Windows ME BSODs (Exception, System Busy, and WARNING) Captured directly from VMware Workstation today.
I had found in the past that it was a bit hard to find clean, unedited, pixel-perfect copies of the Blue Screens of Death, so when my Windows ME VM crashed today I knew it was time to save these for the future. These are pretty much the same as the ones that would appear in Windows 95 and Windows 98.
First the classic “An exception 0D has occurred” BSOD
Next, System is busy
and last but not least, WARNING system has become unstable
Installed a new Ubiquiti USG at an office downtown today, and noticed that in the 8 hours since the system was first up, it had detected over 800 nearby networks. I decided to analyze the data a bit for fun.
First, the channel. I was surprised to see that over half of all detected APs were on channel 6.
It does make sense that there are far more 2.4GHz APs detected, since it has better signal penetration.
Next up, security standards.
Nearly 90% of APs are using WPA2 of some sort, and just over 10% are open. Less than 1% use WPA.
Now, arguably the most interesting – the AP manufacturer, according to OUI lookup.
Note: some OUIs were not recognized, so the dataset is slightly less than 800 here.
In the manufacturer breakdown we se a lot of the usual brands – Cisco, Aruba and Ubiquiti are in the top. Technicolor, Sagemcom, ASUS, HP, Juniper, Sonicwall are all also common network vendors. But what of the others?
The biggest “unusual” vendor we see is Mitsumi. Mitsumi is generally known as an OEM that manufactures PC peripherals and input devices – mice, keyboards, floppy and optical drives, and quite a few game consoles. It’s not surprising that they would make WiFi radios, but I wouldn’t expect their OUI to be used as an OEM.
Looking a bit further at the data, most of the Mitsumi networks’ SSIDs are in the format “WiFi Hotspot 0000” (where 0000 is a random 4 digits). However, a few of them had names such as “Cruze”, “Volt”, “Equinox” and “Malibu”. So, apparently Mitsumi manufactures the WiFi radio for the GM OnStar car hotspots.
Continuing on with the less-known OUI vendors, we also see Visteon, Continen (Continental), Harman/B (Harman/Becker), and AlpsElec (Alps Electric).
Alps, like Mitsumi, is an OEM known for PC peripherals – particularly keyboards and laptop touchpads. In this case, the SSIDs for the Alps APs are all some variant of “MB Hotspot 000” – so they are Mercedes-Benz car hotspots.
So, unsurprisingly, the other 3 are also car electronics OEMs.
Visteon – spun off from Ford in 2000, they specialize in car infotainment and other electronics systems. Continental Automotive Systems – The electronics systems branch of German company Continental Tire. Harman/Becker – a division of Samsung, specialized in car electronics, resulting from the Harman company’s acquisition of Becker, a German car radio manufacturer.
Today’s takeaway: a LOT of cars have WiFi hotspots built-in these days!
Got frustrated that this form wasn’t available as a proper PDF form, so made it myself. Intended only as a convenience for those who wish to use it. PDF is not password-protected or signed.
Download a copy of the 4383-80E pdf form (Patient Enrolment and Consent to Release Personal Health Information) below.
I’ve found multiple guides on how to enable NVIDIA GPU access from lxc containers, however I had to combine the information from multiple sources to get a fully working setup. Here are the steps that worked for me.
Install dkms on your Proxmox host to ensure the nvidia driver can be auto-updated with new kernel versions. # apt install dkms
Download the latest nvidia binary driver supported by nvidia-patch. Head over to https://github.com/keylase/nvidia-patch and get the latest supported driver version listed there.
Download the nvidia-patch repo git clone https://github.com/keylase/nvidia-patch.git
Install the driver from step 1 on the host. For example, ./NVIDIA-Linux-x86_64-455.45.01.run
Run the nvidia-patch/patch.sh script on the host.
Install the same driver in each container that needs access to the Nvidia GPU, but without the kernel module. ./NVIDIA-Linux-x86_64-455.45.01.run --no-kernel-module
Run the nvidia-patch/patch.sh script on the lxc container.
On the host, create a script to initialize the nvidia-uvm devices. Normally these are created on the fly when a program such as ffmpeg calls upon the GPU, but since we need to pass the device nodes through to the containers, they must exist before the containers are started.
I saved the following script as /usr/local/bin/nvidia-uvm-init. Make sure to chmod +x !
#!/bin/bash
## Script to initialize nvidia device nodes.
## https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#runfile-verifications
/sbin/modprobe nvidia
if [ "$?" -eq 0 ]; then
# Count the number of NVIDIA controllers found.
NVDEVS=`lspci | grep -i NVIDIA`
N3D=`echo "$NVDEVS" | grep "3D controller" | wc -l`
NVGA=`echo "$NVDEVS" | grep "VGA compatible controller" | wc -l`
N=`expr $N3D + $NVGA - 1`
for i in `seq 0 $N`; do
mknod -m 666 /dev/nvidia$i c 195 $i
done
mknod -m 666 /dev/nvidiactl c 195 255
else
exit 1
fi
/sbin/modprobe nvidia-uvm
if [ "$?" -eq 0 ]; then
# Find out the major device number used by the nvidia-uvm driver
D=`grep nvidia-uvm /proc/devices | awk '{print $1}'`
mknod -m 666 /dev/nvidia-uvm c $D 0
mknod -m 666 /dev/nvidia-uvm-tools c $D 0
else
exit 1
fi
Next, we create the systemd service files to start this script, and the nvidia-persistenced.
# NVIDIA Persistence Daemon Init Script
#
# Copyright (c) 2013 NVIDIA Corporation
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
#
# This is a sample systemd service file, designed to show how the NVIDIA
# Persistence Daemon can be started.
#
[Unit]
Description=NVIDIA Persistence Daemon
Wants=syslog.target
[Service]
Type=forking
ExecStart=/usr/bin/nvidia-persistenced --user nvidia-persistenced
ExecStopPost=/bin/rm -rf /var/run/nvidia-persistenced
[Install]
WantedBy=multi-user.target
Next, symlink the two service definition files into /etc/systemd/system cd /etc/systemd/system # ln -s /usr/local/lib/systemd/system/nvidia-uvm-init.service # ln -s /usr/local/lib/systemd/system/nvidia-persistenced.service and load the services # systemctl daemon-reload # systemctl start nvidia-uvm-init.service # systemctl start nvidia-persistenced.service
Now you should see all the nvidia device nodes have been created # ls -l /dev/nvidia* crw-rw-rw- 1 root root 195, 0 Dec 6 18:07 /dev/nvidia0 crw-rw-rw- 1 root root 195, 1 Dec 6 18:10 /dev/nvidia1 crw-rw-rw- 1 root root 195, 255 Dec 6 18:07 /dev/nvidiactl crw-rw-rw- 1 root root 195, 254 Dec 6 18:12 /dev/nvidia-modeset crw-rw-rw- 1 root root 511, 0 Dec 6 19:00 /dev/nvidia-uvm crw-rw-rw- 1 root root 511, 0 Dec 6 19:00 /dev/nvidia-uvm-tools
/dev/nvidia-caps: total 0 cr-------- 1 root root 236, 1 Dec 6 18:07 nvidia-cap1 cr--r--r-- 1 root root 236, 2 Dec 6 18:07 nvidia-cap2
Check the dri devices as well # ls -l /dev/dri* total 0 drwxr-xr-x 2 root root 100 Dec 6 17:00 by-path crw-rw---- 1 root video 226, 0 Dec 6 17:00 card0 crw-rw---- 1 root video 226, 1 Dec 6 17:00 card1 crw-rw---- 1 root render 226, 128 Dec 6 17:00 renderD128
Take note of the first number of each device after the group name. In the listings above I have 195, 511, 236 and 226.
Now we need to edit the lxc container configuration file to pass through the devices. Shut down your container, then edit the config file – example /etc/pve/lxc/117.conf. The relevant lines are below the swap: 8192 line
Now, start your container back up. You should be able to use NVENC features. You can test by using ffmpeg: $ ffmpeg -i dQw4w9WgXcQ.mp4 -c:v h264_nvenc -c:a copy /tmp/rickroll.mp4
You should now have working GPU transcode in your lxc container!
If you get the following error, recheck and make sure you have set the correct numeric values for lxc.cgroup.devices.allow and restart your container.
[h264_nvenc @ 0x559f2a536b40] Cannot init CUDA
Error initializing output stream 0:0 -- Error while opening encoder for output stream #0:0 - maybe incorrect parameters such as bit_rate, rate, width
or height
Conversion failed!
Another way to tell the values are incorrect is having blank (———) permission lines for the nvidia device nodes.
$ ls -l /dev/nvidia*
---------- 1 root root 0 Dec 6 18:04 /dev/nvidia0
crw-rw-rw- 1 root root 195, 255 Dec 6 19:02 /dev/nvidiactl
---------- 1 root root 0 Dec 6 18:04 /dev/nvidia-modeset
crw-rw-rw- 1 root root 511, 0 Dec 6 19:02 /dev/nvidia-uvm
crw-rw-rw- 1 root root 511, 1 Dec 6 19:02 /dev/nvidia-uvm-tools
# Install the OpenSSH Server
Add-WindowsCapability -Online -Name OpenSSH.Server~~~~0.0.1.0
# Set service to Automatic startup
Set-Service -Name sshd -StartupType 'Automatic'
# Start service
Start-Service sshd
# Confirm the Firewall rule is configured. It should be created automatically by setup.
Get-NetFirewallRule -Name *ssh*
Update: Even after doing all this the system still locks up randomly when using the amdgpu driver.
I’m running dual AMD FirePro W2100 driving 3 monitors in my workstation. Since installing the cards I’ve been suffering random freezes/graphical lockups that seemed to be related to 3D. They occurred typically during an animation in gnome-shell, or when using Firefox or Chrome with hardware acceleration. Most times, I was able to recover by logging in to the machine via ssh and sending killall -HUP to the appropriate process (usually gnome-shell). Every time this happened, syslog would be full of GPU faults:
I tried updating my kernel (going from Ubuntu 18.04 to 18.10 and even reinstalling with Pop!_OS 18.10 than 19.04) and updating graphics drivers using the oibaf ppa, to no avail. Finally found what seems to be the solution on HackerNews (thanks danieldk) – force the use of the newer amdgpu driver rather than the older radeon driver. The W2100 is a first-generation GCN chip and so is supported by both drivers, and radeon is chosen as the default. To force amdgpu, you need to pass the kernel flags
In Ubuntu, add these to GRUB_CMDLINE_LINUX_DEFAULT in /etc/default/grub, then run update_grub. Pop!_OS doesn’t use grub, so you need to add each flag using kernelstub -a amdgpu.si_support=1 and repeating for each of the 4.
So far, my system seems stable since this change. I will update this post if anything changes.
Install pulseaudio-dlna from Ubuntu default repos, then run pulseaudio-dlna. The first time, it failed, but I think that’s because the receiver was switched off. Console output did show it had been detected though. After switching on the receiver then running pulseaudio-dlna again, it redirected my currently playing audio to the receiver after a few seconds! So far this is the first time I’ve managed to get DLNA live streaming working with this receiver in 5 years of owning it.
Note: this post is more intended for personal reference; as such the quality of writing is not up to my standard, but I’ve decided to publish it as-is anyway since I don’t have time to review and figure it could be useful to someone.
I was looking for a good portable Linux laptop to that I could carry around without being too worried about battery life or the device getting scratched up in my bag. I tried a few different devices, including the HP Folio 13, ThinkPad X131e, GPD Pocket, Dell Latitude E7240. Finally I settled on the Acer C720 Chromebook. The main points in favour of this device are, for me:
Compact, thin and light (11.6″)
Relatively low cost (4GB RAM model can be had for under $150 CAD)
Proper modern low-voltage Intel Core processor (Celeron 2957U – Haswell) and not an Atom core or high-end i7
Dual-band Wi-Fi and Bluetooth
and the main downsides:
Basic TN LCD panel
16GB storage
Runs Chrome OS
Chrome OS keyboard layout
The good news is the downsides are fairly easily remedied.
TN LCD Panel
The Acer C720 uses a standard 11.6″ LED backlit eDP panel. I was able to find an IPS panel to replace it on laptopscreen.com (model N116BCA-EA1). The panel replacement is fairly straightforward; care needs to be taken not to break the LCD bezel or any of the clips.
16GB SSD
The SSD on the C720 is a standard M.2 2242 SATA SSD. The M.2 slot is easily accessible after removing the Chromebook’s bottom cover.
Runs Chrome OS
The Chrome OS/Linux dual-boot process is done by using the chrx script (https://github.com/reynhout/chrx). Remember to remove the write-protect screw while the bottom cover is off.
Chrome OS keyboard layout
This was arguably the hardest part. I chose to run Xubuntu, so the steps I went through apply to that environment. Chrx by default provides good defaults, mapping the back, forward, refresh, brightness and volume keys. The big issue for me was the placement of the power button (above the backspace key)
Prevent systemd-logind from handling the power button
edit /etc/systemd/logind.conf and add the following line:
HandlePowerKey=ignore
Prevent xfce power manager from handling the power button
xfconf-query -c xfce4-power-manager -p /xfce4-power-manager/logind-handle-power-key -s true
Allow xfce power manager to handle lid switch
xfconf-query -c xfce4-power-manager -p /xfce4-power-manager/logind-handle-lid-switch -s false
Remap power key to Delete
xmodmap -e “keycode 124 = Delete”
To find the keycode, stop lightdm, startx to start a basic X session, then use xev to get the keycode for the power key.
Other tweaks
Consistent custom wallpaper/background for boot splash, login, and lock screen
gnome-screensaver uses the Gnome DE’s wallaper setting. Configure this via CLI:
gsettings set org.gnome.desktop.background picture-uri file:///usr/local/share/wallpaper/animevillage.jpg
lightdm-gtk-greeter can be configured using GUI tool lightdm-gtk-greeter-settings or by editing /etc/lightdm/lightdm-gtk-greeter.conf
plymouth/boot splash:
go to /usr/share/plymouth/themes and create a copy of the xubuntu-logo directory. I called mine xubuntu-logo-mod
sudo cp -R xubuntu-logo xubuntu-logo-mod
within the new xubuntu-logo-mod directory, rename and edit config files accordingly
325 cd xubuntu-logo-mod/
326 ls
327 sudo mv xubuntu-logo.plymouth xubuntu-logo-mod.plymouth
328 sudo mv xubuntu-logo.script xubuntu-logo-mod.script
329 sudo nano xubuntu-logo-mod.script
replace wallpaper.png in xubuntu-logo-mod directory with desired image. I used ImageMagick’s convert tool to convert my jpg format image to png.
add new theme we have created as an “alternative” with higher priority
Find out priority and config for current theme
sudo update-alternatives –query default.plymouth
Name: default.plymouth
Link: /usr/share/plymouth/themes/default.plymouth
Status: auto
Best: /usr/share/plymouth/themes/xubuntu-logo-mod/xubuntu-logo-mod.plymouth
Value: /usr/share/plymouth/themes/xubuntu-logo-mod/xubuntu-logo-mod.plymouth
Create new alternative with higher priority
sudo update-alternatives –install /usr/share/plymouth/themes/default.plymouth default.plymouth /usr/share/plymouth/themes/xubuntu-logo-mod/xubuntu-logo-mod.plymouth 200
Where “/usr/local/bin/monitorlayout.sh” is the path to the script saved from arandr.
For sddm, instead edit /usr/share/sddm/scripts/Xsetup.sh and add a line to have it run the same “/usr/local/bin/monitorlayout.sh” script.
#!/bin/sh
# Xsetup - run as root before the login dialog appears
# set screen layout using Xrandr
/usr/local/bin/monitorlayout.sh