Thanks for the details, I wonder if the backtrace files should be re-created as new ones instead, or adding something like a bunch of spaces or date stamps to differentiate between the reports

Ah! I have seen that crazy "shaking" effect too, but I removed it from the report (this one or other similar) because I was unable to reproduce it

In fact I don't think is a typo, there's hundreds of matches about "luncher" in the source code. But talking about UX / usability, maybe the name should be a better one or more descriptive? (in fact i don't know yet what is for about / didn't played with that yet) mmh...

updated, much simpler description and easier to reproduce

I updated the info of the task to make clear "what it is about".

On this video we can see also the issue on the first seconds (the rest of the video is about another issue): https://www.youtube.com/watch?v=dudnIeoRMvA

I have this last file dumped https://pastebin.com/5hpqNpqr which is probably due to this crash, but I cannot recall for sure if is the backtrace for it or not :)

just a note, about the exact version of E in the Elive (actual) packages is located in the "git" part ( 0.2 4.0+18+git5ad58920a6-6buster18 ) of the version of the package, it includes other data like the tag but this one can be imprecise since its not directly related to a commit linearity

Another issue with artifacts:

• composite enabled
• move / resize set to "technical"
• "Avoid server grab" option selected

Nice! it seems to work pretty good here too , Im going to work with it enabled now for some time :)

Ok I did a few tests, these are the results:

Hum, I assume the debian packaging system simply put the "correct" permissions for the files hierarchy when packaging, so a hook of chmod has been needed to add to fix it

seems fixed from T8584, @triantares can you confirm is the same issue?

Seems like it works now, tried with a recent build, will close the report it for now in order to keep the reports clean and smaller :)

alright, I forgot that the theme comes from efl and not E, confirmed the fix here too :)

Can somebody confirm that the bug is fixed? I tried a new build but the issue seems to still here

packaging related issue, it requires to add a chmod 4755, I assume this is not possible to from the makefiles ? In any case I close this bug for now as resolved

It doesn't change, same issue

Thanks for the amazing explanation!

reading that line of "qtdemux", seems like to point to the GST libs, used by emotion, the actual elive build of EFL uses gst as "libgstreamer1.0-0" (1.0.0) for emotion , which is provided from Debian Buster

video of about probably the same bug: http://triantares.ddns.net/Elive/VdesksE23.mp4

Hum, not really, it happens all the time when i move the mouse to the bar (slowly or fast), but when slowly, it seems like to trigger multiple times the jumpy effect

@raster , seems like some systray apps requires xembed? is there possible to use xembed as an option?

little demo uploaded :) https://youtu.be/biiEuGPqWRQ

Ah, finally after to follow your details I was able to set my own keyboard configuration, by just enabling that checkbox it worked...

Ok so first I tried to remove the vaapi gstreammer package, and run it in a virtualbox (so with no specific hardware accelerators aparently), but I seen the same issues

I recorded this video, it seems to happen with any kind of video that I use

Sorry to not have answered before, I was not receiving notifications until now :)

duplicated

I played just a bit with the gifs like it happened to me before and I don't see the flickering anymore... bug fixed? :)

test comment notification

I'm just testing a recent build and I can confirm that systray doesn't works (it doesn't shows anything), not nm-applet or dropbox for example

Im not sure I can add these ones, since im building packages and so they will run in different machines. In fact I can set some options but the packaging system sets the default ones for Debian

Test failed @vtorri , version of libwayland-dev is "1.16.0"

I will try this tomorrow:

I build based on tags, so last tagged version:

[1]   no new version found
[1]   last was  v1.1.1 | fc1aef81432679a33dd37fa3735b7cca08666b85

What about an option in the builds like --build-type=stable or --build-type=development (so the default will be development) where in the stable mode, all the debug and messages are turned off, and the development all the information about reports are on ?

There's also some strange problem with the [ ] temporal checkbox, if you open the options it sometimes open it enabled and sometimes disabled, in the same time looks like sometimes is considered and sometimes no, maybe due to the instance stuff too

the status of this task is "resolved", is this bug stills here? anybody knows if the bug also happens on e17 so i should backport some fixes ?

more than taking long it seems like its "blocked" in some state, mmmh, maybe because it has not finished to load fully while another shell started ? idk

Hum, my zsh is pretty customized and so it could take a little longer to start (maybe 1-2 seconds?), what about if you add a "sleep 4" in your .bashrc to simulate a slow startup ? maybe can simulate the same behaviour

Actually the only issues that I see are, if you open tabs slowly it correctly works, but if you open tabs (ctrl + alt + t ) with the T pressed to continusouly open them, let;'s say opened 70 tabs, you have 2 problems:

• ctrl + next doesn't switch to the next tabs correctly, it has a black window, you can see it easly by resizing terminology
• if you close a shell tab, the next that comes is not focused by default, you need to click with the mouse on terminology to start typing on it