Review Request: Do not wait for remaining Dr.Konqi instances without timeout when ending the KDE session

Rolf Eike Beer kde at opensource.sf-tec.de
Mon Dec 10 12:54:59 GMT 2012


Am , schrieb Jekyll Wu:
> 于 2012年12月10日 19:27, Rolf Eike Beer 写道:
>>> The only downside is some backtrace might be lost. But I don't 
>>> think
>>> that is a big deal. Crashes during shutdown are rare cases 
>>> nowadays(I
>>> hope I'm right), and users noticing those crashes only hours later 
>>> are
>>> the rare case in rare cases (but very annoying).
>>
>> I regularly get them, most times something Kontact related. What 
>> about
>> just saving the backtrace to disk? Next step could be on next 
>> startup to
>> check if there are automatically saved backtraces and ask if to
>> drop/save/report them.
>
> Just to make it clear, when I say "some backtrace might be lost" I
> mean under the cases that bug 126073 complains: "Users think their
> systems will shutdown as expected but hours later find it hasn't
> because startkde waits for Dr.Konqi forever". The proposed change 
> will
> not stand in the way of users who already notice the crash at the
> first place and want to save/report the backtrace.
>
> Saving backtrace to disk is of course good, but unfortunately
> Dr.Konqi doesn't seem to have a dbus call for that operation at the
> monent. As for "Next step", see the comment from George Kiagiadakis.
> That is not an easy task and might take a long time to implement or
> never.

Add one. ;)

> On the other hand, bug 126073 is a big annoyance to anyone who has
> ran across it. And it has been there and annoying users for a long
> time. Comparing the big benefit and small lost of the proposed 
> change,
> I think that "Some backtraces might be lost" is really not a big 
> deal.

Agreed. But please set the timeout to something like 5 minutes. When I 
want to shutdown I want to shutdown, maybe because my batteries are 
nearly empty. Waiting for an hour ist far too long IMHO.

Eike




More information about the kde-core-devel mailing list