
Create a Git branch from an earlier commit
Start a branch at an earlier commit or release tag, keep detached-HEAD work, and understand what happens when the branch later merges into main.
Focused answers for definitions, architecture choices, troubleshooting, pricing, and evaluation.

Start a branch at an earlier commit or release tag, keep detached-HEAD work, and understand what happens when the branch later merges into main.

Split Filebeat inputs and modules into separate YAML files, enable live reloading, and keep output settings in the main configuration.

Use Git remotes, mirrors, and bundles to preserve committed history, while accounting for uncommitted files, LFS objects, repository metadata, and restore tests.

Find a surviving branch tip in local refs, the reflog, or remaining objects, then recreate the branch. Recovery depends on object availability, not a fixed deadline.

Use git add --patch to stage selected changes, inspect the complete staged snapshot, and commit it without accidentally including the rest of the file.

Choose endpoint diffs, merge-base diffs, or commit ranges to compare Git branches, then generate and check a patch when you need to transfer file changes.

Create and push a specific Git tag, distinguish annotated tags from signatures and releases, and handle an incorrect release marker.

Choose a local or remote-tracking branch as a starting point, control upstream tracking, and understand how the new branch diverges.

Export staged, unstaged, new, and binary files as a Git patch; check application, handle conflicts, and preserve commits with format-patch.

In modern Git, both stage additions, modifications, and deletions. Their path scope differs when you run them from a subdirectory.

Fetch a Git tag, make a shallow clone or source archive, preserve detached commits, and verify the revision used for deployment.

Clone nested Git submodules, check out recorded commits, deliberately update dependencies, and configure private-repository access in CI.