fetch problem or was master on kernel.org reset?

fetch problem or was master on kernel.org reset?

[Alex Riesen]

Because I just had to force-update the local reference:

    .../git.git$ git --bare log -g origin/master

    commit 2ebc02d32a4360da2cf69c2b5f5bfad0716d42b0
    Reflog: refs/remotes/origin/master@{1} (Alex Riesen <raa.lkml@gmail.com>)
    Reflog message: fetch git: fast forward
    Author: Junio C Hamano <gitster@pobox.com>
    Date:   Sun Aug 17 15:48:51 2008 -0700

	Start 1.6.1 cycle

	Signed-off-by: Junio C Hamano <gitster@pobox.com>

    commit ea02eef096d4bfcbb83e76cfab0fcb42dbcad35e
    Reflog: refs/remotes/origin/master@{2} (Alex Riesen <raa.lkml@gmail.com>)
    Reflog message: fetch origin: fast forward
    Author: Junio C Hamano <gitster@pobox.com>
    Date:   Sun Aug 17 11:42:10 2008 -0700

	GIT 1.6.0

	Signed-off-by: Junio C Hamano <gitster@pobox.com>


The commit ea02eef096d4bfcbb83e76cfab0fcb42dbcad35e is the current
value of master (on kernel.org). But it definitely was
2ebc02d32a4360da2cf69c2b5f5bfad0716d42b0 some time before (at 11:42)!

Re: fetch problem or was master on kernel.org reset?

[Junio C Hamano]

Alex Riesen <raa.lkml@gmail.com> writes:

> The commit ea02eef096d4bfcbb83e76cfab0fcb42dbcad35e is the current
> value of master (on kernel.org). But it definitely was
> 2ebc02d32a4360da2cf69c2b5f5bfad0716d42b0 some time before (at 11:42)!

I suspect that one of the mirrors still had ea02ee (1.6.0) when you made
the above observation, after 2ebc02 (Start 1.6.1) was pushed into.
According to my push log, "1.6.0" was pushed on 14:04 and "Start 1.6.1" on
19:47 (both on 2008-08-17, US/Pacific).

Currently all git://git[1-4].kernel.org/pub/scm/git/git.git/ mirrors seem
to have the correct 2ebc02.(Start 1.6.1).

Re: fetch problem or was master on kernel.org reset?

[Alex Riesen]

Junio C Hamano, Mon, Aug 18, 2008 20:25:26 +0200:
> Alex Riesen <raa.lkml@gmail.com> writes:
> 
> > The commit ea02eef096d4bfcbb83e76cfab0fcb42dbcad35e is the current
> > value of master (on kernel.org). But it definitely was
> > 2ebc02d32a4360da2cf69c2b5f5bfad0716d42b0 some time before (at 11:42)!
> 
> I suspect that one of the mirrors still had ea02ee (1.6.0) when you made
> the above observation, after 2ebc02 (Start 1.6.1) was pushed into.
> According to my push log, "1.6.0" was pushed on 14:04 and "Start 1.6.1" on
> 19:47 (both on 2008-08-17, US/Pacific).

Ach... Yes. The mirror in question is my own. I updated the repository
(which now refuses to update its local references) _arrround_ the
mirror (directly from kernel.org). Now the mirror is older then my
repository. Simple.

Evening well spent: I just learned that I am far from perfect an have
a lot room for improvement 8)