Re: git caching(?) issues with http(s)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, Nov 21, 2014 at 5:10 PM, Ralf Baechle <ralf@xxxxxxxxxxxxxx> wrote:
> On Fri, Nov 21, 2014 at 03:03:08PM +0100, Jonas Gorski wrote:
>
>> >> I noticed that git over http(s) seems to get stale data (just done a
>> >> few minutes ago):
>> >>
>> >> ~$ git clone http://git.linux-mips.org/pub/scm/ralf/upstream-sfr.git
>> >> Cloning into 'upstream-sfr'...
>> >> ~$ cd upstream-sfr/
>> >> ~/upstream-sfr$ git log -1
>> >> commit 475d5928b79bb78326a645863d46ff95c5e25e5a
>> >> Merge: c6b7b9f 1062080
>> >> Author: Ralf Baechle <ralf@xxxxxxxxxxxxxx>
>> >> Date:   Sat Aug 2 00:07:03 2014 +0200
>> >>
>> >>     Merge branch '3.16-fixes' into mips-for-linux-next
>> >
>> > Odd - but I have an idea what might be wrong.
>> >
>> > Are you seeing this only with the upstream-sfr tree?
>>
>> Hadn't checked anything else, but looks like it also affects other trees:
>>
>> ~# git clone http://git.linux-mips.org/pub/scm/ralf/upstream-linus.git
>> Cloning into 'upstream-linus'...
>> ~$ cd upstream-linus/
>> ~/upstream-linus$ git log -1
>> commit 475d5928b79bb78326a645863d46ff95c5e25e5a
>> Merge: c6b7b9f 1062080
>> Author: Ralf Baechle <ralf@xxxxxxxxxxxxxx>
>> Date:   Sat Aug 2 00:07:03 2014 +0200
>>
>>     Merge branch '3.16-fixes' into mips-for-linux-next
>>
>>
>> (strangely the same commit)
>
> The issue is a stale info/refs on the server.  This file normally gets
> updated when the hook hooks/post-update is ran which in turn calls
> git-update-server-info but that just didn't happen.
>
> Can you retry?

I am now able to clone a new, current tree, but updating my older
tree is broken:

# git pull --rebase
error: Unable to find e213b27f3921a88768201638d6d358f0f0c80419 under
https://git.linux-mips.org/pub/scm/ralf/upstream-sfr.git
Cannot obtain needed blob e213b27f3921a88768201638d6d358f0f0c80419
while processing commit fd37780aa0ee04f74e4323f45492947fcef62c35.
error: Fetch failed.

But I can live with that.

> Btw, why are you using http transport anyway?  The git protocol should
> be more efficient.

General paranoia (https) as well as occasional stupid corporate
firewalls that won't allow anything except 80 or 443.


Jonas





[Index of Archives]     [Linux MIPS Home]     [LKML Archive]     [Linux ARM Kernel]     [Linux ARM]     [Linux]     [Git]     [Yosemite News]     [Linux SCSI]     [Linux Hams]

  Powered by Linux