Xapian discuss - Jun 2007 - Deleted documents not deleted

I seem to be seeing cases where I call db.delete_document(somedocid) with
no error, then flush() and delete the database object, but the document is
still there after process exit. The write lock is normally deleted, so it
appears that the database close finished normally.

If I then then call delete_document(somedocid) from another
command/process, this time it goes away.

I've been seeing this forever (it's not related to a recent xapian
version
or a flint/quartz thing). This seems to be somehow related to the index
state (on a given index, deletes work until they stop working, and I have
to reindex from scratch to get back to normal).

Would someone have any beginning of an idea of what could cause this ? I'm
quite probably doing something stupid.

Regards,
Jean-Francois Dockes

On Tue, Jun 19, 2007 at 07:20:18PM +0200, Jean-Francois Dockes
wrote:
> I seem to be seeing cases where I call db.delete_document(somedocid) with
> no error, then flush() and delete the database object, but the document is
> still there after process exit. The write lock is normally deleted, so it
> appears that the database close finished normally.
> 
> If I then then call delete_document(somedocid) from another
> command/process, this time it goes away.
> 
> I've been seeing this forever (it's not related to a recent xapian
version
> or a flint/quartz thing). This seems to be somehow related to the index
> state (on a given index, deletes work until they stop working, and I have
> to reindex from scratch to get back to normal).
Do you see this in 1.0.0 or later?

This fix went in before 1.0.0, but I can't recall exactly what the symptoms
were, nor find a relevant mail or bug report...

Mon Apr 09 14:50:40 BST 2007  Olly Betts <olly@survex.com>

        * backends/flint/flint_database.cc: Delete the corresponding entry
          (if any) from doclens in delete_document().  Add assertion to
          add_document_() that the corresponding entry in doclens isn't
          already set, but in a non-debug build overwrite any existing
          entry as that's more likely to be correct.
        * backends/quartz/quartz_database.cc: Ditto.

Cheers,
    Olly

Yes, the problem seems to be gone. Any idea why I was apparently the only
one to trigger it ? 

Thanks a lot,

J.F.

Olly Betts writes:
 > On Wed, Jun 20, 2007 at 11:44:30AM +0200, Jean-Francois Dockes wrote:
 > > I seem to have managed to reproduce it with a simple sequence on a
small
 > > data set.
 > 
 > I think this patch should fix the problem - can you try it out and let
 > me know:
 > 
 > http://oligarchy.co.uk/xapian/patches/xapian-delete-document.patch
 > 
 > Cheers,
 >     Olly