Pulp lives on GitHub. The “pulp” repository is for the platform, and then each supported content family (like “rpm” and “puppet”) has its own repository.
Pulp uses a version scheme x.y.z. Pulp’s branching strategy is designed for
bugfix development for older
x.y release streams without interfering with
development or contribution of new features to a future, unreleased
x.y release. This strategy encourages a clear separation of bugfixes and
features as encouraged by Semantic Versioning.
Pulp’s branching model is inspired by the strategy described by Vincent Driessen in this article, but is not identical to it.
This is the latest bleeding-edge code. All new feature work should be done out
of this branch. Typically this is the development branch for future, unreleased
x.y release will have one corresponding branch called
example, all work for the 2.7.z series of releases gets merged into
Within each repository, the
master branch, any branch ending in
-dev, and any
branch ending in
-release should be marked as
on GitHub. The basic protection that disallows force-push and deletion is the
only option that should be enabled. There should be no restrictions on required
status checks or who can push. There is a script at
that will mark all appropriate branches as protected. Any time new branches are
created that should be protected, that script can be run to do the work.
Alpha and Beta releases will be built from the tip of an
x.y-dev branch. If
the beta fails testing, blocking issues will have fixes merged to the
x.y-dev branch like any other bug fix, and then a new build will be made.
Other changes unrelated to the blocking issues may get merged to the
x.y-dev branch between builds, and no effort will be made to “freeze” the
branch. Any such unrelated changes will be included in the next beta build.
Release candidates will be built from the most recent beta tag. GA releases will be built from the most recent release candidate tag.
When a hotfix needs to be made, a branch will be created from the most recent
x.y.z release tag. The fix will be made (via pull request from a personal
fork to the hotfix branch), a new tag will be built from the tip of the hotfix
branch, and the hotfix branch can be merged to
Bug Fix Branches¶
When creating a Pull Request (PR) that fixes a specific bug, title the PR as you would the git commit message.
Similar to bug fix branches, when creating a pull request that holds features until they are merged into a development branch, the pull request branch should be a brief name relevant to the feature. For example, a branch to add persistent named searches might be named “feature/named-searches”.
Choosing an Upstream Branch¶
When creating a bug fix or feature branch, it is very important to choose the right upstream branch. The general rule is to always choose the oldest upstream branch that will need to contain your work.
Commit messages in Pulp should contain a human readable explanation of what was fixed in the commit. They should also follow the standard git message format of starting with a subject line or title (usually wrapped at about 50 chars) and optionally, a longer message (usually wrapped at 72 characters) broken up into paragraphs. For more on what constitutes a good commit message, we recommend Tim Pope’s blog post on the subject.
It’s also recommended that every commit message in Pulp reference an issue in Pulp’s Redmine issue tracker. To do this you should use both a keyword and a link to the issue.
To reference the issue (but not change its state), use
re #123 ref #123
To update the issue’s state to MODIFIED and set the %done to 100, use
fixes #123 closes #123
You can also reference multiple issues in a commit:
fixes #123, #124
Putting this altogether, the following is an example of a good commit message:
Update node install and quickstart The nodes install and quickstart was leaving out an important step on the child node to configure the server.conf on the child node. closes #1392 https://pulp.plan.io/issues/1392
Cherry-picking and Rebasing¶
Don’t do it! Seriously though, this should not happen between release branches. It is a good idea (but not required) for a developer to rebase his or her development branch before merging a pull request. Cherry-picking may also be valuable among development branches. However, master and release branches should not be involved in either.
The reason is that both of these operations generate new and unique commits from the same changes. We do not want pulp-x.y and master to have the same bug fix applied by two different commits. By merging the same commit into both, we can easily verify months later that a critical bug fix is present in every appropriate release branch and build tag.
If you are not sure what “rebasing” and “cherry-picking” mean, Pro Git by Scott Chacon is an excellent resource for learning about git, including advanced topics such as these.