Of course, we also can insist in the current solution of sending everything to /dev/null and avoid the side effects described by HuHa. In that case, it's still true that we have the same two options - fix RedirectToLog which is clearly not working or implementing the redirection as ruby code in YaST. I'll check what's the problem with RedirecToLog. Maybe it's completely useless or maybe it's being called too late or simply not called at all.