From d326d3428da89b943bb5f1d4d396f21b3e999ff7 Mon Sep 17 00:00:00 2001 From: Stan Hu Date: Mon, 29 Aug 2016 07:46:30 -0700 Subject: Optimize branch lookups and force a repository reload for Repository#find_branch If `git gc` runs and `Repository` has an instance to `Rugged::Repository`, a bug in libgit2 may cause the instance to return a stale value or a missing branch. This change not only optimizes the branch lookup so we don't have to iterate through every branch, but it also works around the `git gc` issue by forcing a repository reload every time `Repository#find_branch` is called. See: https://github.com/libgit2/libgit2/issues/1534 Closes #15392, #21470 --- app/models/repository.rb | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) (limited to 'app/models') diff --git a/app/models/repository.rb b/app/models/repository.rb index 91bdafdac99..f891e8374d2 100644 --- a/app/models/repository.rb +++ b/app/models/repository.rb @@ -120,8 +120,21 @@ class Repository commits end - def find_branch(name) - raw_repository.branches.find { |branch| branch.name == name } + def find_branch(name, fresh_repo: true) + # Since the Repository object may have in-memory index changes, invalidating the memoized Repository object may + # cause unintended side effects. Because finding a branch is a read-only operation, we can safely instantiate + # a new repo here to ensure a consistent state to avoid a libgit2 bug where concurrent access (e.g. via git gc) + # may cause the branch to "disappear" erroneously or have the wrong SHA. + # + # See: https://github.com/libgit2/libgit2/issues/1534 and https://gitlab.com/gitlab-org/gitlab-ce/issues/15392 + raw_repo = + if fresh_repo + Gitlab::Git::Repository.new(path_to_repo) + else + raw_repository + end + + raw_repo.find_branch(name) end def find_tag(name) -- cgit v1.2.1