Lucas Bonnet <[email protected]> wrote: >> Okay, I've done some testing with a remote repository. "git >> reset --hard $UUID && git push" will fail with:
>> | ! [rejected] master -> master (non-fast forward) >> but the already mentioned: >> | git push origin +a4d7a62c0fc38ef5ce7b1b68f493b552d835214f:master >> will work just fine as far as I can tell (you could probably >> combine both, i. e. "git reset --hard $UUID && git push >> origin +:master", but I find the explicit naming of the com- >> mit to reset to much more affirmative :-)). >> If no one objects, I'd like to go ahead with this tomorrow >> (Friday) afternoon UTC. (I have made a backup copy of the >> repository :-).) > You are clear to "git push origin", I repeat, your are clear to git > push. > I didn't think you'd take my advice about breaking things this literally :) > Looks like it's time for me to dive into git from a developer's > perspective. Believe me, I would have preferred a different entrée :-). Actually, the resetting today was much scarier than I thought as Savannah /denies/ non-fast forward pushes. So af- ter careful consideration (and one more backup), I followed the path of: | git reset --hard a4d7a62c0fc38ef5ce7b1b68f493b552d835214f | git push origin :master | git push origin master A clone afterwards looks "alright". So please everyone make sure that you base your commits on this one :-). Thanks, Tim _______________________________________________ Emms-patches mailing list [email protected] http://lists.gnu.org/mailman/listinfo/emms-patches
