I've uploaded a couple of extra fixes in knotifications 5.27, could you try
these?
I'm not able to reproduce this issue, so I can't really test this.
On October 17, 2016 8:58:35 AM GMT+02:00, "Alex Dănilă" <nuor...@gmail.com>
wrote:
>Hi,
>
>If there's a way to provide more info (even with minimal
>recompilation),
>I'd be glad to. Fixing this freeze can't happen to soon.
>
>Running kwin and plasmashell from the terminal I get the following
>outputs when the freeze happens:
>
>Currrent active notifications: QHash(("notification 211", "Network
>managementBluetooth (A0:E4:53:AC:F7:07)"))
>Guessing partOf as: 0
> New Notification: "Xperia Z1 Compact Network (nap)" "Connection
>'Xperia Z1 Compact Network (nap)' deactivated." -1 & Part of: 0
>QXcbClipboard::setMimeData: Cannot set X11 selection owner
>kde.systemtray: Wrong IconThemePath
>"/home/alex/.dropbox-dist/dropbox-lnx.x86_64-12.4.22/images" : too
>short
>or does not end with 'icons'
>kde.systemtray: Wrong IconThemePath
>"/home/alex/.dropbox-dist/dropbox-lnx.x86_64-12.4.22/images" : too
>short
>or does not end with 'icons'
>[kde.systemtray: Wrong IconThemePath - Repeated endlessly]
>QXcbConnection: XCB error: 2 (BadValue), sequence: 31522, resource id:
>54525973, major code: 142 (Unknown), minor code: 3
>QXcbConnection: XCB error: 2 (BadValue), sequence: 31554, resource id:
>155189268, major code: 142 (Unknown), minor code: 3
>QXcbConnection: XCB error: 2 (BadValue), sequence: 31580, resource id:
>65011733, major code: 142 (Unknown), minor code: 3
>QXcbConnection: XCB error: 2 (BadValue), sequence: 31593, resource id:
>54525973, major code: 142 (Unknown), minor code: 3
>QXcbConnection: XCB error: 2 (BadValue), sequence: 31606, resource id:
>150994972, major code: 142 (Unknown), minor code: 3
>QXcbConnection: XCB error: 2 (BadValue), sequence: 31621, resource id:
>60817459, major code: 142 (Unknown), minor code: 3
>
>However, the QXcbConnection part seems unrelated, since it is not
>always
>correlated with the notification and the freeze. I've also attached
>.xsession-errors from the last minutes when this reproduce, but to me
>it
>seems to not contain anything related.
>
>Alex
>
>
>On 11/10/16 01:20, Alex Dănilă wrote:
>> Hola!
>>
>> I believe this is rather and interaction between KWin and
>Plasmashell.
>> If I stopped the Plasmashell process, the freeze doesn't happen
>> anymore (those notifications are drawn at the top of the screen in
>> small windows). A further argument for this idea is that after the
>> freeze the screen goes one "frame" back, and that the music and
>cursor
>> are not affected.
>>
>> I'll note that after restarting Plasmashell this doesn't happen as
>> reliably as before.
>>
>> However, if you know a good way of triggering this kind of
>> notifications, it will be my pleasure to do it.
>>
>> Thank you,
>> Alex
>>
>> On 10/10/16 13:14, Maximiliano Curia wrote:
>>> ¡Hola Alex!
>>>
>>> El 2016-10-08 a las 22:32 +0300, Alex Dănilă escribió:
>>>> Package: kwin-x11 Version: 4:5.7.4-1 Severity: normal
>>>
>>>> The entire screens freeze for a couple of seconds (3-4, but
>>>> sometimes seemed close to 10) everytime notifications from the tray
>
>>>> are about to scroll down or disappear. This only affects drawing of
>
>>>> screen content, it doesn't affect sound or cursor motion.
>>>
>>>> If relevant, the particular tray notifications are about bluetooth
>>>> devices, I'll attach a screenshot with this. The particular order
>is
>>>> like this:
>>>> -two messages (M1 - connection deactivated, M2 - connection timed
>>>> out) appear one above the other -2-3 seconds pass -freeze -2
>-freeze
>>>> ends -M1 disappears, M2 animates going down -messages reappear and
>>>> animate again, but without freezes.
>>>
>>>> I'll send a screenshot after the bug is created, and if possible a
>>>> recording.
>>>
>>> As reported it seems that the issue is bluetooth specific, and it
>>> might be caused by an underlying issue of the bluetooth kernel
>module
>>> or hardware device. In my limited tests I couldn't reproduce this
>>> behaviour in my laptop.
>>>
>>> Would it be feasible for you to test this multiple bluetooth
>>> notifications in a different desktop environment in order to rule
>out
>>> a kernel/hardware issue?
>>>
>>> Can you reproduce this triggering a different type of notifications?
>>>
>>> Happy hacking,
>>
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
