This question is in reference to Atlassian Documentation: Automatic branch merging
I've studied this awesome document:
This suggests that release branches should be created only for the endgame of a release cycle. This is how we currently use release branches. However, you also have this automatic branch merging feature that applies only to release branches. If you have a "1.0" release branch for finishing up 1.0, but it's recommended you don't create a "2.0" release branch until the endgame for that release. Given that, when would branch forwarding actually be useful? By the time you create the 2.0 branch, 1.0 has long shipped so there would be nothing to commit to it that would need to get forwarded.
I'm missing something here, so I hope someone can help out!
With multiple release branches, automatic branch merging can be useful when you need to fix an issue on earlier, still supported release. Consider an example where the latest release is
release/2.0, but a critical bug was found in
release/1.0 and this was still a release being used and supported. With automatic branch merging, a fix could be made via a pull request targeting
release/1.0, and this fix would automatically cascade to all later release branches (including
release/2.0), and onto
master (assuming the development branch is set to
Stef has stated the main use case for our team perfectly. In addition, we also create our release branches slightly before the ship date (typically about a week). This way new/risky code not meant for the upcoming release can be merged directly to master to avoid contaminating the release, but any bugfixes/polish on the release branch can still be cascaded to master automatically.
Thanks, that helps! One of the reasons I was asking is that conceptually I like the idea of having a release branch for upcoming known releases and not waiting until the last few weeks. The explicit nature of saying "I'm committing this to 2.0" rather than putting it in master just because there's no release branch yet. I see the downsides (more merging), but it avoids the very common discussion I hear in the hallways ("what is master right now? is it release x or release y?"). Also, it seems easier to drop things on the floor if it always goes to an explicit release branch which cascades down to later releases and to master.
Bitbucket Pipelines helps me manage and automate a number of serverless deployments to AWS Lambda and this is how I do it. I'm building Node.js Lambda functions using node-lambda ...
Connect with like-minded Atlassian users at free events near you!Find a group
Connect with like-minded Atlassian users at free events near you!
Unfortunately there are no AUG chapters near you at the moment.Start an AUG
We're bringing product updates and pro tips on teamwork to ten cities around the world.Save your spot