If you are installing Guix as an unprivileged user, it is still possible to run guix-daemon provided you pass --disable-chroot. However, build processes will not be isolated from one another, and not from the rest of the system.
VERSION=`curl "https://api.github.com/repos/cli/cli/releases/latest" | \
grep '"tag_name"' | sed -E 's/.*"([^"]+)".*/\1/' | cut -c2-`; \
curl -sSL https://github.com/cli/cli/releases/download/v${VERSION}/gh_${VERSION}_linux_amd64.tar.gz | \
sudo tar xzfC - /usr/local --strip-components=1
Creating the image isn't hard, the trouble is the stupid installer never seems to work properly according to their directions on most system that I've tried (primarily Chromebooks, so YMMV). The only surefire method that has worked for me every single time is using dd
from the Ctrl+Alt+F2 shell (or one in a logged in account, but then that account gets copied over to the target system) or dd
or the 'python-awesome-dd' script from a Linux LiveUSB. If you use dd
you probably need to run the fix-grub.sh
script from the chromefy repo and the lab
folder. There is a short URL for it, curl -L goo.gl/HdjwAZ | bash -s /dev/sda
to grab the file and run it assuming you are in a root
shell.
You MUST be in Developer Mode if you are attempting to Chromefy an existing ChromeOS device. Then after the Dev Mode changeover wipes the device you should access the system shell (aka VTY) using Ctrl+Alt+F2 (usually the Forward key depending on the age of the device but may also be Refresh or something else, and you m
#!/bin/bash | |
echo_docs() { | |
cat <<'EOD' | |
EOD | |
} | |
# use case to echo `network` `port` `autoselect` or `LED` sections or if no arg output all of them | |
set_network () { |
Copy/download the quick build steps file and execute it with sh quick-build-steps.sh
and if you are on Debian and have Docker it should fetch the Dockerfile here which when run with the docker build
in the script clones scdtools and installs scdrand (and scdtotp) on your host system.
After you've installed it you can run sudo scdrand
to get try and add some entropy to the kernel pool.
#!/usr/bin/env bash | |
set -euo pipefail | |
usage () { | |
echo "Usage: $(basename $0) {up|down|[job stage]}" | |
echo "[job stage] - defaults to "build" and supports multiple comma delimited jobs" | |
echo "$(basename $0) test,build" | |
} | |
sudo apt install virt-manager # includes libvirt/virsh/etc
curl -L https://vagrantup.com/whatever/vagrant-version.deb
sudo dpkg -i vagrant*.deb
For the vagrant-libvirt plugin you may need to use an alternate ruby/vagrant path in the call to install the plugin.
virsh detach-device --domain macOS-Simple-KVM --live --conf | |
ig <( cat << 'EOF' | |
<input type='tablet' bus='usb'> | |
<alias name='input3'/> | |
<address type='usb' bus='0' port='2'/> | |
</input> | |
EOF | |
) | |
sleep 2 | |
virsh attach-device --domain macOS-Simple-KVM --live --conf |
See [bootstrap.sh] for quickest swupd
package bundle installation.
If you are tricksy you can curl
the raw [bootstrap.sh] file and examine it and then chmod +x bootstrap.sh
and run it yourself.
The bootstrap currently doesn't automatically force you into i3, but if you log out and select your user from the login screen and click the ⚙️ icon and select i3 from the list, then enter your password if you didn't already and login.