On 2013-05-06 23:49, Julian Reschke wrote:
On 2013-05-06 23:22, Julian Reschke wrote:
On 2013-05-06 22:24, Alex Parvulescu wrote:
dunno,

I tried a fresh checkout and I see the same thing

svn co https://dist.apache.org/repos/dist/dev/jackrabbit dist-new
cd dist-new
sh check-release.sh 2.4.4 608ae0c4b61362e72cf95f41bd4bf9732a74a7c1
...


Weird; it sure fails now for my notebook as well, but still passes on my
office machine.

The difference is that that machine has a hosts entry for
svn.apache.org, forcing it not to use the european svn mirror (Jukka
knows the details...). Maybe these two machines are out of sync?


Best regards, Julian

Indeed.

On 160.45.251.2:
svn log InternalVersionManagerBase.java | head -20
------------------------------------------------------------------------
r1479620 | reschke | 2013-05-06 17:16:31 +0100 (Mon, 06 May 2013) | 1
line

[maven-release-plugin]  copy for tag 2.4.4
------------------------------------------------------------------------
r1309924 | reschke | 2012-04-05 17:21:19 +0100 (Thu, 05 Apr 2012) | 1
line

JCR-3286: enhance error handling for empty predecessors property value
(backport
ed to 2.4)
------------------------------------------------------------------------
r1221446 | jukka | 2011-12-20 20:19:26 +0100 (Tue, 20 Dec 2011) | 1 line

[maven-release-plugin]  copy for branch 2.4
------------------------------------------------------------------------
r1186802 | reschke | 2011-10-20 14:46:42 +0100 (Thu, 20 Oct 2011) | 3
lines

JCR-3115: Versioning fixup leaves persistence in a state where the
node can't be
 made versionable again

Extend checker so that it gets the same VersionHistoryInfo object that
other cod
e gets when versioning gets enabled again.

On 140.211.11.4:

r1479620 | reschke | 2013-05-06 17:16:31 +0100 (Mon, 06 May 2013) | 1
line

[maven-release-plugin]  copy for tag 2.4.4
------------------------------------------------------------------------
r1362927 | reschke | 2012-07-18 14:03:22 +0100 (Wed, 18 Jul 2012) | 1
line

JCR-3393: InternalVersionManagerBase.calculateCheckinVersionName may
fail with NPE on broken versioning persistence
------------------------------------------------------------------------
r1309924 | reschke | 2012-04-05 17:21:19 +0100 (Thu, 05 Apr 2012) | 1
line

JCR-3286: enhance error handling for empty predecessors property value
(backported to 2.4)
------------------------------------------------------------------------
r1221446 | jukka | 2011-12-20 20:19:26 +0100 (Tue, 20 Dec 2011) | 1 line

[maven-release-plugin]  copy for branch 2.4

The latter server has r1362927, the other does not.

Best regards, Julian

So it seems that the european SVN server has a corrupted history for 2.4.4.

If anyone wants to check the release, you'll have to use svn.us.apache.org (yes, I asked on #asfinfra).

Best regards, Julian

Reply via email to