d829311a3a
Follow-up to #1826 to reflect new development occurring for Fedora CoreOS using coreos-assembler. Closes: #1831 Approved by: jlebon |
||
---|---|---|
.. | ||
.gitignore | ||
README.md | ||
setup.yml |
Developing using Podman + Vagrant
Note: These instructions are in flux and will be oriented toward a coreos-assembler based workflow, which targets Fedora CoreOS (FCOS). You may find built FCOS images at: http://artifacts.ci.centos.org/fedora-coreos/prod/builds/latest/.
The current tooling here is oriented towards doing builds inside a Fedora 29 pet container, with Vagrant on the host.
You should share the git working directory with the f29 container.
Assuming you have git repositories stored in /srv
, something like:
podman run --name f29dev --privileged -v /srv:/srv --net=host -ti registry.fedoraproject.org/fedora:29 bash
You can start the Vagrant box. To work around "fuse-sshfs" not being built into the Vagrant box, do something like this:
vagrant up; vagrant provision; vagrant halt; vagrant up
Note to run vagrant
as non-root, you'll need to either
add your user to the libvirt
group, or configure polkit.
If working under the /srv
directory, you should first do
sudo chown -R <your-user> /srv/path/to/rpm-ostree/
on your host so
that rootless vagrant
can write there.
Before building, you may need to install the build dependencies in the f29dev container. A quick way to do this is:
ci/installdeps.sh
Now, run autoreconf inside the f29dev container:
./autogen.sh CFLAGS='-ggdb -O0' --prefix=/usr --libdir=/usr/lib64 --enable-installed-tests --enable-gtk-doc
To sync over and install the built binaries to the Vagrant VM:
make vmsync
You may also want to use vmcheck
, like this:
make vmoverlay && make vmcheck
Also see HACKING.md.