From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Jerkovic Subject: Re: git svn clone just stops Date: Thu, 17 Dec 2009 07:47:38 +1100 Message-ID: <1260996458.6475.11.camel@mark.engr.acx> References: <1260915555.6475.8.camel@mark.engr.acx> <20091216081514.GB26038@dcvr.yhbt.net> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: git@vger.kernel.org To: Eric Wong X-From: git-owner@vger.kernel.org Wed Dec 16 21:48:18 2009 Return-path: Envelope-to: gcvg-git-2@lo.gmane.org Received: from vger.kernel.org ([209.132.180.67]) by lo.gmane.org with esmtp (Exim 4.50) id 1NL0nB-0004uy-MD for gcvg-git-2@lo.gmane.org; Wed, 16 Dec 2009 21:48:18 +0100 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753371AbZLPUr4 (ORCPT ); Wed, 16 Dec 2009 15:47:56 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751487AbZLPUrx (ORCPT ); Wed, 16 Dec 2009 15:47:53 -0500 Received: from mail.aconex.com ([203.89.202.182]:58444 "EHLO postoffice2.aconex.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1753371AbZLPUrq (ORCPT ); Wed, 16 Dec 2009 15:47:46 -0500 X-ASG-Debug-ID: 1260996463-7f4300100000-QuoKaX X-Barracuda-URL: http://postoffice2.aconex.com:8000/cgi-bin/mark.cgi Received: from postoffice.aconex.com (localhost [127.0.0.1]) by postoffice2.aconex.com (Spam & Virus Firewall) with ESMTP id 57270B4319; Thu, 17 Dec 2009 07:47:43 +1100 (EST) Received: from postoffice.aconex.com (postoffice.yarra.acx [192.168.102.1]) by postoffice2.aconex.com with ESMTP id MXUW7SLt6RGCDQ6h; Thu, 17 Dec 2009 07:47:43 +1100 (EST) X-Barracuda-Envelope-From: mjerkovic@aconex.com Received: from gatekeeper.aconex.com (gatekeeper.yarra.acx [192.168.102.10]) by postoffice.aconex.com (Postfix) with ESMTP id 11DFEA501BD; Thu, 17 Dec 2009 07:45:46 +1100 (EST) Received: from localhost (localhost.localdomain [127.0.0.1]) by gatekeeper.aconex.com (Postfix) with ESMTP id 291194892DF; Thu, 17 Dec 2009 07:47:43 +1100 (EST) X-Virus-Scanned: amavisd-new at aconex.com Received: from gatekeeper.aconex.com ([127.0.0.1]) by localhost (gatekeeper.aconex.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id FeyWlR74gtce; Thu, 17 Dec 2009 07:47:38 +1100 (EST) Received: from [192.168.7.137] (melho0.aconex.com [203.89.192.141]) by gatekeeper.aconex.com (Postfix) with ESMTP id 2DD344892D7; Thu, 17 Dec 2009 07:47:38 +1100 (EST) X-ASG-Orig-Subj: Re: git svn clone just stops In-Reply-To: <20091216081514.GB26038@dcvr.yhbt.net> X-Mailer: Evolution 2.26.1 X-Barracuda-Connect: postoffice.yarra.acx[192.168.102.1] X-Barracuda-Start-Time: 1260996463 X-Barracuda-Virus-Scanned: by Aconex Staff Email Spam Firewall at aconex.com Sender: git-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org Archived-At: On Wed, 2009-12-16 at 00:15 -0800, Eric Wong wrote: > Mark Jerkovic wrote: > > Hi all, > > > > I am new to Git and I am attempting to clone a SVN repository > > (containing approximately 60,000 commits) to Git. > > > > Using Git v1.6.0.4 with the following command: > > > > git svn clone https://path/to/svn/repo/trunk project.git > > --authors-file=svnauthors.txt --no-metadata > > > > the process just ends without any messages and the Git repository is > > empty. I have attempted this four times already, and each time it has > > stopped after a different revision, so I can't pin it down to a > > particular commit. > > Hi Mark, > > Any chance you're OOM-ing (extremely large file) or running > out of space somewhere? What exit code do you get when it exits? > > > Has anyone experienced this before? Is there a log anywhere that > > records what is happening during this process? > > Not without error messages of some sort. git svn was designed with poor > network conditions in mind and clone is resumable, so you can just > resume like this: > > cd project.git && git svn fetch > That is an excellent tip. Thanks a lot! > I wouldn't use --no-metadata at this point, though, as it makes it much > harder to debug/fix things or recover from errors. > > Once you're comfortable with it, you should be able to write > "git filter-branch" invocation to remove the metadata lines. I have done some more investigating and it appears that at some point a branch was moved to replace our trunk. When the migration process hits this I it goes back to the beginning and starts migrating the branch history too. I think when it gets to the end of it is not able to resume from the "move" point. I will try your resume tip to see if I can get it started again. Many thanks, Mark. >