Ah, nice! I run that at 80 on an i5-6400T for reference. Seems about optimal for my use but that's just a test box, it's not always on.

Good question, I've never seen one to find out. I imagine it would have to be in the QAT driver or possibly the crypto framework.

Thanks for all

I found a link on YouTube:
https://www.youtube.com/watch?v=q2mI0myAEXA
I will try booting from the Linux kernel.
If this is done perfectly, we will discuss the rest of the hardware.
It will be a fun experience

@stephenw10 said in Intel x550-t2 can not disable flow control properly:

hw.ix.flow_control=0

Thank you, this one worked 👍

@glippi Hello, after swapping out the intel 82599 10 gigabit dual port and moving this to the onboard ethernet with PCI passthrough to the VM, it seems everything is stable now.
the AHCI errors, however consistent with disks (as far as I can see), it seems that in this case it was with the ethernet controller.
My guess it was just a faulty card as since moving over I have had no issues now for 18 days.

If you would like any logs, please let me know which one you would like to see for further investigation on this.

@stephenw10 Thank you very much for the detailed response.

@stephenw10 said in NIC issues and advise:

It could be in a shellcmd:
https://docs.netgate.com/pfsense/en/latest/development/boot-commands.html?highlight=shellcmd#shellcmd-option

thanks i managed to find it and remove it.

issue has not happened in a week or so now.

Next step is to figure out the realtek nic issue.

The igc driver/hardware can only be autoselect. For some reason setting it to that specifically can cause it to go into a loop of some sort as your saw.

You could try setting it to 2.5G which will prevent it linking at 1G. It still negotiates that though so if the modem for some reason stops allowing 2.5G it will just fail to link.

@maxster This is the one that I used.

https://store.ui.com/us/en/pro/collections/unifi-accessory-tech-cable-sfp/products/10gbps-direct-attach-cable?variant=UACC-DAC-SFP10-0.5M

@MikeSpears @stephenw10

I confirmed I can issue commands directly from a shell to communicate with the A125 display with pfSense temporarity installed on an external USB SSD on my QNAP TS-653A.

Same shell scripts I created for TrueNAS CORE: https://github.com/elvisimprsntr/QnapFreeLCD

#!/bin/sh NAME="`hostname -s`" stty -f /dev/cuau1.lock 1200 stty -f /dev/cuau1.init 1200 stty -f /dev/cuau1 1200 # clear display printf '\115\015' > /dev/cuau1 sleep 1 # turn on display printf '\115\136\001' > /dev/cuau1 sleep 1 # print hostname on line 1 printf '\115\014\000 '$NAME > /dev/cuau1 sleep 1 # print message on line 2 printf '\115\014\001 System Booting' > /dev/cuau1

It seems there might be a bug in the icp_a106 driver

Mar 2 13:13:11 LCDd 20455 Could not open driver module /usr/local/lib/lcdproc/icp_a106.so: /usr/local/lib/lcdproc/icp_a106.so: Undefined symbol "ftime"

There was a commit in 2019 that appears to have addressed the issue, but was never merged.

https://github.com/lcdproc/lcdproc/commit/781b3113a592d75a29aa5024a94c2fd6b4592f87

Mar 2 13:13:11 LCDd 20298 LCDd version 0.5.9 starting Mar 2 13:13:11 LCDd 20298 Using Configuration File: /usr/local/etc/LCDd.conf Mar 2 13:13:11 LCDd 20455 Listening for queries on 127.0.0.1:13666 Mar 2 13:13:11 LCDd 20455 Could not open driver module /usr/local/lib/lcdproc/icp_a106.so: /usr/local/lib/lcdproc/icp_a106.so: Undefined symbol "ftime" Mar 2 13:13:11 LCDd 20455 Driver [icp_a106] binding failed Mar 2 13:13:11 LCDd 20455 Could not load driver icp_a106 Mar 2 13:13:11 LCDd 20455 There is no output driver Mar 2 13:13:11 LCDd 20455 Critical error while initializing, abort. Mar 2 13:13:12 php 20660 lcdproc: Start client procedure. Error counter: (0) Mar 2 13:13:13 php 20660 lcdproc: Failed to connect to LCDd process Connection refused (61) Mar 2 13:13:13 php 20660 lcdproc: Start client procedure. Error counter: (1) Mar 2 13:13:14 php 20660 lcdproc: Failed to connect to LCDd process Connection refused (61) Mar 2 13:13:14 php 20660 lcdproc: Start client procedure. Error counter: (2) Mar 2 13:13:15 php 20660 lcdproc: Failed to connect to LCDd process Connection refused (61) Mar 2 13:13:15 php 20660 lcdproc: Start client procedure. Error counter: (3) Mar 2 13:13:16 php 20660 lcdproc: Failed to connect to LCDd process Connection refused (61) Mar 2 13:13:16 php 20660 lcdproc: Too many errors, stopping client.

UPDATE

Submitted issue 211

Re: Installing pfSense on F5 BIG-IP 5050
Hi guys, so what you are looking for is the id "14e4" this is a plx for the network card ..

In principle you need the NSL from Broadcom, you can find an open src

"https://github.com/Broadcom-Switch/OpenNSL"

You could look starting at linux-kernel-bde.. this is open ofc, but the F5/Arista/NetApp/ Other big vendors have the full pkg .. because you need an NDA .. and other things .. in the past was a little easy to get it .. now I don't know if you still can.

For everyone's info...

I bought one of these, it has been working GREAT for me so far, week +

https://www.amazon.com/dp/B0C339KVH9?psc=1&ref=ppx_yo2ov_dt_b_product_details

Right, if you are running iperf3 on pfSense you will always see a lower result because it's optimised as a router not a server.

To see what it's capable of you need to test through it. So run the iperf server and client on two different devices connected to two separate interfaces in pfSense.

@Jarhead

From the WEB GUI:

I was able to move the WAN to the 2.5Gb igc0 interface.

1)First I disabled the WAN.

2)Then I reassigned the interface to igc0

3)Finally I reactivated the WAN

I also did a system reboot

aa288371-1db8-47b9-a9e1-e11795fe9b1d-image.png

@Unoptanio

Its a big Machine 👍

@robsonvitorm
First off, to be hyper specific, for my tests HW is a Lenovo m720q with a AJ??929 x8 Riser Adapter and an HP/HPE 331T (Broadcom 5719 4 port)
pfSense on Bare Metal = All 4 Ports work as expected, no PHY error on boot.
pfSense on ProxMox VE 8.1 (Debian based OS) = All 4 ports work as expected, Intel_IOMMU enabled in Grub
pfSense on Ubuntu 22.04 LTS (Debian based OS) = Only bge0 works, bge1 fails to init. Intel_IOMMU and MSR enabled in Grub

Intel_IOMMU:
Edit /etc/default/grub
Append GRUB_CMDLINE_LINUX_DEFAULT with additional "intel_iommu=on" at the end.

MSRs:
Edit /etc/default/grub
Append GRUB_CMDLINE_LINUX_DEFAULT with additional "kvm.ignore_msrs=1" at the end.

After either change:
sudo grub-mkconfig -o /boot/grub/grub.cfg

On the Ubuntu Host, I also blacklisted the bge drivers so the Bare Metal Ubuntu did not load them and attach to the NIC Ports:
EDIT/Create: /etc/modprobe.d/vfio.conf
blacklist tg3
options vfio-pci ids=14e4:1657
# Prevent Broadcom bg3 driver from loading in Bare Metal OS - keep card ports free for VM passthrough
Then push changes to initram: sudo update-initramfs -u
This step did not seem to be needed with Proxmox.

These mods to the OS came from a thread I found on GPU passthrough, which I adapted for NIC passthrough:
https://askubuntu.com/questions/1406888/ubuntu-22-04-gpu-passthrough-qemu
(Though I just noticed that I did not do Step 14, so will have to swap that drive back in and see if it makes any difference)

-Casper

I mean....anything is possible. 😉

That's intended to be a Plus feature though so it's not something I can help you with.

I have a buddy who used to write Linux drivers for a big company, though not sure how much he knows about BSD.
I might hit him up in a few days if I can't figure it out.

If pfSense somehow supported Docker/Podman this would be a non issue, but since it's BSD kernel I don't see that happening.

Always good to details efforts in an edge case. Someone else will be trying this. 😉