KIO scheduler issue

[Paulo Moura Guedes]

On Friday 19 August 2005 11:03, Waldo Bastian wrote:
> Yes, patch is ok, 

Great, I will commit then.

> but I don't think it takes care of the actual bug, only 
> the symptoms. It very much looks like that for some reason the slave gets
> killed twice.

I really made that assumption and tried to fix things on the application side, 
in linkchecker.*, but I can't really find a situation where the job is 
explicitly killed twice. What happens is, in some conditions, the job is 
killed before result but, when it happens, the job object is disconnected and 
set to zero. It would be fun if you can confirm that by checking 
linkchecker.cpp, it's only ~350 loc anyway ;)

Paulo