LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Slackware (https://www.linuxquestions.org/questions/slackware-14/)
-   -   All Things KDE5/Plasma for Slackware Users. (https://www.linuxquestions.org/questions/slackware-14/all-things-kde5-plasma-for-slackware-users-4175670109/)

bassmadrigal 07-21-2021 04:33 PM

Quote:

Originally Posted by igadoter (Post 6268558)
Don't need
Code:

$ ldd /usr/lib64/libKF5Solid.so.5.79.0  | grep libmount
$

but I am running Live Edition of AlienBob. Maybe AlienBob is using own builds of KDE in Live Edition. So at this point I am little out of discussion.

No, Alien Bob doesn't use a different version of KDE in this Slackware Live (at least not until he brings back ktown), you're just running an older version. We already knew the older version didn't have issues. You should probably use slackpkg to upgrade your system or refresh your install with the newer ISO generated on the 16th (the solid update was pushed on the 10th) using iso2ush.sh -r.

Code:

./iso2usb.sh -i slackware64-live-current.iso -r -o /dev/sdc

igadoter 07-21-2021 04:49 PM

Quote:

Originally Posted by LuckyCyborg (Post 6268528)
There are used ONLY functions from libmount, and they DO some safety checks.

So just cut off this part of code and build as separate app. No need to debug all KDE.

igadoter 07-21-2021 04:57 PM

Quote:

Originally Posted by bassmadrigal (Post 6268570)
No, Alien Bob doesn't use a different version of KDE in this Slackware Live (at least not until he brings back ktown), you're just running an older version. We already knew the older version didn't have issues. You should probably use slackpkg to upgrade your system or refresh your install with the newer ISO generated on the 16th (the solid update was pushed on the 10th) using iso2ush.sh -r.

Code:

./iso2usb.sh -i slackware64-live-current.iso -r -o /dev/sdc

Thanks for explanations. You right I should update to new version. But for me six months period between upgrades is reasonably. Person of AlienBob is kind of warranty snapshot he took is stable enough. Direct update with recent -current can brake this stability. So upgrade Live with Live.

j12i 07-21-2021 05:02 PM

Quote:

Originally Posted by ZhaoLin1457 (Post 6268568)
But on Slackware the /var/run is a bind mount of /run, then when we will mount a USB drive, it appears on both locations.

I knew already that this is a cause for the problem, as the code path with the crash is for exactly the case when the partition is mounted in multiple places.
What I didn't knoow is that other distributions do it differently. Also I assumed there's a good reason that Slackware does it this way. I'm going to ask about it in the requests thread. But I think the fault lies with KDE. The check should be there even if it is only needed on (FHS-) non-conforming systems.

marav 07-21-2021 05:16 PM

That would explain the dolphin crashes
Nevertheless it doesn't explain the cases where everything goes well...

It's the "non-systematic" side that leaves me perplexed.

j12i 07-21-2021 05:46 PM

I located the change introducing the offending code in KDE's GitLab and commented there about our issue. I think the commit/change author will be notified about my comment, and will wait for/if he has to something to say about this.

Good night, slackers!

igadoter 07-21-2021 05:50 PM

Quote:

Originally Posted by ZhaoLin1457 (Post 6268568)
In my humble opinion, this is the real issue and that's why the code works for others and not for us sometimes: the way how we handle the /var/run which probably is specific to Slackware.

This is only workaround to solve this particular issue. We have no idea how this change influence others parts of system. To do such things you need to keep history all changes. In emergency you can rollback system to its previous state. That's beyond common Slackware user. We can do some small things. But in general rely on Slackware is evolving without revolutions. We can't keep track of that kind of development.

j12i 07-21-2021 05:54 PM

Related discussion (my comment automatically appeared there).

igadoter 07-21-2021 06:01 PM

In the context of ZhaoLin1457 about binding /var/run this caught my attention from link above
Quote:

With bind mounts on Linux, UDisks returns a 'MountPoints' property that could
include several paths, try to get the actual mount point of the partition
and ignore bind mounts.
if I understand well ignore bind mounts would cause to ignore /var/run.

Edit: For me it looks like this function instead of /var/run returns /run - supposedly or ...nothing.

j12i 07-21-2021 06:10 PM

igadoter, a bind mount is not substantially different from any other mount (man mount may have more info). So if something is mounted with a bind mount (or multiple), those mount points are not in any way inferior to the original one, afaik. (Similar to a file with multiple hardlinks.) Thus, the solid code in question, as far as I can tell, isn't trying to decide which mount point is authoritative, but just tries to make one canonical for solid's sake.

j12i 07-21-2021 07:08 PM

I can't sleep. and Ahmad Samir of solid has already proposed a patch. I suspect it is equivalent to the one from LuckyCyborg's friend, but please test it everyone affected, for this will be in the next version of solid then.

ZhaoLin1457 07-22-2021 01:22 AM

Quote:

Originally Posted by j12i (Post 6268613)
I can't sleep. and Ahmad Samir of solid has already proposed a patch. I suspect it is equivalent to the one from LuckyCyborg's friend, but please test it everyone affected, for this will be in the next version of solid then.

I have tested Solid 5.84 built with the patch proposed by the Solid programmer and everything works fine.

The Plasma5 desktop is loaded properly without crashes and I tested also mounting an USB drive with Dolphin and systray widget.

atelszewski 07-22-2021 01:27 AM

Hi,

Quote:

Originally Posted by j12i (Post 6268613)
I suspect it is equivalent to the one from LuckyCyborg's friend

indeed it is.

--
Best regards,
Andrzej Telszewski

pchristy 07-22-2021 03:01 AM

Excellent that the upstream developer has fixed the issue, and thanks to all on here for their work in finding and fixing this nasty bug!

What a brilliant resource this forum is!

:)

--
Pete

marav 07-22-2021 03:20 AM

And because, I think it's also important to make sure that there's no regression
I recompiled solid with the patch and no issue to report ;-)


All times are GMT -5. The time now is 05:31 PM.