why can I not download the /jdk repository

Thu Jun 28 17:24:00 UTC 2012

Sorry but your problem description appears quite confusing to me.
The error you are getting now seems different from the one in your first post.

Could you just open a DOS-shell and type:

H:\>hg clone  http://hg.openjdk.java.net/jdk7/jdk7/jdk

(in the same way you did "H:\>hg version")
And make sure you have no Cygwin and/or Python in your path.

This took about 8 minutes on my Windows7 with Tortoise Mercurial 2.0.2.

Cloning just http://hg.openjdk.java.net/jdk7/jdk7 took me about 2 seconds.
You wrote it took you about 40 seconds which seems pretty long for
such a small repository.
Based on that number, cloning jdk would require 160minutes on your
side which seems pretty long and which perhaps leads to strange
(timeout?) problems.

Regards,
Volker

On Thu, Jun 28, 2012 at 6:10 PM, Stadelmann Josef
<josef.stadelmann at axa-winterthur.ch> wrote:
> Thank you Kelly, Volker
>
>
>
> I checked tha path and verified that in fact C:\Program
> Files\TortoiseHG\hg.exe is called. the version I am using is
>
> H:\>hg version
>
> Mercurial Distributed SCM (version 2.1.2)
>
> (see http://mercurial.selenic.com for more information)
>
>
>
> Copyright (C) 2005-2012 Matt Mackall and others
>
> This is free software; see the source for copying conditions. There is NO
>
> warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
>
>
>
> Then using TortoiseHG to download the forest one by one reveled the
> fllowing:
>
>
>
> I take RMB down on my OpenJDK Directory and select TortoiseHG / Clone
>
> I enter the source as http://hg.openjdk.java.net/jdk7/jdk7
>
> I enter the target as  E:\OpenJDK\jdk7
>
>
>
> The clone takes place perfect
>
> Then RMB down on the jdk7 directory which is now a clone and asking at the
> context menu to launch
>
> the workbench. Absolute no problem I can see in the Repository List my
> default set to e:\openjdk
>
> and I can see jdk7 as jdk7
>
>
>
> I select jdk7 in the repository list and RMB down and select from the
> context CLONE then
>
> I enter one after the orther. using the README already brought down I
> download (clone)
>
> each remote sub-repository. Simpel and easy
>
>
>
> I am asking for uncompressed data and for pull protocol for the meta data
> and I can see in the details window how all the sources and resources come
> down the line.
>
>
>
> This works for each sub repository "corba langtools hotspot jaxws jaxp" it
> does just not work for "jdk"
>
>
>
> I don’t think the problem is on my system but it has to do with the workload
> of the jdk sub repository.
>
>
>
> excerpt from the clone process of the Tortoise workbench (latest greatest
> version). And as you can see
>
> the previous command completed successful but the clone for jdk always fails
> with URLerror or on a
>
> cmd line version it fails with error 255.
>
>
>
> the jaxp comes down as
>
>
>
> % hg clone --verbose -- http://hg.openjdk.java.net/jdk7/jdk7/jaxp
> E:\OpenJDK\jdk7\jaxp
>
> requesting all changes
>
> adding changesets         <----- about 40 seconds then it continues to
> adding manifest;  but for jdk you can wait forever here . . .
>
> adding manifests
>
> adding file changes
>
> added 233 changesets with 2213 changes to 1984 files
>
> updating to branch default
>
> resolving manifests
>
> getting .hgignore
>
> . . .
>
> . . .
>
> getting nbproject/project.xml
>
> getting nbproject/sqe.properties
>
> getting patches/jaxp_src/README
>
> 19 files updated, 0 files merged, 0 files removed, 0 files unresolved
>
> [command completed successfully Thu Jun 28 17:49:42
> 2012]                                                  ß- for japx
>
> % hg clone --verbose -- http://hg.openjdk.java.net/jdk7/jdk7/jdk
> E:\OpenJDK\jdk7\jdk               ß- for jdk
>
> requesting all changes
>
> URLError:
>
> [command returned code 255 Thu Jun 28 18:02:01 2012]
>
>
>
>
>
> Guess the problem is NOT on my side of the pond,
>
>
>
> or what explains why 1 master and 5 sub-repositories clone perfect and only
> one "jdk" fails?
>
>
>
> Josef
>
>
>
>
>
>
>
>
>
> Von: Kelly O'Hair [mailto:kelly.ohair at oracle.com]
> Gesendet: Mittwoch, 27. Juni 2012 22:03
> An: Volker Simonis
> Cc: Stadelmann Josef; build-dev at openjdk.java.net
>
>
> Betreff: Re: why can I not download the /jdk repository
>
>
>
> I found this old documentation when I worked on the JavaFX project, maybe
> this helps:
>
>
>
> TortoiseHG (Windows)
>
> Get the TortoiseHG Download bundle and install it. It install into the
> directory:
>
> "C:/Program Files/TortoiseHG/".
>
> After installing it:
>
> 1.      Edit the file "C:/Program Files/TortoiseHG/Mercurial.ini" and make
> sure the ssh command used is from CYGWIN and not Plink. Unless of course you
> want to use Plink, it's up to you.
>
> 2.      Prepend "C:/Program Files/TortoiseHG/" to your PATH, make sure it is
> before /usr/bin, so that running hg version tells you it is TortoiseHG.
>
> Mercurial Tips
>
> ·         On Windows, the default CYGWIN hg will not work on some of the
> repositories, so you can downgrade the CYGWIN hg version to 1.0.2 or switch
> to use the TortoiseHG build of Mercurial, which is a 1.3.1 or newer version.
> The problem centers around the Windows limits to full pathnames. Mercurial
> .hg/ files can end up with much longer filenames than the file they
> represent in the working set, or the files you edit. The most recent
> releases of Mercurial have optimized these path lengths, but versions like
> 1.1 had made the problem worse, resulting in some of our repositories not
> cloning. Version 1.0.2 was ok, but 1.3.1 or newer is best. Unfortunately, at
> this time, 1.3.1 is not available with CYGWIN.
>
> ·         On Windows, the CYGWIN hg is a Python script and doesn't play well
> with the native Windows system. Some ant scripts try and run hg and ant will
> fail when running a Python script. Using TortoiseHG solves this problem
> because it provides a hg.exe.
>
> -kto
>
>
>
> On Jun 27, 2012, at 10:32 AM, Volker Simonis wrote:
>
>
>
> Hi,
>
> this may be related to Cygwin. The error "C770817 at C036357
> /cygdrive/e/OpenJDK/jdk8" indicates that you are using a
> "Cygwin-Mercurial". I would recommend to install and use a native
> Windows Mercurial (e.g.
> http://tortoisehg.bitbucket.org/download/index.html) and try with that
> one.
>
> I never had problems cloning with tortoisehg, also I didn't succeed to
> push with it (I think because of some ssh/private-key issues). So I
> use the Cygwin hg for pushing, but that's not very stable for me
> either...
>
> Regards,
> Volker
>
> On Wed, Jun 27, 2012 at 6:59 PM, Stadelmann Josef
> <josef.stadelmann at axa-winterthur.ch> wrote:
>
> Just o inform you before I give up.
>
>
>
> At the README of jdk7 or jdk8 one can read
>
>
>
> ---------------
>
>
>
> This one root repository can be obtained with something like:
>
>
>
>     hg clone http://hg.openjdk.java.net/jdk8/jdk8 openjdk8
>
>
>
>
>
>
>
>   To make sure you have all the nested repositories, you can run the
>
>
>
>   get_source.sh script located in the same respository as this file:
>
>
>
>     cd openjdk8 && sh ./get_source.sh
>
>
>
> ---------------
>
>
>
> Since weeks, running the get_source.sh, download takes place for all
>
>
>
> sub-repositories except and always fails for the jdk*/jdk repository.
>
>
>
> "hg clone http://hg.openjdk.java.net/jdk7/jdk7/ C:/OpenJDK/jdk7/jdk"
>
>
>
> Watching the network, there is an immediate transfer of the .hg subdirectory
>
>
>
> but then after about 20 minutes I get an abort error
>
>
>
> C770817 at C036357 /cygdrive/e/OpenJDK/jdk8
>
>
>
> $ hg clone http://hg.openjdk.java.net/jdk8/jdk8/jdk jdk
>
>
>
> requesting all changes
>
>
>
> abort: error:
>
>
>
> WHY?
>
>
>
> Once in the past 4 weeks, I was able to download it on a Saturday at home.
>
>
>
> Josef
>
>