iwlwifi fails occasionally with "Hardware error", "Queue stuck" and "Microcode SW error"

Question

I have Ubuntu 18.04 (kernel: 5.4.0-53-generic) installed on my laptop with Intel(R) Dual Band Wireless AC 9462, REV=0x354 as its wireless card. Sometimes I notice my wifi acting strangely with intermittent periods of extreme lag and sometimes disconnection and then failure to reconnect to the access point.

First I noticed the following two lines in the output of my dmesg (see the complete output here):

[    6.572592] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-jf-b0-50.ucode failed with error -2
[    6.572606] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-jf-b0-49.ucode failed with error -2

I noticed that iwlwifi-QuZ-a0-jf-b0-48.ucode is the last existing version in my /lib/firmware/. By downloading and adding iwlwifi-QuZ-a0-jf-b0-50.ucode and also 53, 55 and 59 to that directory I could fix the issue at first. So no problem was reported in dmesg:

mhr@mhr-Vostro-5490:~$ dmesg | grep iwlwifi
[    9.090633] iwlwifi 0000:00:14.3: enabling device (0000 -> 0002)
[    9.107845] iwlwifi 0000:00:14.3: TLV_FW_FSEQ_VERSION: FSEQ Version: 58.3.35.22
[    9.108038] iwlwifi 0000:00:14.3: loaded firmware version 50.3e391d3e.0 op_mode iwlmvm
[    9.165959] iwlwifi 0000:00:14.3: Detected Intel(R) Dual Band Wireless AC 9462, REV=0x354
[    9.290913] iwlwifi 0000:00:14.3: base HW address: 4c:1d:96:30:83:6a
[    9.663805] iwlwifi 0000:00:14.3 wlp0s20f3: renamed from wlan0

But after about a week the problem came back. My wifi works fine since the startup to some random while and then starts to act strangely (poor connection, low speed, disconnecting from the access point) and of course sometimes it never happens and everything is fine. As soon as the failures in my network starts, I can see the iwlwifi producing lengthy and inscrutable error messages reported in dmesg (see complete output):

[  123.483358] iwlwifi 0000:00:14.3: Hardware error detected. Restarting.
[  134.234398] iwlwifi 0000:00:14.3: Queue 2 is stuck 225 184
[  134.234633] iwlwifi 0000:00:14.3: Microcode SW error detected. Restarting 0x0.

Answer 1

To add some context. I also reguarly suffer from this. My adapter is a Wi-Fi 6 AX200 (v1). Errors are similar to this one. I only get this when I connect to my Mesh WiFI network. And not via a regular WiFI network. Not sure if that is related though.

Answer 2

I wrote a bash script to conveniently restart my iwlwifi module whenever the problem comes back.

I am sharing it here just as a temporary fix that worked for me. This script accepts input parameters to be passed to modprobe iwlwifi. I haven't still found a parameter that fixes the problem for sure, thus I leave it blank for now.

#!/bin/bash

params=""
delay=5

# if [ "$EUID" -ne 0 ]; then
if [ $USER != 'root' ]; then
    echo "Must be run as root!"
    exit
fi

# if [ $# -eq 0 ]; then
#     echo "Script run with no input arguments..."
# fi

for i in $*; do
    params=" $params $i"
done

echo "removing iwlwifi module..."
modprobe -r iwlwifi

if [ -z "$params" ]; then
    echo "iwlwifi will be reinserted with no additional settings specified..."
else
    echo "iwlwifi will be reinserted with the following params:$params"
fi

echo -n "reinserting in $delay seconds"
for ((i=0; i<$delay; i++)); do
    sleep 1
    echo -n .
done
echo

modprobe iwlwifi $params

echo "end of script" `basename "$0"`