Re: Fwd: [Daniel Shahaf: Long-standing corruption on svn.apache.org]

2011-10-02 Thread Daniel Shahaf
And here's another pair: [[[ harmonia,0:~% /root/bin/dump-noderev.pl /x1/svn/asf /subversion r965497 id: 0-836401.0.r965497/8198 type: dir pred: 0-836401.0.r965496/3747 count: 47068 text: 965497 7817 368 368 e18a7fbe8bedde647c30c288bbc02c1d props: 866503 8246 73 0 d34667037ecfd7bab466410a1c52bda6

Re: [RFC] - Proper encoding for patch file?

2011-10-02 Thread Branko Čibej
On 08.09.2011 20:07, Mark Phippard wrote: This is a JavaHL issue. See the attached patch which resolves the problem I face. If I use the JavaHL diff API to produce a patch it fails if there are paths in the patch with UTF8 characters in the name. Here is an example of the Exception:

Re: SQL indices a WC format bump and 1.7

2011-10-02 Thread Branko Čibej
On 02.09.2011 16:30, Philip Martin wrote: Bert also suggests changing our other indices by adding wc_id and/or local_relpath thus allowing them to be UNIQUE. Can anyone confirm that UNIQUE indices are better? Just imagine, if the UNIQUE constraint did not imply an index, every INSERT or

Re: SQL indices a WC format bump and 1.7

2011-10-02 Thread Branko Čibej
On 02.09.2011 22:56, Greg Stein wrote: On Fri, Sep 2, 2011 at 13:18, Mark Phippard markp...@gmail.com wrote: Pardon my ignorance, but in a scenario like this where we want to just change some of the indexes, aren't we able to just bump the WC format on the fly automatically? IOW, can't we

Re: SQL indices a WC format bump and 1.7

2011-10-02 Thread Mark Mielke
On 10/02/2011 03:26 PM, Branko Čibej wrote: On 02.09.2011 16:30, Philip Martin wrote: Bert also suggests changing our other indices by adding wc_id and/or local_relpath thus allowing them to be UNIQUE. Can anyone confirm that UNIQUE indices are better? Just imagine, if the UNIQUE constraint

Re: Fwd: [Daniel Shahaf: Long-standing corruption on svn.apache.org]

2011-10-02 Thread Daniel Shahaf
Here are some more instances of the mergeinfo count bug: % f() { /root/bin/dump-noderev.pl /x1/svn/asf / $1 | grep -a minfo-cnt } REVISION minfo-cnt of /'s noderev r891679minfo-cnt: 34417890202 r905000minfo-cnt: 34417892519 r907500minfo-cnt:

Slight tweak to Nightlies

2011-10-02 Thread Gavin McDonald
Hi All, To make future maintenance easier, I propose moving all 1st of Month nightlies to a separate http://ci.apache.org/projects/subversion/nightlies/dist/archives area away from the current monthly set available in http://ci.apache.org/projects/subversion/nightlies/dist . As far as the page

First theory for explaining both kinds of corruption Re: Fwd: [Daniel Shahaf: Long-standing corruption on svn.apache.org]

2011-10-02 Thread Daniel Shahaf
tldr: The first part of this mail lists a few more instances. The second half of this mail offers a theory that explains both modes of corruption. --- I've run 'svn log -ql 7000' for the following fspaths and repositories: asf:/hadoop asf:/openejb asf:/james asf:/jackrabbit asf:/karaf

[PATCH] FSFS sanity check

2011-10-02 Thread Daniel Shahaf
I'm trying add a basic sanity check to cover the ridiculous bogosity where the root noderev thinks its predecessor is about 20 revisions old... [[[ Index: subversion/libsvn_fs_fs/fs_fs.c === --- subversion/libsvn_fs_fs/fs_fs.c

Re: [PATCH] FSFS sanity check

2011-10-02 Thread Nothing to see here, move along folks
Daniel Shahaf wrote on Mon, Oct 03, 2011 at 01:13:52 +0200: + if (root_noderev-predecessor_count != -1 + root_noderev-predecessor_count != rev); Nothing to see here; stray semicolon. I disable Will never be executed warnings so my compiler didn't warn me about it. Building on another

[svnbench] Revision: 1178283 compiled Oct 3 2011, 00:21:28

2011-10-02 Thread neels
/home/neels/svnbench/20111003-002436 Started at Mon Oct 3 00:24:36 UTC 2011 - Results for dir levels: 5 spread: 5 Timings for 5x5_1.7.x N min max avgoperation (unit is seconds) 6 323.44 330.10 327.03