måndag den 9 maj 2011 klockan 17:40 skrev Jonathan Nieder detta:
> Mats Erik Andersson wrote:
>
> > In comparison,
> >
> > $ git clone git://git.sv.gnu.org/gnulib.git
> > [...]
> > Resolving deltas: 51% (48907/95596)
> >
> > and the uses much time for internal computations.
>
> On second thought, are you saying this hangs, too?
No, only that it consumes very much time to complete.
A noticeable amount on kfreebsd-amd64 and a disturbing
amount on my weaker i386 system, so it is easy to be
fooled when running the test.
> > Exactly the same behaviour is present on kfreebsd-amd64
> > with identical Git package. As long as this latter system
> > had 1:1.7.2.3-2 shallow cloning worked correctly.
> [...]
> > Contrast this to my observations that git_1:1.7.2.5-1 works
> > correctly for Squeeze with i386 and amd64.
>
> That part is puzzling, too --- the relevant variable would be threads
> vs processes (i.e. pre-1.7.2 vs post-1.7.2) if I had diagnosed it
> correctly. So there is more to figure out.
>
> Anyway, let's see how 1.7.5.1 does, and then go from there.
The result is very disappointing. This is the complete test run,
using manual commanding. The version 1.7.5.1 is available for i386,
but not for kfreebsd-amd64, so only the former can be investigated.
* Success: git clone --depth 2 git://git.sv.gnu.org/gnulib.git
* Failure: git clone --depth 1 ..../gnulib.git
Aborted
* Failure: git clone --depth 3 ..../gnulib.git
Aborted
* Failure: git clone --depth 2 ..../gnulib.git
Aborted
* (Reference check on much smaller repository)
Success: git clone --depth 3 ..../inetutils.git
* Failure: git clone --depth 1 ..../gnulib.git
Aborted
* Wait ten minutes
* Failure: git clone --depth 1 ..../gnulib.git
Aborted
* Wait twenty minutes
* Failure: git clone --depth 1 ..../gnulib.git
Aborted
* Wait thirty minutes
* Success: git clone ..../gnulib.git
(Execution time: five minutes plus, but does make in the end.)
It is puzzling to have the initial success with depth=2, followed
by all failures in shallow mode. The success in deep mode at least
helps in reducing the scope somewhat, but the conclusion escapes me.
Regards,
Mats E A
(For the protocol: Is the strange, combined dependency on 'git-man'
intended as stated, one '>>' clause and one '<<'? Correct versioning?
No response needed, but it provokes manual intervention at upgrade.)
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
