You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by kf...@collab.net on 2005/10/13 03:55:02 UTC
1.3.0 RC1 ready to be rolled
The magic pathrev is:
http://svn.collab.net/repos/svn/branches/1.3.x/ @ r16689
Ben Reser, is today good for you to roll? I know we originally said
Wednesday, but then we got stuck on those Windows bindings problems
for a bit, which pushed us into today (Thursday).
Thanks to quite a few people in IRC, for that final push of change
reviewing and voting. I won't mention everyone's name here, but You
Know Who You Are... and if You don't Know Who You Are, well, that's
what we have 'svn log' for.
-Karl
--
www.collab.net <> CollabNet | Distributed Development On Demand
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by John Szakmeister <jo...@szakmeister.net>.
On Thursday 13 October 2005 03:13, John Szakmeister wrote:
> On Wednesday 12 October 2005 23:55, kfogel@collab.net wrote:
> > The magic pathrev is:
> >
> > http://svn.collab.net/repos/svn/branches/1.3.x/ @ r16689
> >
> > Ben Reser, is today good for you to roll? I know we originally said
> > Wednesday, but then we got stuck on those Windows bindings problems
> > for a bit, which pushed us into today (Thursday).
> >
> > Thanks to quite a few people in IRC, for that final push of change
> > reviewing and voting. I won't mention everyone's name here, but You
> > Know Who You Are... and if You don't Know Who You Are, well, that's
> > what we have 'svn log' for.
>
> Is this okay considering that the Ruby bindings are current failing on
> Jani's machine?
s/current/currently/
-John
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by John Szakmeister <jo...@szakmeister.net>.
On Wednesday 12 October 2005 23:55, kfogel@collab.net wrote:
> The magic pathrev is:
>
> http://svn.collab.net/repos/svn/branches/1.3.x/ @ r16689
>
> Ben Reser, is today good for you to roll? I know we originally said
> Wednesday, but then we got stuck on those Windows bindings problems
> for a bit, which pushed us into today (Thursday).
>
> Thanks to quite a few people in IRC, for that final push of change
> reviewing and voting. I won't mention everyone's name here, but You
> Know Who You Are... and if You don't Know Who You Are, well, that's
> what we have 'svn log' for.
Is this okay considering that the Ruby bindings are current failing on Jani's
machine?
-John
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by John Peacock <jp...@rowman.com>.
Peter N. Lundblad wrote:
> I can't reproduce this using latest 1.3.x against svn.collab.net using
> neon 0.24.7. Are you sure this isn't a local config problem?
I wonder if that is it then. I used neon 0.25.0 (since it was specifically
mentioned as supported in the INSTALL docs). I also retrieved apr and apr-util
from branches/0.9.x (also per INSTALL) to replace the oddly configured libraries
that SUSE comes with.
I'll wash/rinse/repeat with 0.24.7 and let you know as soon as I can.
John
--
John Peacock
Director of Information Research and Technology
Rowman & Littlefield Publishing Group
4720 Boston Way
Lanham, MD 20706
301-459-3366 x.5010
fax 301-429-5747
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by "Peter N. Lundblad" <pe...@famlundblad.se>.
On Thu, 13 Oct 2005, John Peacock wrote:
> kfogel@collab.net wrote:
> > The magic pathrev is:
> >
> > http://svn.collab.net/repos/svn/branches/1.3.x/ @ r16689
> >
>
> I don't suppose you noticed the problem that I reported yesterday with a 1.3.x
> client vs. a 1.2.1 server, did you? I snagged the above path (well, actually at
> r16683) and built/tested/installed. Now anytime I try and connect to, say the
> master svn repository, I get the following error:
>
> > $ svn ls http://svn.collab.net/repos/svn/
> > subversion/libsvn_ra_dav/util.c:826: (apr_err=175002)
> > svn: PROPFIND request failed on '/repos/svn'
> > subversion/libsvn_ra_dav/util.c:296: (apr_err=175002)
> > svn: PROPFIND of '/repos/svn': 207 Multi-Status (http://svn.collab.net)
>
> I think the compatibility guarantee says a 1.3.x client should be able to talk
> to a 1.2.x server, right? :-0
>
I can't reproduce this using latest 1.3.x against svn.collab.net using
neon 0.24.7. Are you sure this isn't a local config problem?
Regards,
//Peter
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by John Peacock <jp...@rowman.com>.
Philip Martin wrote:
>
> I have a 1.3.x@HEAD build that uses neon 0.25.0 and it seems to work
>
I'm building with the following:
> gcc (GCC) 4.0.2 20050901 (prerelease) (SUSE Linux)
> Subversion branches/1.3.x @ 16706
> Neon 0.25.0
> apr/branches/0.9.x @314989
> apr-util/branches/0.9.x @314989
and the following configure options:
> "./configure" \
> "--enable-maintainer-mode" \
> "--without-berkeley-db" \
> "--with-ssl" \
> "--disable-shared" \
and I get the same error as before. The ethereal trace (attached) doesn't
provide much more detail that I can recognize. Let me know if there is anything
else I can provide (up to an including a login on the box in question, if so
required).
John
--
John Peacock
Director of Information Research and Technology
Rowman & Littlefield Publishing Group
4720 Boston Way
Lanham, MD 20706
301-459-3366 x.5010
fax 301-429-5747
Re: 1.3.0 RC1 ready to be rolled
Posted by Philip Martin <ph...@codematters.co.uk>.
kfogel@collab.net writes:
> John Peacock <jp...@rowman.com> writes:
>> After testing, it seems it _was_ Neon 0.25.0 that broke things. I
>> don't have time to get an ethereal trace right now, but I'll try to
>> rebuild and get a trace tonight or tomorrow.
>
> Oooooh. Okay, thanks for narrowing it down. That's a 1.3.0
> showstopper IMHO, at least until we know more. I.e., if it turns out
> to be a bug in Neon 0.25.x, then we might as well ship, because no
> better Neon is available. Users can just choose 0.24 or 0.25.
>
> However, my instinct is it's a bug in SVN, since it shows itself when
> client and server have relative version skew.
I have a 1.3.x@HEAD build that uses neon 0.25.0 and it seems to work
$ /usr/local/subversion2/bin/svn --version -q
1.3.0-dev
$ ldd /usr/local/subversion2/bin/svn | grep neon
libneon.so.25 => /usr/lib/libneon.so.25 (0x402c8000)
$ /usr/local/subversion2/bin/svn ls http://svn.collab.net/repos/svn/
README
branches/
tags/
trunk/
--
Philip Martin
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by kf...@collab.net.
John Peacock <jp...@rowman.com> writes:
> After testing, it seems it _was_ Neon 0.25.0 that broke things. I
> don't have time to get an ethereal trace right now, but I'll try to
> rebuild and get a trace tonight or tomorrow.
Oooooh. Okay, thanks for narrowing it down. That's a 1.3.0
showstopper IMHO, at least until we know more. I.e., if it turns out
to be a bug in Neon 0.25.x, then we might as well ship, because no
better Neon is available. Users can just choose 0.24 or 0.25.
However, my instinct is it's a bug in SVN, since it shows itself when
client and server have relative version skew.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by John Peacock <jp...@rowman.com>.
Ben Collins-Sussman wrote:
> Sure, most of us build svn trunk everyday and continue to work against
> svn.collab.net...
I was kidding; I knew people were using trunk. I was just trying to alert the
list that there was a problem somewhere. ;-)
> Can you get a network trace with ethereal?
After testing, it seems it _was_ Neon 0.25.0 that broke things. I don't have
time to get an ethereal trace right now, but I'll try to rebuild and get a trace
tonight or tomorrow.
John
--
John Peacock
Director of Information Research and Technology
Rowman & Littlefield Publishing Group
4720 Boston Way
Lanham, MD 20706
301-459-3366 x.5010
fax 301-429-5747
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by Ben Collins-Sussman <su...@red-bean.com>.
On 10/13/05, John Peacock <jp...@rowman.com> wrote:
> kfogel@collab.net wrote:
> > The magic pathrev is:
> >
> > http://svn.collab.net/repos/svn/branches/1.3.x/ @ r16689
> >
>
> I don't suppose you noticed the problem that I reported yesterday with a 1.3.x
> client vs. a 1.2.1 server, did you? I snagged the above path (well, actually at
> r16683) and built/tested/installed. Now anytime I try and connect to, say the
> master svn repository, I get the following error:
>
> > $ svn ls http://svn.collab.net/repos/svn/
> > subversion/libsvn_ra_dav/util.c:826: (apr_err=175002)
> > svn: PROPFIND request failed on '/repos/svn'
> > subversion/libsvn_ra_dav/util.c:296: (apr_err=175002)
> > svn: PROPFIND of '/repos/svn': 207 Multi-Status (http://svn.collab.net)
>
> I think the compatibility guarantee says a 1.3.x client should be able to talk
> to a 1.2.x server, right? :-0
Sure, most of us build svn trunk everyday and continue to work against
svn.collab.net...
Can you get a network trace with ethereal?
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by John Peacock <jp...@rowman.com>.
kfogel@collab.net wrote:
> The magic pathrev is:
>
> http://svn.collab.net/repos/svn/branches/1.3.x/ @ r16689
>
I don't suppose you noticed the problem that I reported yesterday with a 1.3.x
client vs. a 1.2.1 server, did you? I snagged the above path (well, actually at
r16683) and built/tested/installed. Now anytime I try and connect to, say the
master svn repository, I get the following error:
> $ svn ls http://svn.collab.net/repos/svn/
> subversion/libsvn_ra_dav/util.c:826: (apr_err=175002)
> svn: PROPFIND request failed on '/repos/svn'
> subversion/libsvn_ra_dav/util.c:296: (apr_err=175002)
> svn: PROPFIND of '/repos/svn': 207 Multi-Status (http://svn.collab.net)
I think the compatibility guarantee says a 1.3.x client should be able to talk
to a 1.2.x server, right? :-0
I used another machine to get the diff between r16683 and r16689 to make sure
that it wasn't something weird while you were merging changes in (but looking at
the diff, I don't see anything likely). FWIW, I set up and ran davcheck and
everything worked 100%.
John
--
John Peacock
Director of Information Research and Technology
Rowman & Littlefield Publishing Group
4720 Boston Way
Lanham, MD 20706
301-459-3366 x.5010
fax 301-429-5747
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by David James <ja...@gmail.com>.
On 10/13/05, Jani Averbach <ja...@jaa.iki.fi> wrote:
> On 2005-10-13 12:57-0400, David James wrote:
> > I'm not able to reproduce the test failures on my machine now. "make
> > check-swig-rb" passes for me on both the 1.3.x branch and on trunk
> > using r16666.
>
> Just make sure, did you compile your apr with "--enable-pool-debug=yes"?
After recompiling Subversion and APR with --enable-pool-debug=yes, I
can reproduce the problem. The test_session test from test_ra.rb fails
with a segfault, as shown below:
test_session(SvnRaTest):
/nobackup/clgrp/james/trunk/subversion/bindings/swig/ruby/test/test_ra.rb:108:
[BUG] Segmentation fault
ruby 1.8.3 (2005-06-20) [i686-linux]
make: *** [check-swig-rb] Error 134
Kouhei, could you take a look at this?
Cheers,
David
--
David James -- http://www.cs.toronto.edu/~james
Re: 1.3.0 RC1 ready to be rolled
Posted by Jani Averbach <ja...@jaa.iki.fi>.
On 2005-10-13 12:57-0400, David James wrote:
> I'm not able to reproduce the test failures on my machine now. "make
> check-swig-rb" passes for me on both the 1.3.x branch and on trunk
> using r16666.
Just make sure, did you compile your apr with "--enable-pool-debug=yes"?
BR, Jani
--
Jani Averbach
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by David James <ja...@gmail.com>.
On 10/13/05, Jani Averbach <ja...@jaa.iki.fi> wrote:
> On 2005-10-13 10:10-0500, kfogel@collab.net wrote:
> >
> > Dang. Erm, any Ruby-enabled people out there know what's going on?
>
> Ruby has been failing since:
>
> Date: Mon, 29 Aug 2005 22:41:57 -0600
> Subject: swig-rb trunk r15982: PASS (x86_64-unknown-linux-gnu shared)
>
> That was the last time when it has passed on trunk, if this is any help.
> David James reproduced it on his machine at one point.
I'm not able to reproduce the test failures on my machine now. "make
check-swig-rb" passes for me on both the 1.3.x branch and on trunk
using r16666.
Cheers,
David
--
David James -- http://www.cs.toronto.edu/~james
Re: 1.3.0 RC1 ready to be rolled
Posted by Jani Averbach <ja...@jaa.iki.fi>.
On 2005-10-13 10:10-0500, kfogel@collab.net wrote:
>
> Dang. Erm, any Ruby-enabled people out there know what's going on?
Ruby has been failing since:
Date: Mon, 29 Aug 2005 22:41:57 -0600
Subject: swig-rb trunk r15982: PASS (x86_64-unknown-linux-gnu shared)
That was the last time when it has passed on trunk, if this is any help.
David James reproduced it on his machine at one point.
BR, Jani
--
Jani Averbach
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by kf...@collab.net.
David James <ja...@gmail.com> writes:
> John Peacock pointed out some build issues with the Perl bindings,
> where installed libraries link to build-dir libraries. Max, will
> reverting r16486 fix the problem? If so, perhaps we should revert
> r16486 and propose the revert for backport to the 1.3.x branch.
>
> See: http://svn.haxx.se/dev/archive-2005-10/0556.shtml
Independently of Perl bindings stuff, John also reminded us of some
Ruby bindings problems. The tail of 'swig-rb trunk r16692' results
from svntest@jaa.iki.fi say this:
cd subversion/bindings/swig/ruby && /bin/sh \
/srv/svntest/svntest/obj-sh/libtool --tag=CC --silent --mode=link \
x86_64-pc-linux-gnu-gcc -shrext .so -shared -rpath \
/usr/lib/ruby/site_ruby/1.8/x86_64-linux/svn/ext -avoid-version \
-module -o wc.la svn_wc.lo
../../../../subversion/bindings/swig/ruby/libsvn_swig_ruby/\
libsvn_swig_ruby-1.la
../../../../subversion/libsvn_wc/libsvn_wc-1.la
../../../../subversion/libsvn_subr/libsvn_subr-1.la
libtool: link: warning: `/usr/lib/libgdbm.la' seems to be moved
libtool: link: warning: `/usr/lib/libdb-4.3.la' seems to be moved
libtool: link: warning: `/usr/lib/libexpat.la' seems to be moved
libtool: link: warning: `/usr/lib/libgdbm.la' seems to be moved
libtool: link: warning: `/usr/lib/libdb-4.3.la' seems to be moved
libtool: link: warning: `/usr/lib/libexpat.la' seems to be moved
libtool: link: warning: `/usr/lib/libgdbm.la' seems to be moved
libtool: link: warning: `/usr/lib/libdb-4.3.la' seems to be moved
libtool: link: warning: `/usr/lib/libexpat.la' seems to be moved
cd /srv/svntest/svntest/obj-sh/subversion/bindings/swig/ruby; \
/usr/bin/ruby -I \
/srv/svntest/svntest/svn/svn_trunk/subversion/bindings/swig/ruby \
/srv/svntest/svntest/svn/svn_trunk/subversion/bindings/ \
swig/ruby/test/run-test.rb --verbose=normal
Loaded suite test
Started
........................................................................................../srv/svntest/svntest/svn/svn_trunk/subversion/bindings/swig/ruby/test/test_ra.rb:108: [BUG] Segmentation fault
ruby 1.8.3 (2005-09-21) [x86_64-linux]
make: *** [check-swig-rb] Error 134
Dang. Erm, any Ruby-enabled people out there know what's going on?
-Karl
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: 1.3.0 RC1 ready to be rolled
Posted by David James <ja...@gmail.com>.
On 10/12/05, kfogel@collab.net <kf...@collab.net> wrote:
> The magic pathrev is:
>
> http://svn.collab.net/repos/svn/branches/1.3.x/ @ r16689
>
> Ben Reser, is today good for you to roll? I know we originally said
> Wednesday, but then we got stuck on those Windows bindings problems
> for a bit, which pushed us into today (Thursday).
>
> Thanks to quite a few people in IRC, for that final push of change
> reviewing and voting. I won't mention everyone's name here, but You
> Know Who You Are... and if You don't Know Who You Are, well, that's
> what we have 'svn log' for.
John Peacock pointed out some build issues with the Perl bindings,
where installed libraries link to build-dir libraries. Max, will
reverting r16486 fix the problem? If so, perhaps we should revert
r16486 and propose the revert for backport to the 1.3.x branch.
See: http://svn.haxx.se/dev/archive-2005-10/0556.shtml
Cheers,
David
--
David James -- http://www.cs.toronto.edu/~james