git-pull fails after git upgrade

3 messages, 3 authors, 2016-06-15 · open the first message on its own page

git-pull fails after git upgrade

From: Coywolf Qi Hunt <hidden>
Date: 2016-06-15 22:42:16

Hello,

gemini:~/linux/linux-2.6> git --version
git version 1.0.8

gemini:~/linux/linux-2.6> git-pull
fatal: unexpected EOF
Fetch failure: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git

What am I missing? thanks
--
Coywolf Qi Hunt

Linus repository at git.kernel.org???

From: Junio C Hamano <hidden>
Date: 2016-06-15 22:42:16

Coywolf Qi Hunt [off-list ref] writes:
gemini:~/linux/linux-2.6> git-pull
fatal: unexpected EOF
Fetch failure: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
The failure does not seem to happen on all repositories, but
just is with Linus repository, so I doubt "git update" on either
your end or kernel.org public machines have anything to do with
this.  And Linus repository from gitweb is also missing.

I do not have more access to the mirrored public machines than
you have.  I can try snooping the traffic from git-peek-remote
to git://git.kernel.org/ but cannot examine the filesystem over
there, so I cannot tell you more than this: I looked at Linus
repository on the mirror source machine and did not see anything
obviously wrong with it.  It seemed to have updated around 3-4
hours ago.

Re: Linus repository at git.kernel.org???

From: Linus Torvalds <torvalds@osdl.org>
Date: 2016-06-15 22:42:16


On Wed, 11 Jan 2006, Junio C Hamano wrote:
Coywolf Qi Hunt [off-list ref] writes:
quoted
gemini:~/linux/linux-2.6> git-pull
fatal: unexpected EOF
Fetch failure: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
The failure does not seem to happen on all repositories, but
just is with Linus repository, so I doubt "git update" on either
your end or kernel.org public machines have anything to do with
this.  And Linus repository from gitweb is also missing.
The mirroring was delayed for a while, but seems to be ok now.

I repacked my archive (exactly because mirroring was so slow) and while 
the result was always a valid archive on the master site, if you pulled 
when mirroring was starting to delete the unpacked objects but had not yet 
mirrored the new pack, you'd get something like the above - git-daemon 
would exit because of a "missing" object).

I think it should be ok now (at least it seems to have mirrored out things 
over-night, and gitweb now matches my local head again).

		Linus
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help