MAINTENANCE MODE

Due to the move of Bitcoin Core to Guix, this repository is switching to maintenance mode. Only serious bugs (including security issues) will be considered going forward.

Gitian

Read about the project goals at the project home page.

This package can do a deterministic build of a package inside a VM.

Deterministic build inside a VM

This performs a build inside a VM, with deterministic inputs and outputs. If the build script takes care of all sources of non-determinism (mostly caused by timestamps), the result will always be the same. This allows multiple independent verifiers to sign a binary with the assurance that it really came from the source they reviewed.

Prerequisites:

Arch:

sudo pacman -S python2-cheetah qemu rsync
sudo pacman -S lxc libvirt bridge-utils # for lxc mode

From AUR:

apt-cacher-ng (you may have to play with permissions (chown to apt-cacher-ng) on files to get apt-cacher-ng to start)
debootstrap
dpkg
gnupg1
multipath-tools (for kpartx)

Non-AUR packages:

debian-archive-keyring (for making Debian guests)
ubuntu-keyring (for making Ubuntu guests)

From newroco on GitHub:

vmbuilder

Also, I had to modify the default /etc/sudoers file to uncomment the secure_path line, because vmbuilder isn't found otherwise when the env -i ... sudo vmbuilder ... line is executed (because the i flag resets the environment variables including the PATH).

Gentoo:

layman -a luke-jr  # needed for vmbuilder
sudo emerge dev-vcs/git net-misc/apt-cacher-ng app-emulation/vmbuilder dev-lang/ruby
sudo emerge app-emulation/qemu
export KVM=qemu-system-x86_64

Ubuntu:

This pulls in all pre-requisites for KVM building on Ubuntu:

sudo apt-get install git apache2 apt-cacher-ng python-vm-builder ruby qemu-utils

If you'd like to use LXC mode instead, install it as follows:

sudo apt-get install lxc

If you'd like to use docker mode instead, install it as follows:

sudo apt-get install docker-ce

Debian:

See Ubuntu, and also run the following on Debian Jessie or newer:

sudo apt-get install ubuntu-archive-keyring

On Debian Wheezy you run the same command, but you must first add backports to your system, because the package is only available in wheezy-backports.

OSX with MacPorts:

sudo port install ruby coreutils
export PATH=$PATH:/opt/local/libexec/gnubin  # Needed for sha256sum

OSX with Homebrew:

brew install ruby coreutils
export PATH=$PATH:/opt/local/libexec/gnubin

VirtualBox:

Install virtualbox from http://www.virtualbox.org, and make sure VBoxManage is in your $PATH.

Debian Guests

Gitian supports Debian guests in addition to Ubuntu guests. Note that this doesn't mean you can allow the builders to choose to use either Debian or Ubuntu guests. The person creating the Gitian descriptor will need to choose a particular distro and suite for the guest and all builders must use that particular distro and suite, otherwise the software won't reproduce for everyone.

To create a Debian guest:

bin/make-base-vm --distro debian --suite jessie

There is currently no support for LXC Debian guests. There is just KVM support. LXC support for Debian guests is planned to be added soon.

Only Debian Jessie guests have been tested with Gitian. If you have success (or trouble) with other versions of Debian, please let us know.

If you are creating a Gitian descriptor, you can now specify a distro. If no distro is provided, the default is to assume Ubuntu. Since Ubuntu is assumed, older Gitian descriptors that don't specify a distro will still work as they always have.

Create the base VM for use in further builds

NOTE: requires sudo, please review the script

KVM

bin/make-base-vm
bin/make-base-vm --arch i386

LXC

bin/make-base-vm --lxc
bin/make-base-vm --lxc --arch i386

Set the USE_LXC environment variable to use LXC instead of KVM:

export USE_LXC=1

Docker

bin/make-base-vm --docker
bin/make-base-vm --docker --arch i386

Set the USE_DOCKER environment variable to use DOCKER instead of KVM:

export USE_DOCKER=1

VirtualBox

Command-line VBoxManage must be in your $PATH.

Setup:

make-base-vm cannot yet make VirtualBox virtual machines ( patches welcome, it should be possible to use VBoxManage, boot-from-network Linux images and PXE booting to do it). So you must either get or manually create VirtualBox machines that:

Are named Gitian-<suite>-<arch> -- e.g. Gitian-xenial-i386 for a 32-bit, Ubuntu 16 machine.
Have a booted-up snapshot named Gitian-Clean . The build script resets the VM to that snapshot to get reproducible builds.
Has the VM's NAT networking setup to forward port localhost:2223 on the host machine to port 22 of the VM; e.g.:

    VBoxManage modifyvm Gitian-xenial-i386 --natpf1 "guestssh,tcp,,2223,,22"

The final setup needed is to create an ssh key that will be used to login to the virtual machine:

ssh-keygen -t rsa -f var/id_rsa -N ""
ssh -p 2223 ubuntu@localhost 'mkdir -p .ssh && chmod 700 .ssh && cat >> .ssh/authorized_keys' < var/id_rsa.pub

Then log into the vm and copy the ssh keys to root's authorized_keys file.

ssh -p 2223 ubuntu@localhost
# Now in the vm
sudo bash
mkdir -p .ssh && chmod 700 .ssh && cat ~ubuntu/.ssh/authorized_keys >> .ssh/authorized_keys

Set the USE_VBOX environment variable to use VBOX instead of KVM:

export USE_VBOX=1

Sanity-testing

If you have everything set-up properly, you should be able to:

PATH=$PATH:$(pwd)/libexec
make-clean-vm --suite xenial --arch i386

# on-target needs $DISTRO to be set to debian if using a Debian guest
# (when running gbuild, $DISTRO is set based on the descriptor, so this line isn't needed)
DISTRO=debian

# For LXC:
LXC_ARCH=i386 LXC_SUITE=xenial on-target ls -la

# For KVM:
start-target 32 xenial-i386 &
# wait a few seconds for VM to start
on-target ls -la
stop-target

Building

Copy any additional build inputs into a directory named inputs.

Then execute the build using a YAML description file (can be run as non-root):

export USE_LXC=1 # LXC only
bin/gbuild <package>.yml

or if you need to specify a commit for one of the git remotes:

bin/gbuild --commit <dir>=<hash> <package>.yml

The resulting report will appear in result/<package>-res.yml

To sign the result, perform:

bin/gsign --signer <signer> --release <release-name> <package>.yml

Where <signer> is your signing PGP key ID and <release-name> is the name for the current release. This will put the result and signature in the sigs/<package>/<release-name>. The sigs/<package> directory can be managed through git to coordinate multiple signers.

After you've merged everybody's signatures, verify them:

bin/gverify --release <release-name> <package>.yml

Poking around

Log files are captured to the var directory
You can run the utilities in libexec by running PATH="libexec:$PATH"
To start the target VM run start-target 32 xenial-i386 or start-target 64 xenial-amd64
To ssh into the target run on-target (after setting $DISTRO to debian if using a Debian guest) or on-target -u root
On the target, the build directory contains the code as it is compiled and install contains intermediate libraries
By convention, the script in <package>.yml starts with any environment setup you would need to manually compile things on the target

TODO:

disable sudo in target, just in case of a hypervisor exploit
tar and other archive timestamp setter

LXC tips

bin/gbuild runs lxc-execute or lxc-start, which may require root. If you are in the admin group, you can add the following sudoers line to prevent asking for the password every time:

%admin ALL=NOPASSWD: /usr/bin/lxc-execute
%admin ALL=NOPASSWD: /usr/bin/lxc-start

Right now lxc-start is the default, but you can force lxc-execute (useful for Ubuntu 14.04) with:

export LXC_EXECUTE=lxc-execute

Recent distributions allow lxc-execute / lxc-start to be run by non-privileged users, so you might be able to rip-out the sudo calls in libexec/*.

If you have a runaway lxc-start command, just use kill -9 on it.

The machine configuration requires access to br0 and assumes that the host address is 10.0.2.2:

sudo brctl addbr br0
sudo ifconfig br0 10.0.2.2/24 up

Tests

Not very extensive, currently.

python -m unittest discover test

devrandom / gitian-builder

readme