Devise system to back up all project data stored on GitHub

[andrewdavidwong]

(Branched from #3958 (comment))

The Qubes OS Project has a lot of important data stored (only) on GitHub. This is a single point of failure. We should devise a system (ideally an automated one) to back this data up regularly. Examples of data that should be backed up:

• Issues (text, labels, milestones, closing and reopening actions)
• Pull requests (message if different from commit message, merging and closing actions)
• Comments (on issues, commits, PRs, and everywhere else)

[marmarek]

I've downloaded data from github-issues repo using migration API. It's ~200MB tar.gz containing a set of json files with issues data, comments etc. Plus a directory with issues attachments. And a git repo itselt.
Looks like a good method for a backup. I consider publishing the archive, but need to make sure there is no private data inside first. For example threre is users.json.

Creating the archive (on github side) took like 10min, for this single repo. Including all of them, will probably take much more time. But it shouldn't be a problem.

[marmarek]

I've downloaded data from github-issues repo using migration API. It's ~200MB tar.gz containing a set of json files with issues data, comments etc. Plus a directory with issues attachments. And a git repo itselt.
Looks like a good method for a backup. I consider publishing the archive, but need to make sure there is no private data inside first. For example threre is users.json.

Creating the archive (on github side) took like 10min, for this single repo. Including all of them, will probably take much more time. But it shouldn't be a problem.