Purpose

• The purpose of this KB article is to provide instructions for performing repository backups, both on a per-repository basis as well as in bulk.
• As per our backup policy, we perform hourly snapshot backups of your repositories and store these in AWS - our support team is able to restore your core repository content within a 7-day window of data loss.

Procedure #1 - Single repository backup

If you are seeking to perform a local backup of a singular (or a few) repositories, this can most simply be achieved by performing a mirror clone.

• A mirror clone is a complete backup clone of your repository, it contains the full commit history, the working directory, and all the extended refs of the remote repository - it will also maintain remote branch tracking configuration.

• To perform a mirror clone - you would simply add the --mirror flag to your regular clone command against the repository that you wish to backup:
  

  // Clone via SSH
  git clone --mirror git@bitbucket.org:workspaceID/repositoryslug.git
  // Clone via HTTPS
  git clone --mirror https://username@bitbucket.org/workspaceID/repositoryslug.git

Procedure #2 - Bulk repository backup

If you are seeking to perform a local backup of all of your remote repositories, this can be achieved in bulk by making use of a script - for example:

We have an unofficial script that can be used to handle this, along with its instructions located here: Bulk backup script

• Before executing the bash script, you will need to create an App Password if you do not already have one - this is because the script uses our API to extract a list of the repositories belonging to your workspace

• Once you have an App Password, you will then need to ensure that the bash script is executable by giving it the necessary permissions:

  chmod 777 account_repo_backup.sh

• Once complete, you can then execute the script - including the necessary authentication and workspace location parameters as follows:

  ./account_repo_backup.sh <Username> <AppPassword> <WorkspaceID>