On Thu, 11 Dec 2008, drago01 wrote: > That's not true a daemon that runs does not automatically degrade > performance ... most of them spend 99.9% of the time sleeping, waiting > to wake up when something happens. > Now tell me how does this affect performance in any way? > Increased start up time is the only thing .. as for memory they will > be simply paged out when a non sleeping process needs the memory. In theory that's all correct. But anyway there are bugs and/or issues inside of mcstransd (when taking it here as an example), which are not really fixed and still exist, even that the bug report is maybe closed now. Reproducing that is as far as I know still possible as described there. - https://bugzilla.redhat.com/show_bug.cgi?id=195916 - https://bugzilla.redhat.com/show_bug.cgi?id=457179 > Bugs happen and this was not a feature update but a security! one that > broke things, so doing "just bugfixes" wouldn't have avoided this .. Didn't we learn in the beginnung of my started thread, that it takes anyway ~ 24 hours for a push? So why then a daemon which checks more often? Sorry, but the possible reason of "having some own repository" still makes IMHO not necessary to check more often than once or at least twice per 24 hours for available updates... Greetings, Robert -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
