Problem in update from the linux-2.6 repository

Problem in update from the linux-2.6 repository

[Marcel Holtmann]

Hi guys,

I run into a weird problem when trying to update my clone of the
linux-2.6 repository from Linus:

---
# cg-update -f
17:09:00 URL:http://www.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/HEAD [41/41] -> "refs/heads/.origin-fetching" [1]
Getting alternates list
Getting pack list
error: Unable to find 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c under http://www.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/

Cannot obtain needed object 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
while processing commit 0000000000000000000000000000000000000000.
cg-fetch: objects fetch failed
---

I use the latest available git and cogito for this and I have no idea
which is at fault. Using rsync as transport it even looks more worse:

---
# cg-update -f
WARNING: The rsync access method is DEPRECATED and will be REMOVED in the future!

receiving file list ... done

sent 111 bytes  received 774 bytes  354.00 bytes/sec
total size is 41  speedup is 0.05

receiving file list ... done
progress: 2631 objects, 9754798 bytes, 100% done
info/packs

sent 58020 bytes  received 10252045 bytes  74982.29 bytes/sec
total size is 121158464  speedup is 11.75

receiving file list ... done

sent 118 bytes  received 1095 bytes  346.57 bytes/sec
total size is 902  speedup is 0.74
Tree change: f912696ab330bf539231d1f8032320f2a08b850f:2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
error: unable to find 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
fatal: git-cat-file 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c: bad file
error: unable to find 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
fatal: git-cat-file 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c: bad file
Invalid tree id: 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c

Applying changes...
error: unable to find 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
fatal: git-cat-file 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c: bad file
error: unable to find 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
fatal: git-cat-file 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c: bad file
Invalid commit id: 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
---

Maybe this is a problem of the kernel.org synchronization, but otherwise
it should be fixed. Will there be a way to get some useful messages to
the end user in case of an error or some sanity checks that the tree you
are pulling from is in sane state or not?

Regards

Marcel

Re: Problem in update from the linux-2.6 repository

[Marcel Holtmann]

Hi,

> I run into a weird problem when trying to update my clone of the
> linux-2.6 repository from Linus:
> 
> ---
> # cg-update -f
> 17:09:00 URL:http://www.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/HEAD [41/41] -> "refs/heads/.origin-fetching" [1]
> Getting alternates list
> Getting pack list
> error: Unable to find 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c under http://www.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/
> 
> Cannot obtain needed object 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
> while processing commit 0000000000000000000000000000000000000000.
> cg-fetch: objects fetch failed
> ---

forget about this thing. The reason for this problem was the kernel.org
mirroring and the repository wasn't in a sane state. However it would be
great if we can detect this somehow.

Regards

Marcel

Re: Problem in update from the linux-2.6 repository

[Petr Baudis]

  Hi,

Dear diary, on Mon, Nov 07, 2005 at 11:18:51PM CET, I got a letter
where Marcel Holtmann <marcel@holtmann.org> told me that...
> > I run into a weird problem when trying to update my clone of the
> > linux-2.6 repository from Linus:
> > 
> > ---
> > # cg-update -f
> > 17:09:00 URL:http://www.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/HEAD [41/41] -> "refs/heads/.origin-fetching" [1]
> > Getting alternates list
> > Getting pack list
> > error: Unable to find 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c under http://www.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/
> > 
> > Cannot obtain needed object 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
> > while processing commit 0000000000000000000000000000000000000000.
> > cg-fetch: objects fetch failed
> > ---
> 
> forget about this thing. The reason for this problem was the kernel.org
> mirroring and the repository wasn't in a sane state. However it would be
> great if we can detect this somehow.

  well, we did detect this and error out. ;-) What else would you like
GIT/Cogito to do? Anything particular that could make this less
confusing?

-- 
				Petr "Pasky" Baudis
Stuff: http://pasky.or.cz/
VI has two modes: the one in which it beeps and the one in which
it doesn't.

Re: Problem in update from the linux-2.6 repository

[Marcel Holtmann]

Hi Petr,

> > > I run into a weird problem when trying to update my clone of the
> > > linux-2.6 repository from Linus:
> > > 
> > > ---
> > > # cg-update -f
> > > 17:09:00 URL:http://www.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/HEAD [41/41] -> "refs/heads/.origin-fetching" [1]
> > > Getting alternates list
> > > Getting pack list
> > > error: Unable to find 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c under http://www.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git/
> > > 
> > > Cannot obtain needed object 2ed5e6d09e266bd2288d49aaaf240ed8c468c13c
> > > while processing commit 0000000000000000000000000000000000000000.
> > > cg-fetch: objects fetch failed
> > > ---
> > 
> > forget about this thing. The reason for this problem was the kernel.org
> > mirroring and the repository wasn't in a sane state. However it would be
> > great if we can detect this somehow.
> 
>   well, we did detect this and error out. ;-) What else would you like
> GIT/Cogito to do? Anything particular that could make this less
> confusing?

I actually have no real idea, but the "object fetch failed" doesn't tell
me anything besides that an object seems to be missing. Maybe this is a
problem we have to live with when using the kernel.org mirroring.

Regards

Marcel

Re: Problem in update from the linux-2.6 repository

[Nick Hengeveld]

On Mon, Nov 07, 2005 at 11:18:51PM +0100, Marcel Holtmann wrote:

> forget about this thing. The reason for this problem was the kernel.org
> mirroring and the repository wasn't in a sane state. However it would be
> great if we can detect this somehow.

Is this an issue with the two servers (or virtual servers?) behind the
kernel.org DNS name being out of sync?

Would it make sense to configure the individual servers to be http-alternates
of each other?

-- 
For a successful technology, reality must take precedence over public
relations, for nature cannot be fooled.