Terrance's questions

This morning 2 of my Ubuntu installs that are both 22.04.3 LTS running HWE upgraded to Kernel 6.5.0-14-generic. After rebooting both systems the network cards that are RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller did not come up. Booting to the older 6.2.0-39-generic Kernel the network card works just fine. Upon trying to reinstall the r8168-dkms driver which installs the driver version 8.049.02 fails. And running sudo lshw -C network is showing the card as UNCLAIMED.

P.S. I will not use the r8169 driver for my r8168 chipset as it will work OK in the beginning but after a little bit will give unpredictable behavior, some to the point of needing a hard reboot of my system to bring the interface back up.

My system is running 20.04.3 LTS, and I am using it as a Plex Media Server (amongst other services). I am using a bonded network for adaptive load balancing on the system. I am bumping into an issue with the system when it receives a new kernel update.

There are 2 network cards in the system. One is in a slot (RTL8169 PCI Gigabit Ethernet Controller) and the other is onboard (RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller). The issue I am bumping into is that the RTL8111/8168/8411 does not like the r8169 driver (unpredictable things will happen like not being able to communicate with the internet) so I have to run the following commands to unbind the r8169 driver and bind the r8168 driver:

address=$(lspci | awk '/8168/ {print "0000:"$1}')
echo $address | sudo tee /sys/bus/pci/drivers/r8169/unbind
echo $address | sudo tee /sys/bus/pci/drivers/r8168/bind

After running these commands the driver shows up correctly in the lshw command and I don't have any issues when the right driver is set for the right card:

terrance@Intrepid:~$ sudo lshw -C network
  *-network                 
       description: Ethernet interface
       product: RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller
       vendor: Realtek Semiconductor Co., Ltd.
       physical id: 0
       bus info: pci@0000:02:00.0
       logical name: eth0
       version: 03
       serial: c8:3a:35:d6:ec:a0
       size: 1Gbit/s
       capacity: 1Gbit/s
       width: 64 bits
       clock: 33MHz
       capabilities: pm msi pciexpress msix vpd bus_master cap_list ethernet physical tp 10bt 10bt-fd 100bt 100bt-fd 1000bt-fd autonegotiation
       configuration: autonegotiation=on broadcast=yes driver=r8168 driverversion=8.048.00-NAPI duplex=full latency=0 link=yes multicast=yes port=twisted pair slave=yes speed=1Gbit/s
       resources: irq:25 ioport:d800(size=256) memory:fdfff000-fdffffff memory:fdff8000-fdffbfff
  *-network
       description: Ethernet interface
       product: RTL8169 PCI Gigabit Ethernet Controller
       vendor: Realtek Semiconductor Co., Ltd.
       physical id: 5
       bus info: pci@0000:03:05.0
       logical name: eth1
       version: 10
       serial: bc:5f:f4:79:66:fe
       size: 1Gbit/s
       capacity: 1Gbit/s
       width: 32 bits
       clock: 66MHz
       capabilities: pm bus_master cap_list rom ethernet physical tp mii 10bt 10bt-fd 100bt 100bt-fd 1000bt 1000bt-fd autonegotiation
       configuration: autonegotiation=on broadcast=yes driver=r8169 driverversion=5.13.0-30-generic duplex=full latency=64 link=yes maxlatency=64 mingnt=32 multicast=yes port=twisted pair slave=yes speed=1Gbit/s
       resources: irq:20 ioport:e800(size=256) memory:febffc00-febffcff memory:febc0000-febdffff
  *-network
       description: Ethernet interface
       physical id: 1
       logical name: bond0
       serial: bc:5f:f4:79:66:fe
       capabilities: ethernet physical
       configuration: autonegotiation=off broadcast=yes driver=bonding driverversion=5.13.0-30-generic duplex=full firmware=2 ip=10.0.0.220 link=yes master=yes multicast=yes

This lasts through reboots just fine until the next kernel update. But, if I run those exact same commands in a script it tells me "No such device"

terrance@Intrepid:~/bin$ sudo ./driver_fix.bsh 
0000:02:00.0
tee: /sys/bus/pci/drivers/r8169/unbind: No such device
0000:02:00.0
tee: /sys/bus/pci/drivers/r8168/bind: No such device

What would be the best way that I can make this permanent so either it makes this change during startup or it changes only when a kernel update happens?

Thanks!

I understand that there are or may be similar questions on here, but this one does not pertain to firejail nor does Firefox have any delays when starting.

OS Info:

Xubuntu 20.04
DISTRIB_DESCRIPTION="Ubuntu 20.04.3 LTS"
NAME="Ubuntu"
VERSION="20.04.3 LTS (Focal Fossa)"
VERSION_ID="20.04"
$ uname -a
Linux terrance-ubuntu 5.11.0-40-generic #44~20.04.2-Ubuntu SMP Tue Oct 26 18:07:44 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

I run a script from a crontab that will launch my ADP login page and clock me in and out at specific times of the day. This used to work flawlessly until Firefox 94 was released. Now, I understand that for remote Mozilla decided to no longer go with X11 but D-Bus instead. For the life of me, I cannot figure out what is meant by using D-Bus instead of X11, other than they claim that it is simpler to use. I am assuming that this may be due to Wayland which I do not use.

If I run the following script from a command line terminal at the specific times it works perfect, but if I run the script from the crontab I get the following message:

The script (still a work in progress):

#!/bin/bash

#This function checks the path of the app on a Mac.
realpath1() {
        [[ $1 = /* ]] && echo "$1" || echo "$PWD/${1#./}"
}

#This function matches the day of the week and returns 0 if match, 1 if weekend.
function dowcheck(){
case " ${daysofweek[@]} " in
    *\ ${DOW}\ *)
        return 0;;
    *)
        return 1;;
esac
}

#This function matches if the clock in or out time is a match with 0 or 1 if not.
function timecheck(){
case " ${timesofday[@]} " in
    *\ ${HM}\ *)
        return 0;;
    *)
        return 1;;
esac
}

#This function matches days off to today.  If a match return 0 meaning day off, 1 means not a day off.
function daysoffcheck(){
case " ${daysoff[@]} " in
    *\ ${daymdy}\ *)
        return 0;;
    *)
        return 1;;
esac
}

#Check the OS type.
OS_TYPE=$(uname -a | awk '{print $1}')
if [[ ${OS_TYPE} == "Linux" ]]; then
        OS=$(grep -i ^name= /etc/*release | awk -F= '{print $2}' | sed 's/\"//g')
else
        OS=$(system_profiler SPSoftwareDataType | awk '/System Version:/ {print $3}')
fi
if [ "${OS}" = "CentOS Linux" ]; then
        OS=Fedora
fi

#Set working directories and set Display for running in a CRONJOB.
case $OS in
        macOS) apppath=/Applications/Firefox.app/Contents/MacOS
        export DISPLAY="/private/tmp/com.apple.launchd.*/org.macosforge.xquartz:0"
                PWD=$(dirname $(realpath1 $(which $0)));;
        *) apppath=/usr/bin
        DM=$(/usr/bin/basename $(/bin/cat /etc/X11/default-display-manager))
        case $DM in
            lightdm)
                export DISPLAY=:0;;
            gdm3)
                grep -E "# AutomaticLogin|AutomaticLoginEnable = false" /etc/$DM/*.conf >/dev/null && export DISPLAY=:1 || export DISPLAY=:0;;
            *);;
        esac
                PWD=$(dirname $(realpath $(which $0)));;
esac

#Set variables for matching functions.
DOW=$(date +%a)
HM=$(date +%H:%M)
daymdy=$(date +%m-%d-%Y)
#If today is newer than day off remove last day off.
if [[ "${daymdy}" > "$(head -1 $PWD/daysoff.txt)" ]]; then
    sed -i '1d' $PWD/daysoff.txt
fi

#Declare arrays.
declare -a daysofweek=('Mon' 'Tue' 'Wed' 'Thu' 'Fri')
declare -a timesofday=('08:00' '12:00' '12:30' '16:30')
declare -a inout=('in' 'out for lunch' 'in from lunch' 'out for the day')
declare -a daysoff=($(cat $PWD/daysoff.txt))

#Get in or out.
for i in "${!timesofday[@]}"; do
    if [[ "${timesofday[$i]}" == "${HM}" ]]; then
        inorout="${inout[$i]}";
    fi;
done

#Run functions and return 0 or 1.
daysoffcheck
doff=$?
dowcheck
dow=$?
timecheck
time=$?

#Finish up and send information or launch Firefox if need be.
if [[ $doff != "1" ]]; then
        echo "Today is a day off!  Why are you trying to clock in?"
        exit 1
elif [[ $dow != "0" ]]; then
        echo "It's the weekend!  Why are you trying to clock in?"
        exit 1
elif [[ $time != "0" ]]; then
        echo "It is $DOW at $HM.  It is not time to clock in or out."
        exit 1
else
        echo "It's ${HM}. Time to clock ${inorout}." | mail -s "Time clock" [email protected]
        echo "It's ${HM}. Time to clock ${inorout}." | mail -s "Time clock" [email protected] 
        xdotool mousemove --sync 677 1011
        $apppath/firefox --new-tab https://workforcenow.adp.com/workforcenow/login.html &
        $PWD/clock_in_out.bsh
        wait
fi

If anyone has any ideas that I could make Firefox work with D-Bus like it was with X11 before version 94 I would greatly appreciate it!

The xfce4-weather-plugin for Xubuntu 18.04 and now 20.04 has stopped working and gives "No Data".

This is a Q&A for when you are using the Graphics Drivers PPA at greater than the 390 driver as well as 18.04.3 LTS.

I am running Ubuntu 18.04.3 and I have a NVIDIA GeForce GTX 750 Ti. My video driver currently is nvidia-driver-440 that I installed from the Graphics Drivers PPA and I did not use the ubuntu-drivers autoinstall for the driver.

Recently my kernel updated to 5.3.0-26-generic and my video resolution is now stuck at 640x480. And it appears as though my video driver is installed, but did not load with the kernel.

Is there anyway to fix this?

EDIT: I have added a new answer below for Ubuntu 20.04 LTS and CUDA 11.1.1. See below.

I have a system with a NVIDIA card that has a compute support of 3.5+ compared on https://developer.nvidia.com/cuda-gpus. How do I install CUDA and the NVIDIA drivers in Ubuntu without downloading the .deb files from NVIDIA?

Update 2019-01-09: Please see answer below as there is a way to fix this with CUPS 2.2.8.

Update 2020-10-17: Upgrading from 18.04 LTS to 20.04 LTS will fix it as it upgrades the version of CUPS from 2.2.7 (which contains the bug) to version 2.3.1 which does not have the bug. Since some users may not be able to upgrade to 20.04 LTS right now this is still very valid for the 18.04 LTS users.

I have Ubuntu 18.04 LTS installed clean on this system. When I try to add a printer through CUPS 2.2.7 address of http://localhost:631 in the Chrome browser, I can't add the printer as I keep getting Unauthorized. It never asks for a login. How do I add my printer through CUPS using Google Chrome?

BTW, Firefox, Chromium and other browsers so far seem to work fine for adding a printer in CUPS 2.2.7. It appears to be a bug between CUPS 2.2.7 and Google Chrome.

This is a Q&A that I recently have figured out, so I have posted the answer below.

After installing Windows 10, every time I get a kernel update or I run the update-grub2 it always shows Windows 7 or Windows Recovery Environment instead of Windows 10. How do I fix this permanently?