[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-pim
Subject:    Re: [Kde-pim] Testing out 4.10.2 against our identified blockers
From:       Volker Krause <vkrause () kde ! org>
Date:       2013-04-08 18:25:46
Message-ID: 3598010.0McKADQggz () vkpc5
[Download RAW message or body]

On Monday 08 April 2013 10:30:32 Georg C. F. Greve wrote:
> Hi all,
> 
> Since 4.10.2 is out, we probably out to figure out how we did against the
> blockers for advertising KDE PIM more widely.
> 
> Unfortunately, I have to report that
> 
> 	https://bugs.kde.org/show_bug.cgi?id=312460
> 
> can still be reproduced with 4.10.2. In other words, I still have to restart
> Kontact multiple times a day to get to my mail (since I see I have new
> mail, but no way to reach it), including the usual 5-6 minutes of Kontact
> being fully unusable while the calendar is being built up.
> 
> Will try switching to individual applications to make restart less painful.
> 
> Since everyone seems to say Stephen Kelly is the only one who can fix this:
> 
> Dear Stephen, can you pretty, pretty please take a look at this? :)

well, a way to reproduce this might also enable others to help fixing it :)

I for example do not see this here, and I expect my setup to be somewhat 
similar to yours (large offline IMAP accounts). Kontact on my laptop usually 
does not get restarted many days in a row, so there's definitely enough 
opportunity to see this I think.

So (as usual) this is probably not an universal problem but it obviously 
depends on some yet unknown conditions to be triggered. That part needs to be 
done at least with someone affected involved, I can only point you to the 
places I'd look at:

- check akonadiconsole debugger and debug output for retrieval errors

- unread count updates, so we can assume monitor signals arrive from the 
server, might still be worth checking if they make it out of Akonadi::Monitor 
or if they are filtered out (gdb or debug output in Monitor or maybe even 
better in ETM)

- use GammaRay to check if the items don't exist in ETM or if they are dropped 
somewhere above that in the proxy stack

The last two should massively narrow down the places to look at, basically 
pointing to a single class already.

regards,
Volker

_______________________________________________
KDE PIM mailing list kde-pim@kde.org
https://mail.kde.org/mailman/listinfo/kde-pim
KDE PIM home page at http://pim.kde.org/
[prev in list] [next in list] [prev in thread] [next in thread]