Browser failed to start

[Micheal5311]

I installed sugar as a desktop environment to test it before I give a laptop to my niece with it and the browser is failing to start

[walterbender]

Could you please open the Log activity and see if there is an entry for the Browse activity. It may provide a clue as to what is causing the issue.

[chimosky]

Agreed, could you also tell us what version of browse you're running? You can see the version in the list view from the home view.

[Micheal5311]

Browse is version 208

[quozl]

Top right of photograph, failure to load the shared library libwebkit2gtk-4.x is probably a root cause. The message is truncated, so I can't isolate further. I suggest pasting the log rather than using a photograph.

This may be a mistake of Linux distribution rather than Sugar. Such as missing dependencies. What release of Linux are you using? Is there a libwebkit2gtk*so file in /usr/lib anywhere?

[Micheal5311]

I am using fedora 40, and there is no libwebkit2gtk

[Micheal5311]

my assumtion is that is the problem

[quozl]

Is odd, as there is a clear declaration of dependency. https://packages.fedoraproject.org/pkgs/sugar-browse/sugar-browse/fedora-40.html

And the file should be there if the package is installed. https://packages.fedoraproject.org/pkgs/webkitgtk/webkit2gtk4.1/fedora-40-updates.html

Please check if you have Fedora package webkit2gtk4.1 installed, and tell us what version of package it is?

I've not had a chance to test Fedora 40, sorry.

[Micheal5311]

i do have the package installed, I need it for lutris to work, the version is 2.44.1-2.fc40

[chimosky]

i do have the package installed, I need it for lutris to work, the version is 2.44.1-2.fc40

Can you open a python REPL and import it to be sure Python can find it?

[SnehaPoojary20]

I also encountered this issue while testing the Browse activity. Since the Log Activity option is unavailable, I attempted to check logs manually but couldn't access the terminal. I suspect this could be due to a missing WebKit dependency or a Python GI issue. Has anyone tried debugging this using an alternative method?

[quozl]

I've not seen this issue myself, but I've always been able to use SSH, kernel command line break-in, live-boot, or other methods to get logs.