Howard Chu wrote:

I have committed a patch to syncprov.c for this, but I haven't reproduced the original problem so I can't verify that it is fixed. Running test050 by itself 20 times never showed any problem for me. (Dual core OpenSUSE 10.2)

Note that I reported it especially for RE24 since I've only tested it there. I will check HEAD more thoroughly with your patch in HEAD. Any chance to see exactly this patch in RE24 any time soon so that I can test with this particular code-base? Yes, I could port your patch to RE24 myself but I'd prefer if we have exactly the same code-base.

Ciao, Michael.