From: Denys Dmytriyenko <denis@denix.org>
To: Christian Gagneraud <chgans@gna.org>
Cc: "meta-arago@arago-project.org" <meta-arago@arago-project.org>,
yocto@yoctoproject.org
Subject: Re: [meta-arago] GCC SVN repo too slow?
Date: Mon, 19 Aug 2013 11:53:07 -0400 [thread overview]
Message-ID: <20130819155307.GA32524@denix.org> (raw)
In-Reply-To: <5211B27F.9070707@gna.org>
On Mon, Aug 19, 2013 at 05:51:59PM +1200, Christian Gagneraud wrote:
> CC: meta-arago@arago-project.org
>
>
> On 19/08/13 17:16, Khem Raj wrote:
> >
> >On Aug 18, 2013, at 7:34 PM, Christian Gagneraud <chgans@gna.org> wrote:
> >
> >>Hi all,
> >>
> >>I'm having problems with the GCC SVN repo being so slow to checkout, that svn failed (and so my build too).
> >>I first suspected the proxy we have here, but i tried a checkout from an outside machine (but still in New Zealand) and it seems the problem comes from gcc.gnu.org.
> >>
> >>I am using the arago layer, which requires svn://gcc.gnu.org/svn/gcc/branches;module=gcc-4_5-branch;protocol=http for gcc-crosssdk-intermediate.
> >>
> >>By slow, I really mean slow: 1 to 2 files are checkout per second!?!
> >>
> >>Has anyone encountered the same kind of problems?
> >
> >
> >thats a known issue and one of pressing reasons to not use svn version on later versions if you see new recipes
> >best case you could hit a mirror from yocto project and download the source mirror
>
>
> From the logs, I can see that bitbake is trying (among others):
> downloads.yoctoproject.org
> sources.openembedded.org
>
> but hits a 404, so it fall-backs to GCC's SVN repo directly.
>
> On the side, from the bitbake logs, I end up with a rather obscure error:
> svn: E175002: REPORT of '/svn/gcc/!svn/vcc/default': 200 OK
> (http://gcc.gnu.org)
>
> When trying a checkout manually from the outside machine, it fails
> too, but the error message is a bit more friendly:
> svn: REPORT of '/svn/gcc/!svn/vcc/default': Could not read response
> body: connection was closed by server (http://gcc.gnu.org)
>
> I found plenty of "black magic" workarounds on the web while looking
> for the strange "Error: [...] 200 OK", among them was to increase
> Apache time out! Which won't even really solve my problem, because
> the checkout will takes hours, if not days!
>
> Right now, I'm really stuck. The only solutions I can think of are:
> - Ask kindly Arago not to use GCC code from SVN, or maybe to upload
> an archive on yocto/oe mirrors.
> - Check with GNU/GCC guys if they can do something about it (maybe
> it's a temporary issue with their server)
> - Wait and see, cross my fingers. :(
Chris,
I'm just wondering why do you even need gcc-4.5? Unless you specifically use
gcc-4.5 based toolchain with meta-arago. That is the only reason I keep
gcc-4.5 recipes in meta-arago to support old products with old toolchains.
We've moved to 4.7 Linaro quite some time ago - when you use that, you won't
need to download 4.5 version.
If you really need gcc-4.5 and it's not just a misconfiguration, then I have
a tarball available that can be uploaded to a mirror...
--
Denys
WARNING: multiple messages have this Message-ID (diff)
From: Denys Dmytriyenko <denis@denix.org>
To: Christian Gagneraud <chgans@gna.org>
Cc: "meta-arago@arago-project.org" <meta-arago@arago-project.org>,
yocto@yoctoproject.org, Khem Raj <raj.khem@gmail.com>
Subject: Re: [yocto] GCC SVN repo too slow?
Date: Mon, 19 Aug 2013 11:53:07 -0400 [thread overview]
Message-ID: <20130819155307.GA32524@denix.org> (raw)
In-Reply-To: <5211B27F.9070707@gna.org>
On Mon, Aug 19, 2013 at 05:51:59PM +1200, Christian Gagneraud wrote:
> CC: meta-arago@arago-project.org
>
>
> On 19/08/13 17:16, Khem Raj wrote:
> >
> >On Aug 18, 2013, at 7:34 PM, Christian Gagneraud <chgans@gna.org> wrote:
> >
> >>Hi all,
> >>
> >>I'm having problems with the GCC SVN repo being so slow to checkout, that svn failed (and so my build too).
> >>I first suspected the proxy we have here, but i tried a checkout from an outside machine (but still in New Zealand) and it seems the problem comes from gcc.gnu.org.
> >>
> >>I am using the arago layer, which requires svn://gcc.gnu.org/svn/gcc/branches;module=gcc-4_5-branch;protocol=http for gcc-crosssdk-intermediate.
> >>
> >>By slow, I really mean slow: 1 to 2 files are checkout per second!?!
> >>
> >>Has anyone encountered the same kind of problems?
> >
> >
> >thats a known issue and one of pressing reasons to not use svn version on later versions if you see new recipes
> >best case you could hit a mirror from yocto project and download the source mirror
>
>
> From the logs, I can see that bitbake is trying (among others):
> downloads.yoctoproject.org
> sources.openembedded.org
>
> but hits a 404, so it fall-backs to GCC's SVN repo directly.
>
> On the side, from the bitbake logs, I end up with a rather obscure error:
> svn: E175002: REPORT of '/svn/gcc/!svn/vcc/default': 200 OK
> (http://gcc.gnu.org)
>
> When trying a checkout manually from the outside machine, it fails
> too, but the error message is a bit more friendly:
> svn: REPORT of '/svn/gcc/!svn/vcc/default': Could not read response
> body: connection was closed by server (http://gcc.gnu.org)
>
> I found plenty of "black magic" workarounds on the web while looking
> for the strange "Error: [...] 200 OK", among them was to increase
> Apache time out! Which won't even really solve my problem, because
> the checkout will takes hours, if not days!
>
> Right now, I'm really stuck. The only solutions I can think of are:
> - Ask kindly Arago not to use GCC code from SVN, or maybe to upload
> an archive on yocto/oe mirrors.
> - Check with GNU/GCC guys if they can do something about it (maybe
> it's a temporary issue with their server)
> - Wait and see, cross my fingers. :(
Chris,
I'm just wondering why do you even need gcc-4.5? Unless you specifically use
gcc-4.5 based toolchain with meta-arago. That is the only reason I keep
gcc-4.5 recipes in meta-arago to support old products with old toolchains.
We've moved to 4.7 Linaro quite some time ago - when you use that, you won't
need to download 4.5 version.
If you really need gcc-4.5 and it's not just a misconfiguration, then I have
a tarball available that can be uploaded to a mirror...
--
Denys
next prev parent reply other threads:[~2013-08-19 15:53 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-08-19 2:34 GCC SVN repo too slow? Christian Gagneraud
2013-08-19 5:16 ` Khem Raj
2013-08-19 5:51 ` Christian Gagneraud
2013-08-19 5:51 ` [yocto] " Christian Gagneraud
2013-08-19 5:56 ` Khem Raj
2013-08-19 5:56 ` [yocto] " Khem Raj
2013-08-19 15:05 ` [meta-arago] " Maupin, Chase
2013-08-19 15:05 ` [yocto] " Maupin, Chase
2013-08-19 15:53 ` Denys Dmytriyenko [this message]
2013-08-19 15:53 ` Denys Dmytriyenko
2013-08-19 16:01 ` Khem Raj
2013-08-19 16:01 ` [yocto] " Khem Raj
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20130819155307.GA32524@denix.org \
--to=denis@denix.org \
--cc=chgans@gna.org \
--cc=meta-arago@arago-project.org \
--cc=yocto@yoctoproject.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.