Dolphin causing kioslave5 to crash system

Following several unsuccessful attempts to upgrade from 13.2 to 14.0, I seem to have corrupted the original system in some way. Rolling back to the original using beadm has usually worked for me - but not this time. Maybe I did something without realising. Currently, on my 13.2 if I have Firefox running (for example) and then run Dolphin, the system crashes. Looking at the output of top -n -o swap 183 | less works, I can see that kioslave5 use climbs relentlessly, and unless I kill Dolphin it will crash KDM.
Using Thunar has no such effect.
So two questions:
  1. What can I do to fix this? (I don't like upgrading semi-broken systems)
  2. Should I try to upgrade to 14.1 instead (there seems to be some bugs around 14.0)
This is completely outside my previous (scant) experience.

All advice welcome.
 
All versions of FreeBSD use the exact same ports tree. Thus have the exact same packages available to them.

What can I do to fix this? (I don't like upgrading semi-broken systems)
The OS has very little to do with third party applications (ports/packages) because they all have the same available to them.
Should I try to upgrade to 14.1 instead (there seems to be some bugs around 14.0)
You probably should, 13.2 is now EoL and 14.0 will be EoL soon.
 
Something you have to keep in mind is that a boot environment doesn't include /home. So, if you have some corrupt or incorrect settings there, beadm can do nothing against that. Try to rename some possibly implicated config files in your profile to see what happen or create a new user and test it.
 
All versions of FreeBSD use the exact same ports tree. Thus have the exact same packages available to them.


The OS has very little to do with third party applications (ports/packages) because they all have the same available to them.

You probably should, 13.2 is now EoL and 14.0 will be EoL soon.
Thanks, you guys are just amazing - it took me longer to write, than you guys to answer.
However, here is another question to embarrass myself with.
After all my failed attempts to upgrade from 13.2 to 14.0, I have a long list of boot environments of 14 that don't work. I presume I should delete these using beadm or bectl. Are there any rules or guidlines for this - for example, should I start at the most recent and start deleting them back to the one that works?
Or do I need to delete any of them at all (apart from space reasons)?
Many thanks as always.
 
Something you have to keep in mind is that a boot environment doesn't include /home. So, if you have some corrupt or incorrect settings there, beadm can do nothing against that. Try to rename some possibly implicated config files in your profile to see what happen or create a new user and test it.
Please see my reply to SirDice. Very much appreciated.
 
After all my failed attempts to upgrade from 13.2 to 14.0, I have a long list of boot environments of 14 that don't work. I presume I should delete these using beadm or bectl. Are there any rules or guidlines for this - for example, should I start at the most recent and start deleting them back to the one that works?
Or do I need to delete any of them at all (apart from space reasons)?
Before an upgrade (minor or major) I always check the BEs and clean up everything. Check which BE is active right now, bectl list, look for a BE marked with NR. That's your current BE. If that one works, remove all the others. If it's not named 'default' rename it (that will avoid future confusion).

Ideally you do this clean up after you upgraded/updated and are sure everything's working.
 
BTW, there is a bug in KDE Dolphin according to some in the KDE (Linux) community.
A solution is given here: Mint Dolphin bug
I'm not going to bother as I've switched to Thunar (which is superior in some ways) and I hope to upgrade to 14.1 in the very near future.
Thanks again for your help.
 
For reasons too boring and embarrassing to detail, it looks like I have to fix my broken KDE installation before I can move on to upgrade etc.
The problem is presumably a config file associated with Dolphin, KDE or Qt.
When I run the following as a user (or even root).
Code:
dolphin -v
Qt: Session management error: None of the authentication protocols specified are supported
dolphin 23.08.5

I can't see help either here or on the Linux forums, but they certainly report the error.
If I don't get any useful feedback, do you think it worth raising again as a more specific question?
And what about KDE itself - is that a useful avenue?
All help appreciated.
 
Last edited by a moderator:
The same with CURRENT:

Code:
% dolphin -v
Qt: Session management error: None of the authentication protocols specified are supported
dolphin 23.08.5
% uname -aKU
FreeBSD mowa219-gjp4-zbook-freebsd 15.0-CURRENT FreeBSD 15.0-CURRENT main-n271559-c1287a3bb1b7 GENERIC-NODEBUG amd64 1500023 1500023
%

… failed attempts to upgrade from 13.2 to 14.0, I have a long list of boot environments of 14 that don't work. …

freebsd-version -kru ; uname -aKU

For us to know what's running. Thanks.


At a glance, I don't see relevance to kioslave5.
 
The same with CURRENT:

Code:
% dolphin -v
Qt: Session management error: None of the authentication protocols specified are supported
dolphin 23.08.5
% uname -aKU
FreeBSD mowa219-gjp4-zbook-freebsd 15.0-CURRENT FreeBSD 15.0-CURRENT main-n271559-c1287a3bb1b7 GENERIC-NODEBUG amd64 1500023 1500023
%



freebsd-version -kru ; uname -aKU

For us to know what's running. Thanks.



At a glance, I don't see relevance to kioslave5.
Thanks for your interest.
I now have 14.1 working, but I don't remember doing anything special. It could be it just took a long time for several of the attempts, and I was impatient.
To answer your question:
Code:
14.1-RELEASE
14.1-RELEASE
14.1-RELEASE-p2
FreeBSD 4Dec22 14.1-RELEASE FreeBSD 14.1-RELEASE releng/14.1-n267679-10e31f0946d8 GENERIC amd64 1401000 1401000
The kioslave comment came from a suggestion / comment on another thread. I ran dolphin and then looked at dmesg and got the following:
Code:
Security policy loaded: MAC/ntpd (mac_ntpd)
pid 1919 (baloo_file), jid 0, uid 1001: exited on signal 6 (core dumped)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2060 (dolphin)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2060 (dolphin)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2060 (dolphin)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2078 (kioslave5)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2078 (kioslave5)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2078 (kioslave5)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2078 (kioslave5)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2078 (kioslave5)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2078 (kioslave5)

Hopes that helps.
And thanks again for your interest.
 
balooctl disable

Instead, you can use deskutils/recoll with real-time indexing.

Please see bug 256269 and the linked reports.
Many thanks for taking the time out to advise.
I'll devote some time to it later today and let you know how it went.
Regardless, thanks again.
 
Many thanks for taking the time out to advise.
I'll devote some time to it later today and let you know how it went.
Regardless, thanks again.
Sorry to say that doesn't seem to be it.
I'll write more later for those interested, but the KDE handbook seems to be wrong as well.
I can't seem to configure dolphin on FreeBSD the way the handbook says.
No matter how I try to open dolphin, it always tries to open the file specified and an old file from long ago.
But as said, thanks for the idea - and that search utility looks really useful.
 
Can you share a link?

There's no mention of Dolphin in the FreeBSD Handbook (PDF).
That's the point. I can only see a handbook from KDE - and I'll send the link tomorrow. However, it doesn't work for FreeBSD - it doesn't translate (unless I'm missing something). Too late here now. More tomorrow. Many thanks again for your interest.
 
Can you share a link?

There's no mention of Dolphin in the FreeBSD Handbook (PDF).
Sorry to take so long. I've looked spasmodically at this issue, and all the handbook type information assumes that everything is working and gives you options.
It doesn't ever seem to cover how to fix misconfiguration and how to recover. I had the idea of re-loading Dolphin - but even that's not possible from pkg.
My next effort is to follow up on the KDE site advice about using VALGRIND - but it looks formidable for someone who just wants to use something rather than someone who wants to re-design it. Anyway, thanks for your efforts to help.
All ideas welcome.
 
If able, you could make a fresh user to test from. If it works there, then you are looking at either a configuration issue or possibly a bad interaction with 1+ files in the user directory. I've sometimes seen crashes I suspected were corrupted/incomplete files being read to get properties, make thumbnails, etc. Sometimes I have also tracked excessive kioslave activity which usually matches a sluggish dolphin and only clears by closing dolphin. Testing could involve setting permissions to block content from being seen by dolphin under your normal user or copying configurations (or parts of) from one user to the other (think kde5 keeps stuff in the .config folder but isn't the only thing writing there). Having a backup or copying configs to a new name as a backup can be helpful to go back to if you don't remember all tweaks that may have been done to relevant parts of the system.
 
Thanks to both Cath O'Deray and Mirror176, but I've stumbled across something interesting here. I will try the "add a new user" later today or tomorrow, but in my spare time I've been reading the KDE website - and it only offers help for FreeBSD installed for ports and asks you to recompile the concerned packages WITH-DEBUG, and points you at FreeBSD/FAQ. That means there's no real avenue for FreeBSD problems installed from packages. If I'm wrong here, please put me right.
FWIW, before I realised this I installed VALGRIND and ran it - although I couldn't install the debug packages as recommended in the FreeBSD/FAQ.
The output is in the log.txt file attached. It seems to point to the error, but my knowledge level isn't up to it.
If anyone can help me re-configure Dolphin to get it working again, I would be immensely grateful.
The bigger issue of KDE help for systems installed from packages, I'll leave to others to comment.
Thanks.
 

Attachments

  • log.txt
    8.8 KB · Views: 15
Especially for Cath O'Deray and Mirror176. This whole issue seems to revolve around Polkit configuration, and unfortunately nothing is particularly clear.
There has been three versions of "polkit-124" with no real explanation or advice from the developers or anyone else.
From my experience, that's very unusual; one of the reasons I use FreeBSD is the advice, guidance and patience.
But not in this case, it seems.
 
For anyone still reading this worn and desperate thread:
My latest attempt to get dolphin to run (even with balooctl disable as suggested by Cath O'Deray) running dolphin creates the following multiple lines of dmesg output before crashing the machine:
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2350 (dolphin)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2367 (kioslave5)

How dolphin's configuration has changed so much to create these errors is a mystery to me - but then I'm a newbie.
All suggestions welcome.
 
Code:
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2350 (dolphin)
WARNING: autofs_trigger_one: request for /net/ completed with error 5, pid 2367 (kioslave5)

Remove this line from /etc/auto_master:
Code:
/net            -hosts          -nobrowse,nosuid,intr
Then restart automountd(8).
 
SirDice, that must be the fastest reply ever.
Unfortunately, although it did stop those /net/ errors, it created the following (seen on dmesg)
============================================================
da0: <Generic USB SD Reader 1.00> Removable Direct Access SCSI device
da0: Serial Number 058F312D81B
da0: 40.000MB/s transfers
da0: Attempt to query device size failed: NOT READY, Medium not present
da0: quirks=0x2<NO_6_BYTE>
da1 at umass-sim0 bus 0 scbus10 target 0 lun 1
da1: <Generic USB CF Reader 1.01> Removable Direct Access SCSI device
da1: Serial Number 058F312D81B
da1: 40.000MB/s transfers
da1: Attempt to query device size failed: NOT READY, Medium not present
da1: quirks=0x2<NO_6_BYTE>
da2 at umass-sim0 bus 0 scbus10 target 0 lun 2
da2: <Generic USB SM Reader 1.02> Removable Direct Access SCSI device
da2: Serial Number 058F312D81B
da2: 40.000MB/s transfers
da2: Attempt to query device size failed: NOT READY, Medium not present
da2: quirks=0x2<NO_6_BYTE>
da3 at umass-sim0 bus 0 scbus10 target 0 lun 3
da3: <Generic USB MS Reader 1.03> Removable Direct Access SCSI device
da3: Serial Number 058F312D81B
da3: 40.000MB/s transfers
da3: Attempt to query device size failed: NOT READY, Medium not present
================================================================
There must be something more basic wrong in some other configuration path. And my very basic level of knowledge doesn't help me.
I thought at one stage it was something to do with Polkit (as that's been patched three times without explanation - see pkg polkit-124_3. But that's as far as I go.


My /etc/auto_master: now looks like
# Automounter master map, see auto_master(5) for details.
#
# /net -hosts -nobrowse,nosuid,intr
# When using the -media special map, make sure to edit devd.conf(5)
# to move the call to "automount -c" out of the comments section.
#/media -media -nosuid,noatime,autoro
#/- -noauto

Thanks again for such a prompt reply, I'll now try and understand devd.conf - but more advice would be most welcome.
 
Back
Top