Bug 34964 - xfce: sometimes file manager windows pops up fullscreen for some partitions when installing kernel
Summary: xfce: sometimes file manager windows pops up fullscreen for some partitions w...
Status: NEW
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: Cauldron
Hardware: All Linux
Priority: Normal normal
Target Milestone: ---
Assignee: Mageia Bug Squad
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2026-01-10 13:06 CET by Morgan Leijström
Modified: 2026-01-25 21:23 CET (History)
2 users (show)

See Also:
Source RPM:
CVE:
Status comment: Seems to be related to system speed


Attachments

Description Morgan Leijström 2026-01-10 13:06:26 CET
Description of problem: Annoying!
Running xfce desktop: when a kernel is removed, file manager windows pops up fullscreen for all partitions

I saw this on 10alpha Live, bit I have noticed it for long.

Now i realise it may be a real disturbance if rok (remove-old-kernels), runs in auto mode like it is by default and "suddenly" from user point of view a bunch of file managers pops up fullscreen for no "to user" obvious reason.

Is is possible to by default avoid this behaviour at kernel removals (or whatever cause it)?  And still have the otherwise nice automount behaviuor in xfce?
Comment 1 Morgan Leijström 2026-01-10 21:18:15 CET
CC registered maintainer for xfce

CC: (none) => jani.valimaa

Comment 2 Lewis Smith 2026-01-20 21:47:49 CET
Morgan
Can you say whether this follows:
- automatic 'remove-old-kernels'
- manual use of same
- manual removal of a kernel package with (say) urpme.

CC: (none) => lewyssmith

Comment 3 Morgan Leijström 2026-01-22 00:08:02 CET
Testing this using Live 10alpha 64bit Xfce with persistence.
Another machine, another (my fastest biggest) USB stick.

After several in and uninstallations of kernels, and using both drakrpm, urpmi, urpme, rok:

§ Never a problem uninstalling

§ Installing most often no problem, but once with drakrpn and once with urpme it  happened: thunar pupped fullscreen

What it showed was the MSWindows10 partition on this computer. (When I expereinced the problem earlier it was always several thunar windows, and that computer is a dualboot WinXP / Mageia with several partitions.)

Looking in system journal i see:
 Jan 21 23:00:30 localhost ntfs-3g[31731]: Mounted /dev/sda3 (Read-Write, label "", NTFS 3.1)

And Xfce by default opens new plugged in partitions in thunar, so that part is logical.

But what makes the system mount it - sometimes but not always, when installing kernel?
 (and i think i have previously seen it when uninstalling kernel)

Zooming out in journal i see the below.
(Hm that sdb it errors on is the Mageia Live in use.  But anyway it seems irrelevant for this bug, as that appear once during boot and also during installs and removal of kernels when no partitions gets (re)mounted.) 

Jan 21 23:00:15 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sda
Jan 21 23:00:15 localhost bootloader-config[27613]: test_for_bad_drives(/dev/sda on sector #62)
Jan 21 23:00:15 localhost kernel:  sda: sda1 sda2 sda3 sda4
Jan 21 23:00:15 localhost bootloader-config[27613]: found a gpt partition table on /dev/sda at sector 0
Jan 21 23:00:15 localhost bootloader-config[27613]: running: udevadm settle
Jan 21 23:00:15 localhost kernel:  sda: sda1 sda2 sda3 sda4
Jan 21 23:00:15 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sda1
Jan 21 23:00:15 localhost bootloader-config[27613]: blkid gave: vfat 409F-E1C8
Jan 21 23:00:15 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sda2
Jan 21 23:00:15 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sda3
Jan 21 23:00:15 localhost bootloader-config[27613]: blkid gave: ntfs-3g F692A0D492A09A9F
Jan 21 23:00:15 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sda4
Jan 21 23:00:15 localhost bootloader-config[27613]: blkid gave: ntfs-3g 34FC2E0BFC2DC846
Jan 21 23:00:15 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sdb
Jan 21 23:00:15 localhost bootloader-config[27613]: blkid gave: iso9660 2025-12-20-14-23-01-00 Mageia-10-a1-Live-Xfce-x86_64
Jan 21 23:00:15 localhost bootloader-config[27613]: test_for_bad_drives(/dev/sdb on sector #62)
Jan 21 23:00:15 localhost udisksd[835]: Error probing device: Error sending ATA command IDENTIFY DEVICE to '/dev/sdb': Unexpected sense data returned:
                                        0000: 70 00 05 00  00 00 00 0a  00 00 00 00  20 00 00 00    p........... ...
                                        0010: 00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00    ................
                                         (g-io-error-quark, 0)
Jan 21 23:00:17 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sdb
Jan 21 23:00:17 localhost bootloader-config[27613]: blkid gave: iso9660 2025-12-20-14-23-01-00 Mageia-10-a1-Live-Xfce-x86_64
Jan 21 23:00:17 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sdb
Jan 21 23:00:17 localhost bootloader-config[27613]: blkid gave: iso9660 2025-12-20-14-23-01-00 Mageia-10-a1-Live-Xfce-x86_64
Jan 21 23:00:17 localhost bootloader-config[27613]: found a dos partition table on /dev/sdb at sector 0
Jan 21 23:00:17 localhost bootloader-config[27613]: guess_geometry_from_partition_table sdb: 30400/255/63
Jan 21 23:00:17 localhost bootloader-config[27613]: is_geometry_valid_for_the_partition_table failed for (sdb1, 8291023): 1023,63,31 vs 516,23,34 with geometry 30400/255/63
Jan 21 23:00:17 localhost bootloader-config[27613]: running: udevadm settle
Jan 21 23:00:17 localhost udisksd[835]: Error probing device: Error sending ATA command IDENTIFY DEVICE to '/dev/sdb': Unexpected sense data returned:
                                        0000: 70 00 05 00  00 00 00 0a  00 00 00 00  20 00 00 00    p........... ...
                                        0010: 00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00    ................
                                         (g-io-error-quark, 0)
Jan 21 23:00:17 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sdb1
Jan 21 23:00:17 localhost bootloader-config[27613]: blkid gave: iso9660 2025-12-20-14-23-01-00 Mageia-10-a1-Live-Xfce-x86_64
Jan 21 23:00:17 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sdb2
Jan 21 23:00:17 localhost bootloader-config[27613]: blkid gave: vfat F7FA-7802 MGAISO-ESP
Jan 21 23:00:17 localhost bootloader-config[27613]: running: blkid -o udev -p /dev/sdb3
Jan 21 23:00:17 localhost bootloader-config[27613]: blkid gave: ext4 77a1f039-b14b-4c57-a49e-218fb0cc9871 mgalive-persist
Jan 21 23:00:17 localhost bootloader-config[27613]: running: mkinitrd -f /boot/initrd-6.18.4-desktop-1.mga10.img 6.18.4-desktop-1.mga10
Jan 21 23:00:30 localhost polkit-agent-helper-1[31666]: pam_unix(polkit-1:auth): user [root] has blank password; authenticated without it
Jan 21 23:00:30 localhost ntfs-3g[31731]: Version 2022.10.3 external FUSE 29
Jan 21 23:00:30 localhost ntfs-3g[31731]: Mounted /dev/sda3 (Read-Write, label "", NTFS 3.1)
Jan 21 23:00:30 localhost ntfs-3g[31731]: Cmdline options: rw,uhelper=udisks2,nodev,nosuid,uid=1000,gid=1000,windows_names
Jan 21 23:00:30 localhost ntfs-3g[31731]: Mount options: uhelper=udisks2,nodev,nosuid,allow_other,nonempty,relatime,rw,default_permissions,fsname=/dev/sda3,blkdev,blksize=4096
Jan 21 23:00:30 localhost ntfs-3g[31731]: Global ownership and permissions enforced, configuration type 7
Jan 21 23:00:30 localhost udisksd[835]: Mounted /dev/sda3 at /run/media/live/F692A0D492A09A9F on behalf of uid 1000
Jan 21 23:00:31 localhost dbus-daemon[1732]: [session uid=1000 pid=1732] Activating via systemd: service name='org.freedesktop.thumbnails.Thumbnailer1' unit='tumblerd.service' requested by ':1.15' (uid=1000 pid=1843 comm="Thunar --sm-client-id 2721fcc56-38be-468f-b77e-f07")
Jan 21 23:00:31 localhost systemd[1623]: Starting Thumbnailing service...
Jan 21 23:00:31 localhost dbus-daemon[1732]: [session uid=1000 pid=1732] Successfully activated service 'org.freedesktop.thumbnails.Thumbnailer1'
Jan 21 23:00:31 localhost systemd[1623]: Started Thumbnailing service.
Jan 21 23:01:05 localhost [RPM][27606]: install kernel-desktop-6.18.4-1.mga10-1-1.mga10.x86_64: success
Jan 21 23:01:06 localhost [RPM][27606]: install kernel-desktop-6.18.4-1.mga10-1-1.mga10.x86_64: success
Jan 21 23:01:06 localhost [RPM][27606]: Transaction ID 69714c6c finished: 0

Summary: xfce - when a kernel is removed, file manager windows pops up fullscreen for all partitions => xfce: sometimes file manager windows pops up fullscreen for some partitions when installing kernel

Comment 4 Morgan Leijström 2026-01-25 00:12:55 CET
Observation on timing: the filemanager pops up (showing a partition on main disk) between downloading the kernel package and before unpacking (i see xz consuming CPU).

The problem is pretty repeatable on the slow Acer D260 running an updated 10alpha1 on a USB.
Comment 5 Lewis Smith 2026-01-25 18:57:30 CET
I think this matters. It suggest that the FM reacts in a time-dependant way to some filesesystem change: if things are fast enough, this does not trigger; if slow enough, it does.
Also, you are playing with a USB system (relatively slow). Most people will have disc [equivalents] (relatively fast), and hence not see it.

What think you?

Some naive but painful questions:
- Can you try different file managers?
On my multi-desktop system, but running Xfce, sometimes I see Thunar, sometimes PCManFM.
- Can you add & try other desktops, say LXDE, LxQt

Status comment: (none) => Seems to be related to system speed

Comment 6 Morgan Leijström 2026-01-25 20:28:50 CET
I think so too:
Some timeout, or execution order thing.

I believe it is only when using xfce that this behaviour gets noted, because it by default automatically opens new drives (partition) in a new file manager window.

I do not have time to test more now.
Comment 7 Lewis Smith 2026-01-25 21:23:19 CET
Thanks. But you should be able to try another file manager, to see whether that or Xfce itself is reponsible. When you have time, of course.
TIA

Note You need to log in before you can comment on or make changes to this bug.