Re: [egit-dev] EGit Corruption @ GitHub

["Shawn O. Pearce" <spearce@xxxxxxxxxxx>]

"Shawn O. Pearce" <spearce@xxxxxxxxxxx> wrote:
> Robin Rosenberg <robin.rosenberg@xxxxxxxxxx> wrote:
> > s??ndagen den 31 januari 2010 18.27.53 skrev  Chris Aniszczyk:
> > > Anyone have any insights on this?
> > >     http://help.github.com/egit-corruption/
> > 
> > Yes. This is https://bugs.eclipse.org/286653
> 
> I'm not so sure.

Hmmph.  I wish GitHub helped us more with troubleshooting the
software bugs their customers are running into, rather than just
blaming us.


Google searching for one of the commits in that error message leads
me to what looks like the original problem report that spawned this
help thread with GitHub:

  http://support.github.com/discussions/repos/1586-cloning-mrnfragments-fails

It seems the owner of that repository doesn't use EGit.  Yet he
managed to screw up the repository anyway by breaking a link
between two commits.  And screwing up his local repository.

If he doesn't use EGit... I don't think we can blame this particular
corruption problem on EGit.


I'm actually starting to wonder if GitHub is having some respository
corruption on their own side, that doesn't necessarily require
EGit to destroy the repository.  E.g. these support issues mention
that there is corruption, possibly caused by a fork being deleted,
but the owners haven't used anything other C Git:

  https://github.tenderapp.com/discussions/repos/471-git-upload-pack-aborting-due-to-possible-repository-corruption-on-the-remote-side
  https://github.tenderapp.com/discussions/repos/501-repository-corruption-git-push-fails-git-clone-fails
  https://github.tenderapp.com/discussions/repos/1389-remveeemacs-rails-mangled  

On the other hand, yes, this looks like the JGit based push failure
in bug 286653, and should be fixed by my commit:

 https://github.tenderapp.com/discussions/repos/513-git-pull-fatal-git-upload-pack-aborting-due-to-possible-repository-corruption-on-the-remote-side 


So, unless someone can point to another example, I think this isn't
an issue for us anymore, and once we release 0.7.0 we should ask
GitHub to update their help page to encourage their users to upgrade
to 0.7.0 as soon as possible.

-- 
Shawn.