Re: GIT get corrupted on lustre

[Eric Chamberland <Eric.Chamberland@giref.ulaval.ca>]

Hi!

I still have the corruption problems....

We just compiled a git without threads to try... (by the way, 
--without-pthreads doesn't work, you have to do a --disable-pthreads 
instead).

And to remove the warnings about threads at "git gc" execution, I did a:

git config --local pack.threads 1

and cloned a repository and started to do:

git gc

once every hour.

Then this night (at 05:35:02 exactly), the same error as usual occurred:

error: index file 
.git/objects/pack/pack-bf0748cee64a1964be0a1061c82aca51c993b825.idx is 
too small
error: refs/heads/master does not point to a valid object!

So now I am convinced that it is not a thread problem....

I am kind of discouraged, we like to use git, but in this case we have 
this error which seems unsolvable!

Anyone has a new idea?

Thanks,

Eric


On 01/09/2013 04:20 PM, Eric Chamberland wrote:
> Hi Brian,
>
> On 01/08/2013 11:11 AM, Eric Chamberland wrote:
>> On 12/24/2012 10:11 AM, Brian J. Murrell wrote:
>>> Have you tried adding a "-q" to the git command line to quiet down git's
>>> "feedback" messages?
>>>
>>
>
> I moved to git 1.8.1 and added the "-q" to the command "git gc" but it
> occured to return an error, so the "-q" option is not avoiding the
> problem here... :-/
>
> command in crontab:
>
> cd /rap/jsf-051-aa/ericc/tests_git_clones/GIREF && for i in seq 10; do
> /software/apps/git/1.8.1/bin/git gc -q || true;done
>
> results:
> error: index file
> .git/objects/pack/pack-1f09879c88cd71a15dcc891713cf038d249830ad.idx is
> too small
> error: refs/remotes/origin/BIB_Branche_1_4_x does not point to a valid
> object!
>
> and this clone was a "clean" clone in which only "git qc -q" has been
> run on....
>
> I still have a doubt on threads....
>
> Eric
>
> --
> To unsubscribe from this list: send the line "unsubscribe git" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html