David Kastrup <d...@gnu.org> writes: > David Kastrup <d...@gnu.org> writes: > >> Graham Percival <gra...@percival-music.ca> writes: >> >>> On Sat, Oct 29, 2011 at 08:43:20AM +0200, David Kastrup wrote: >>>> >>>> After having my local staging up-to-date, >>>> I just pushed staging~1 to origin/dev/staging. >>> >>> thanks, checked and pushed to master. >> >> Sigh. Again it has not been fast-forwarded, but the history has been >> straightened out, dissolving my merge commit. >> >> So much for keeping bijection intact. We need to figure out what >> happens here: I should have thought that the history was perfectly >> fast-forwardable. > > Maybe somebody with shell server access (is there anybody? or a tool?) > can run git reflog on the server so that one can take a look at the > first page or so?
Anyway, for now I propose that Patchy _only_ tries straight pushes of dev/staging to master. No rebase of any kind. If staging is not a proper descendant of master, this can always be fixed manually. Perhaps let Patchy send a mail to the list when master or staging changed since the last attempt, and the push still fails. -- David Kastrup _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
