Contributing to libvirt
This page provides guidance on how to contribute to the
libvirt project
The libvirt project is always looking for new contributors to
participate in ongoing activities. While code development is a
major part of the project, assistance is needed in many other
areas including documentation writing, bug triage, testing,
application integration, website / wiki content management,
translation, branding, social media and more. The only
requirement is an interest in virtualization and desire to
help.
The following is a non-exhaustive list of areas in which
people can contribute to libvirt. If you have ideas for
other contributions feel free to follow them.
- Software development. The core library / daemon (and
thus the bulk of coding) is written in C, but there are
language bindings written in Python, Perl, Java, Ruby,
Php, OCaml and Go. There are also higher level wrappers
mapping libvirt into other object frameworks, such GLib,
CIM and SNMP. For those interested in working on the core parts of
libvirt, the contributor guidelines are
mandatory reading
- Translation. All the libvirt modules aim to support
translations where appropriate. All translation is
handling outside of the normal libvirt review process,
using the Fedora
instance of the Zanata tool. Thus people wishing
to contribute to translation should join the Fedora
translation team
- Documentation. There are docbook guides on various
aspects of libvirt, particularly application development
guides for the C library and Python, and a virsh command
reference. There is thus scope for work by people who are
familiar with using or developing against libvirt, to
write further content for these guides. There is also a
need for people to review existing content for copy editing
and identifying gaps in the docs
- Website / wiki curation. The bulk of the website is
maintained in the primary GIT repository, while the wiki
site uses mediawiki. In both cases there is a need for
people to both write new content and curate existing
content to identify outdated information, improve its
organization and target gaps.
- Testing. There are a number of tests suites that can run
automated tests against libvirt. The coverage of the tests
is never complete, so there is a need for people to create
new test suites and / or provide environments to actually
run the tests in a variety of deployment scenarios.
- Code analysis. The libvirt project has access to the coverity
tool to run static analysis against the codebase, however,
there are other types of code analysis that can be useful.
In particular fuzzing of the inputs can be very effective
at identifying problematic edge cases.
- Security handling. Downstream (operating system) vendors
who distribute libvirt may wish to propose a person to
be part of the security handling team, to get early access
to information about forthcoming vulnerability fixes.
- Evangalism. Work done by the project is of no benefit
unless the (potential) user community knows that it
exists. Thus it is critically important to the health
and future growth of the project, that there are a people
who evangalise the work created by the project. This can
take many forms, writing blog posts (about usage of features,
personal user experiences, areas for future work, and more),
syndicating docs and blogs via social media, giving user
group and/or conference talks about libvirt.
- User assistance. Since documentation
is never perfect, there are inevitably cases where users
will struggle to attain a deployment goal they have, or
run into trouble with managing an existing deployment.
While some users may be able to contact a software vendor
to obtain support, it is common to rely on community help
forums such as libvirt users
mailing list, or sites such as
stackoverflow.
People who are familiar with libvirt and have ability &
desire to help other users are encouraged to participate in
these help forums.
For full details on contacting other project contributors
read the contact page. There
are two main channels that libvirt uses for communication
between contributors:
The project has a number of
mailing lists for
general communication between contributors.
In general any design discussions and review
of contributions will take place on the mailing
lists, so it is important for all contributors
to follow the traffic.
Contributors to libvirt are encouraged to join the
IRC channel used by
the project, where they can have live conversations
with others members.
Since 2016, the libvirt project directly participates as an
organization in the Google Summer of Code program. Prior to
this the project had a number of students in the program
via a joint application with the QEMU project. People are
encouraged to look at both the libvirt and QEMU programs
to identify potentially interesting projects to work on.