V on Windows 10?

That's odd. i350 is very widely used. Unless it was one of the many fake Intel cards that are about...

@Riano-desouza did that work for you?