On 01/07/2012 at 16:48, chymian <chym...@gmx.net> wrote: > BUT: the point here: > >> plasma should definitly NOT freeze on a lost network connection. > > or even, if a plasmoid can't update it's content!
This is probably caused by plasma design - i.e. all plasmoids run under single process. So, if one plasmoid tries to connect, it virtually hangs whole desktop for that time. This issue was discussed some time ago on planet KDE, but I can't find that post right now. I have found that brainstorm discussion from few years back: http://forum.kde.org/viewtopic.php?f=90&t=45255 It seems that rationale was: plasmoid crash causes whole desktop to crash, so we should run them in separate processes. But plasmoids don't crash that often now, so we do not have to run them in separate processes. Your case shows that there is still pretty strong reason to run plasmoids in separate processes (if only my suppositions are right). You should definitely post your findings in KDE forums - either in linked thread, or in new thread (which may get more attention). But you should not post a bug report to Debian's BTS - there is really not much that Debian's KDE packaging team can do about it. As for corrupted configplasma-desktop-appletsrc file - it would be nice if you could provide step-by-step instructions how to reproduce that. I will check on Arch Linux with KDE to ensure that this is Debian issue. As KDE packaging team recently said on that mailing list, they prefer that KDE bugs goes directly to KDE, not Debian BTS. Debian BTS is appropriate for packaging issues and things that do not happen outside Debian. -- Best regards Mirosław Zalewski -- To UNSUBSCRIBE, email to debian-kde-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/201207020044.40863.mini...@poczta.onet.pl
