[EdLUG] Kubuntu Desktop broken after 'apt-get dist-upgrade'

Roy roy at crossford.net
Tue Sep 27 19:29:23 UTC 2022 

Alistair,

As KDE is broken both of you and a new user, it sounds like something 
went wrong with the KDE-plasma install.

Run the KDE-plasma part of the install again. Its odd that it will not 
start for a new user.

I expect that plasma has its own logs somewhere but as a MATE user, I 
don't know what I'm looking for.

Regards,

Roy Bamford.



On 27/09/2022 20:16, Alistair wrote:
> Roy
>
> You are correct, once I fail to get a useful graphic session working I change to a second console to login and investigate.
> I tried another user (created last week).
> The only difference was that the spinning cog wheel disappeared  quickly, but the black screen with the white outline mouse pointer did come up.
> After a while it also timed-out to the graphic lockscreen.
> Again the log file for this user was in the home folder.
>
> Regards,
> Alistair
>
> On Tuesday, 27 September 2022 19:18:42 BST Roy wrote:
>> On 27/09/2022 19:13, Roy wrote:
>>> Alistair,
>>>
>>> Looking at dmesg first
>>>
>>> [    2.145084] [drm] Initialized i915 1.6.0 20201103 for 0000:00:02.0 on minor 0
>>>
>>> That's the kernel creating /dev/dri/card0 and
>>>
>>> [    2.277515] Console: switching to colour frame buffer device 240x67
>>> [    2.306335] i915 0000:00:02.0: [drm] fb0: i915drmfb frame buffer device.
>>>
>>> The kernel is good.
>>>
>>> Now your Xorg.0.log. The error about /dev/dri/card0 is gone and the
>>> modesetdriver is in use.
>>>
>>> That's correct too.
>>>
>>> At the end the log shows
>>>
>>> [    37.620] (II) modeset(0): Modeline "1920x1080"x0.0  110.92  1920 1968 2000 2080  1080 1083 1088 1111 +hsync -vsync (53.3 kHz e)
>>> [    80.054] (**) Option "fd" "27"
>>> [    80.054] (II) event2  - Power Button: device removed
>>>
>>> So Xorg ran for about 43 seconds, then shut down. There are no errors
>>> there either.
>>>
>>> The file name
>>>
>>> ~/.local/share/xorg/Xorg.0.logfile
>>>
>>> is significant.
>>>
>>> When the login screen is show, Xorg runs as root and the log is in
>>> /var/log.
>>>
>>> The file you posted is in your /home directory. As Xorg is running as
>>> you, you must have logged in already.
>>>
>>> That points to kde-plasma having problems because the kernel and Xorg
>>> are good.
>>>
>>> The potential problems fall into two sorts. The software install and
>>> the per user setup, the per user setup is the easiest to test.
>>>
>>> Add a new user to your system, just with the defaults that your system
>>> provides.
>>>
>>> Using the new user, does everything appear to work?
>>>
>>>
>>> Regards,
>>>
>>> Roy Bamford
>>>
>> Alistair,
>>
>> "So Xorg ran for about 43 seconds, then shut down." is not correct.
>>
>> [    80.062] (II) AIGLX: Suspending AIGLX clients for VT switch
>>
>> You switched to a text console leaving Xorg running. The conclusion is
>> not changed.
>>
>> Test with a new user.
>>
>>
>> Regards,
>>
>> Roy Bamford.
>>
>>
>>> On 27/09/2022 12:49, Alistair wrote:
>>>> Roy,
>>>>
>>>> I thought that if I am getting a screenlock / login screen, then the X window system must be starting, and thus creating a logfile...
>>>> I searched and found it at:
>>>> ~/.local/share/xorg/Xorg.0.logfile
>>>> The timestamp is: Sept 27 12:14
>>>>
>>>> Two new files have been uploaded to the dpaste site:-
>>>>
>>>> >From today's login, t480-dmesg-2022-09-27:
>>>> https://dpaste.com/77VLC9YMS
>>>>
>>>> >From today's login, t480-Xorg.0.log-2022-09-27
>>>> https://dpaste.com/8HSS7RK5S
>>>>
>>>> Hope these help.
>>>>
>>>> Regards,
>>>> Alistair
>>>>
>>>> On Sunday, 25 September 2022 21:37:13 BST Roy wrote:
>>>>> Alistair,
>>>>>
>>>>> None of the files in /dev, /sys and /proc are real files. They are the
>>>>> kernel exposing some of its internal data structures as files.
>>>>>
>>>>> Here I have
>>>>>
>>>>> $ ls -l /dev/dri/ total 0 crw-r----- 1 root video 226, 0 Sep 28 2021
>>>>> card0 crw-rw---- 1 root video 226, 64 Sep 12 2021 controlD64
>>>>>
>>>>> On card0, the 226 is the kernel major device number, the 0 is the minor
>>>>> device number. There are no file sizes there.
>>>>>
>>>>> You really don't want to know about the timestamps but other readers may
>>>>> be curious. I run this system with a static /dev. /dev entries are made
>>>>> by hand with mknod and the file timestamp is the create time. They do
>>>>> not indicate my uptime.
>>>>>
>>>>> Anyway, /dev/dri/card0 is the way that Xorg communicates with the kernel
>>>>> part of the video driver.
>>>>>
>>>>> If its present with the right owner/group/permissions you should be good.
>>>>>
>>>>> $ mount | grep devtmpfs dev on /dev type devtmpfs (...
>>>>>
>>>>> Run the above command. The sample output shows that a dynamically
>>>>> managed /dev generated by the kernel is in use. udev gets a signal when
>>>>> devices are added and removed, the kernel updates /dev and udev changes
>>>>> permissions and adds/removes symlinks.
>>>>>
>>>>> If its all automatic like that, I don't understand how you get
>>>>>
>>>>>    > [ 272.782] (EE) open /dev/dri/card0: No such file or directory
>>>>>
>>>>> That means that /dev/dri/card0 is missing. Incorrect permissions
>>>>> generate a Permission Denied error.
>>>>>
>>>>> If the file is missing, that's a kernel problem. If its there, Xorg
>>>>> should see it.
>>>>>
>>>>> That contradiction makes me ask if the dmesg and Xorg.0.log you shared
>>>>> are a self consistent set. Did they come from the same boot of the system?
>>>>>
>>>>> Gentoo does not have x11-apps but
>>>>> https://packages.debian.org/stretch/x11-apps  shows it includes
>>>>>
>>>>> - oclock and xclock, graphical clocks;
>>>>>      
>>>>>
>>>>> Providing you have twm and xterm, default startx will start.
>>>>>
>>>>> Regards,
>>>>>
>>>>> Roy Bamford
>>>>>
>>>>> On 25/09/2022 16:12, Alistair wrote:
>>>>>> Roy,
>>>>>>
>>>>>>> [ 272.782] (EE) open /dev/dri/card0: No such file or directory
>>>>>> This (zero size) file does exist, owner: root; group: video.
>>>>>>
>>>>>> I tried the startplasma-x11 command:
>>>>>>
>>>>>> $ startplasma-x11
>>>>>>
>>>>>> $DISPLAY is not set or cannot connect to the x server
>>>>>>
>>>>>> Searching on that error took me to the arch forum and wiki
>>>>>>
>>>>>> systemctl status sddm.service
>>>>>>
>>>>>> sddm service
>>>>>>
>>>>>> loaded: masked (Reason: unit sddm.service is masked.)
>>>>>>
>>>>>> Active: inactive (dead)
>>>>>>
>>>>>> (On a working system I get:
>>>>>>
>>>>>> ●sddm.service - Simple Desktop Display Manager     Loaded: loaded
>>>>>> (/lib/systemd/system/sddm.service; enabled; vendor preset: enabled)
>>>>>>       Active: active (running)since Sun 2022-09-25 08:55:39 BST...)
>>>>>>
>>>>>> I followed the trail of text files from:
>>>>>>
>>>>>> systemctl get-target
>>>>>>
>>>>>> graphical.target
>>>>>>
>>>>>> All contained similar to a working system, but did not lead me
>>>>>> anywhere useful.
>>>>>>
>>>>>> Trying:
>>>>>>
>>>>>> systemctl enable sddm.service
>>>>>>
>>>>>> (input password three separate times)
>>>>>>
>>>>>> ...
>>>>>>
>>>>>> ...
>>>>>>
>>>>>> ...
>>>>>>
>>>>>> Each get "===AUTHENTICATION COMPLETE==="
>>>>>>
>>>>>> The final line stated:
>>>>>>
>>>>>> Failed to enable unit: unit file /etc/systemd/system/sddm.service is
>>>>>> masked.
>>>>>>
>>>>>>> to test with startx, you need to install twm, xterm
>>>>>>> and xclock. Be warned that tmw only uses the primary
>>>>>>> mouse button.
>>>>>> Checking for these three and installing as required, xclock "has no
>>>>>> installation candidate" and suggested x11-apps instead.
>>>>>>
>>>>>> This has made no difference.
>>>>>>
>>>>>> Regards,
>>>>>>
>>>>>> Alistair
>>>>>>
>>>>>> On Thursday, 22 September 2022 17:41:29 BST Roy wrote:
>>>>>>
>>>>>>> Alistair,
>>>>>>> First, your t480-dmesg-2022-09-22:https://dpaste.com/F626WA62N
>>>>>> kernel log.
>>>>>>
>>>>>>> The i915 lines starting with
>>>>>>> [ 2.043003] i915 0000:00:02.0: vgaarb: deactivate vga console
>>>>>>> are good
>>>>>>> [ 2.233768] fbcon: i915drmfb (fb0) is primary device
>>>>>>> [ 2.245684] Console: switching to colour frame buffer device 240x67
>>>>>>> Tells that kernel support for your Intel GPU in the kernel is correct
>>>>>>> and working.
>>>>>>> That's good as the Xorg driver will use it.
>>>>>>> Now your Xorg.0.log-2022-09-22
>>>>>>> Xorg is doing its automatic thing, that works for a single display setup
>>>>>>> [ 272.780] (==) Matched modesetting as autoconfigured driver 0
>>>>>>> [ 272.780] (==) Matched fbdev as autoconfigured driver 1
>>>>>>> [ 272.780] (==) Matched vesa as autoconfigured driver 2
>>>>>>> Its worked out that there are three possible drivers for your GPU.
>>>>>>> They all loaded.
>>>>>>> [ 272.782] (EE) open /dev/dri/card0: No such file or directory
>>>>>>> That's a bad thing. That /dev file is created by the kernel. The kernel
>>>>>>> part of the driver started OK. Is that file really missing?
>>>>>>> Its required for hardware video acceleration.
>>>>>>> Eventually Xorg gets going on the fbdev driver.
>>>>>>> [ 272.782] (II) FBDEV(0): Creating default Display subsection in
>>>>>> Screen section
>>>>>>
>>>>>>> "Default Screen Section" for depth/fbbpp 24/32
>>>>>>> [ 272.782] (==) FBDEV(0): Depth 24, (==) framebuffer bpp 32
>>>>>>> The log ends with
>>>>>>> [ 401.971] (II) Server terminated successfully (0). Closing log file.
>>>>>>> which suggests that Xorg started, found nothing to do then exited.
>>>>>>> That's expected when you run startx as there is no ~.xinitrc file
>>>>>>> to tell Xorg what to do, so it tries to start three xterms and an
>>>>>>> analogue clock all wrapped up in twm. Quite likely, none of those
>>>>>>> things are installed.
>>>>>>> In short Xorg, worked but somewhat suboptimally.
>>>>>>> Rather than testing with startx, try startplasma-x11
>>>>>>> to test with startx, you need to install twm, xterm
>>>>>>> and xclock. Be warned that tmw only uses the primary
>>>>>>> mouse button.
>>>>>>> I use MATE. I suspect that plasma will not start properly
>>>>>>> without that  /dev/dri/card0 file because it will want hardware
>>>>>>> acceleration.
>>>>>>> The file is created by the kernels DEVTMPFS option.
>>>>>>> udev may create symlinks and fiddle with permissions but
>>>>>>> your file appears to be missing completely.
>>>>>>> Check that as a next step. If its missing, try another
>>>>>>> kernel.
>>>>>>> Regards,
>>>>>>> Roy Bamford.
>>>>>>> On 22/09/2022 11:19, Alistair wrote:
>>>>>>>> Roy
>>>>>>>> I think the Xorg.0.log file is all there is.
>>>>>>>> This time I am including the Xorg.0.log.old file for comparison.
>>>>>>>> On the T480 the Xorg.0.log file is only 2/3 the size of the
>>>>>> Xorg.0.log.old file.
>>>>>>
>>>>>>>> The machine is a Thinkpad T480 with a i5-8350U CPU, and onboard
>>>>>> Intel UHD Graphics 620.
>>>>>>
>>>>>>>> I saved/copied the files to USB then copied to my good machine
>>>>>> before "select all; copy; paste." to the dpaste site.
>>>>>>
>>>>>>>> Uploaded files:
>>>>>>>> Xorg.0.log.old-2022-09-22:https://dpaste.com/787GM4L7X
>>>>>>>> Xorg.0.log-2022-09-22:https://dpaste.com/9A3LVBGCY
>>>>>>>> t480-dmesg-2022-09-22:https://dpaste.com/F626WA62N
>>>>>>>> Hope there is some useful data here.
>>>>>>>> Regards,
>>>>>>>> Alistair
>>>>>>>> On Wednesday, 21 September 2022 22:12:10 BST Roy wrote:
>>>>>>>>> Alistair,
>>>>>>>>> Both logs are truncated. They will be several screenfuls of each log.
>>>>>>>>> There are no errors there.
>>>>>>>>> Save them to some removable media then post the whole files.
>>>>>>>>> We can see that Xorg is doing its automatic thing and that you have a
>>>>>>>>> 8086:5917 GPU.
>>>>>>>>> 8086 means Intel.
>>>>>>>>> 5917 is the PCI product ID of your GPU. That's a UHD Graphics
>>>>>> 620. All
>>>>>>
>>>>>>>>> the interesting things are further down the log,
>>>>>>>>> Current logs would be best as they will show the current state of
>>>>>> your
>>>>>>
>>>>>>>>> system.
>>>>>>>>> Regards,
>>>>>>>>> Roy Bamford
>>>>>>>>> On 21/09/2022 19:24, Alistair wrote:
>>>>>>>>>> Roy,
>>>>>>>>>> I copied out the contents of dmesg about a week (~13 boots)
>>>>>> after the event, and the file date on the Xorg log is about the same
>>>>>> day. I have just copied it out this evening.
>>>>>>
>>>>>>>>>> The files are at:
>>>>>>>>>> dmesg:https://dpaste.com/4NKF4VDTX
>>>>>>>>>> Xorg.0.log:https://dpaste.com/FDSEQTEWN
>>>>>>>>>> I hope these are useful, and you can explain what I should look for.
>>>>>>>>>> Regards,
>>>>>>>>>> Alistair
>>>>>>>>>> On Wednesday, 21 September 2022 17:26:17 BST Roy wrote:
>>>>>>>>>>> Alistair,
>>>>>>>>>>> I'm a Gentoo user so I can't tell you the steps to fix it but
>>>>>> extracting
>>>>>>
>>>>>>>>>>> some debug information is distro agnostic.
>>>>>>>>>>> The X server and desktop are two separate bits. It's possible
>>>>>> for the X
>>>>>>
>>>>>>>>>>> Server to start and then for the desktop to fail. Since the desktop
>>>>>>>>>>> depends on a working X Server, the other way round is not possible.
>>>>>>>>>>> Step 1 is narrow the problem space.
>>>>>>>>>>>   From you failed update system post the output of the dmesg
>>>>>> command. If
>>>>>>
>>>>>>>>>>> you have the wgetpaste script, that will do all the hard work
>>>>>> and return
>>>>>>
>>>>>>>>>>> a URL that you can share, like
>>>>>>>>>>> $ wgetpaste -c dmesg
>>>>>>>>>>> Your paste can be seen here:http://dpaste.com/49PQNMDVM
>>>>>>>>>>> It will tell what your kernel did when it started. Errors or
>>>>>> omissions
>>>>>>
>>>>>>>>>>> here make give a few hints.
>>>>>>>>>>> Xorg also has a startup log file. If root runs Xorg, its at
>>>>>>>>>>> /var/log/Xorg.0.log. If a normal user runs Xorg its in
>>>>>>>>>>> /home/<username>/... but I don't know where. That will be good
>>>>>> to see too.
>>>>>>
>>>>>>>>>>> $ wgetpaste -s 0x0 /var/log/Xorg.0.log
>>>>>>>>>>> Your paste can be seen here:https://0x0.st/oVYj.0.log
>>>>>>>>>>> If you use a graphical login, you will have both. The graphical
>>>>>> login
>>>>>>
>>>>>>>>>>> runs Xorg as root, then when you log in, its run as the logged
>>>>>> in user.
>>>>>>
>>>>>>>>>>> The idea, is just to get a few pointers to whats broken.
>>>>>>>>>>> Those links are my real examples.
>>>>>>>>>>> Regards,
>>>>>>>>>>> Roy Bamford.
>>>>>>>>>>> On 21/09/2022 14:16, Alistair wrote:
>>>>>>>>>>>> Kevin,
>>>>>>>>>>>> I created a new user, shutdown, and then booted normally.
>>>>>>>>>>>> On logging in as the new user I tried to start the desktop,
>>>>>> with the same result.
>>>>>>
>>>>>>>>>>>> Last week I extracted the .xsession-errors file from the machine.
>>>>>>>>>>>> Unfortunately I had booted the box several times before then,
>>>>>> such that the first shutdown recorded shows errors and is different to
>>>>>> the sequence recorded on a good machine. It is different to all the
>>>>>> subsequent entries immediately prior to attempting to startx.
>>>>>>
>>>>>>>>>>>> I have yet to find anything meaingful (to me at least) when
>>>>>> searching for these messages.
>>>>>>
>>>>>>>>>>>> Last error before first startx attempt:
>>>>>>>>>>>> XIO: fatal IO error 4 (Interrupted system call) on X server ":0"
>>>>>>>>>>>> First error after first startx attempt:
>>>>>>>>>>>> QDBusConnection: error: could not send signal to service ""
>>>>>> path "//home/alistair/.kde/share/config/kdeglobals" interface
>>>>>> "org.kde.kconfig.notify" member "ConfigChanged": Invalid object path:
>>>>>> //home/alistair/.kde/share/config/kdeglobals
>>>>>>
>>>>>>>>>>>> The file '/home/alistair/.kde/share/config/kdeglobals' does
>>>>>> exist and is similar to a good one. Is the extra leading '/' an error?
>>>>>> and if so where is the file it is read from?
>>>>>>
>>>>>>>>>>>> Second error:
>>>>>>>>>>>> Qt: Session management error: networkIdsList argument is NULL
>>>>>>>>>>>> Andrew,
>>>>>>>>>>>> I cannot now remember if the upgrade included a kernel update,
>>>>>> but 'linux-generic' is one of 25 packages now being held back by
>>>>>> 'apt-get dist-upgrade'.
>>>>>>
>>>>>>>>>>>> This box has kernels 5.15.0-46, and 5.15.0-47 installed.
>>>>>>>>>>>> I booted into 5.15.0-46 and tried to start the desktop without
>>>>>> success.
>>>>>>
>>>>>>>>>>>> It boots to the console, and after logging in and then
>>>>>> starting X, I get the "plasma by kde"(?) black screen with a moveable
>>>>>> mouse pointer.
>>>>>>
>>>>>>>>>>>> Regards
>>>>>>>>>>>> Alistair
>>>>>>>>>>>> On Tuesday, 20 September 2022 22:40:10 BST Andrew Robinson wrote:
>>>>>>>>>>>>> Did the kernel upgrade at the same time? Did you try booting
>>>>>> into the older kernel and still get the problem? I have had something
>>>>>> similar and I needed to add a boot flag, correctly boot and then
>>>>>> update again but it’s tough to say without access. You’re able to
>>>>>> login? So you can see a GUI at a login?
>>>>>>
>>>>>>>>>>>>> I know this is not the time to say such a thing as it’s of no
>>>>>> help to you now but it’s also a great place to plug using Btrfs and
>>>>>> snapshots, you can have it automatically snapshot whenever you do a
>>>>>> package manager command, if you destroy it, you can boot to the old
>>>>>> snapshot. I recently did this while playing around with new wayland
>>>>>> settings, destroyed it, couldn’t boot, went back to the last snapshot
>>>>>> (taken automatically when I updated a wayland package) and I was back
>>>>>> in the game. (running Arch)
>>>>>>
>>>>>>>>>>>>> From: Kevin Davidson
>>>>>>>>>>>>> Sent: 20 September 2022 22:26
>>>>>>>>>>>>> To: Edinburgh Linux Users Group
>>>>>>>>>>>>> Subject: Re: [EdLUG] Kubuntu Desktop broken after 'apt-get
>>>>>> dist-upgrade'
>>>>>>
>>>>>>>>>>>>> I think I missed the beginning of this, but if I were
>>>>>> troubleshooting this, the first thing I'd try is to log in as a
>>>>>> different user and see if you still have the problem, Create another
>>>>>> temporary user account if you need to. If that works, then it’s not
>>>>>> the software installation that’s the problem, it’s the configuration.
>>>>>> Start looking through all the dot files and see what’s different
>>>>>> between the config that works and the one that doesn’t. The
>>>>>> .xsession-errors file ought to be a useful log of what went wrong.
>>>>>>
>>>>>>>>>>>>> Kevin
>>>>>>>>>>>>> On 20 Sep 2022, at 21:54,
>>>>>> Alistair<alistair at archerry.plus.com>  wrote:
>>>>>>
>>>>>>>>>>>>> I am still struggling to understand how to recover the
>>>>>> desktop in a refined manner.
>>>>>>
>>>>>>>>>>>>> Some of the sites I have found discuss re-installing the
>>>>>> display manager and/or window manager and/or the desktop environment,
>>>>>> but non of them mention kde plasma explicitly.
>>>>>>
>>>>>>>>>>>>> (I see the kde plasma logo and spinning cog wheel when I
>>>>>> issue the 'startx' command. It just stops at a black screen.)
>>>>>>
>>>>>>>>>>>>> Taking a "blunderbuss" approach, I have looked at simply
>>>>>> re-installing the kubuntu desktop.
>>>>>>
>>>>>>>>>>>>> Is this an acceptable approach? Or are there further
>>>>>> questions I can ask to establish the system state and what needs fixing?
>>>>>>
>>>>>>>>>>>>> Using the simulation option of apt-get, I can compare what
>>>>>> will be installed / re-installed on the system compared to another
>>>>>> working setup:
>>>>>>
>>>>>>>>>>>>> Broken T480:
>>>>>>>>>>>>> 'sudo apt-get install -s --reinstall kubuntu-desktop
>>>>>>>>>>>>> Reading package lists...
>>>>>>>>>>>>> Building dependency tree...
>>>>>>>>>>>>> Reading state information...
>>>>>>>>>>>>> The following additional packages will be installed:
>>>>>>>>>>>>> fonts-noto-unhinted haveged ibus-data kde-style-oxygen-qt5
>>>>>> kgamma5
>>>>>>
>>>>>>>>>>>>> kubuntu-settings-desktop libhavege2 libibus-1.0-5
>>>>>> liboxygenstyle5-5
>>>>>>
>>>>>>>>>>>>> liboxygenstyleconfig5-5 libscim8v5 libxcb-record0 plasma-desktop
>>>>>>>>>>>>> plasma-desktop-data plasma-thunderbolt plasma-widgets-addons sddm
>>>>>>>>>>>>> sddm-theme-breeze
>>>>>>>>>>>>> Suggested packages:
>>>>>>>>>>>>> ibus quota
>>>>>>>>>>>>> The following NEW packages will be installed
>>>>>>>>>>>>> fonts-noto-unhinted haveged ibus-data kde-style-oxygen-qt5
>>>>>> kgamma5
>>>>>>
>>>>>>>>>>>>> kubuntu-desktop kubuntu-settings-desktop libhavege2 libibus-1.0-5
>>>>>>>>>>>>> liboxygenstyle5-5 liboxygenstyleconfig5-5 libscim8v5
>>>>>> libxcb-record0
>>>>>>
>>>>>>>>>>>>> plasma-desktop plasma-desktop-data plasma-thunderbolt
>>>>>> plasma-widgets-addons
>>>>>>
>>>>>>>>>>>>> sddm sddm-theme-breeze
>>>>>>>>>>>>> 0 to upgrade, 19 to newly install, 0 to remove and 67 not to
>>>>>> upgrade.'
>>>>>>
>>>>>>>>>>>>> Working T450:
>>>>>>>>>>>>> 'sudo apt-get install -s --reinstall kubuntu-desktop
>>>>>>>>>>>>> Reading package lists... Done
>>>>>>>>>>>>> Building dependency tree
>>>>>>>>>>>>> Reading state information... Done
>>>>>>>>>>>>> 0 to upgrade, 0 to newly install, 1 reinstalled, 0 to remove
>>>>>> and 0 not to upgrade.
>>>>>>
>>>>>>>>>>>>> Inst kubuntu-desktop [1.398] (1.398 Ubuntu:20.04/focal [amd64])
>>>>>>>>>>>>> Conf kubuntu-desktop (1.398 Ubuntu:20.04/focal [amd64])'
>>>>>>>>>>>>> Regards,
>>>>>>>>>>>>> Alistair
>>>>>>>>>>>>> On Tuesday, 13 September 2022 21:10:09 BST Ken Robson wrote:
>>>>>>>>>>>>> OK my bad .Xauthority is right one.
>>>>>>>>>>>>> ⁣Ken
>>>>>>>>>>>>> Get BlueMail for Android ​
>>>>>>>>>>>>> On 13 Sep 2022, 19:11, at 19:11,
>>>>>> Alistair<alistair at archerry.plus.com>  wrote:
>>>>>>
>>>>>>>>>>>>> Ken,
>>>>>>>>>>>>> Thanks for the suggestion.
>>>>>>>>>>>>> I cannot find a file called .xauth.
>>>>>>>>>>>>> In the /home/alistair/ folder are the files:
>>>>>>>>>>>>> .Xauthority and .xsession-errors.
>>>>>>>>>>>>> I have 'rsync' these, and the output of 'dmesg' to another
>>>>>> box so that
>>>>>>
>>>>>>>>>>>>> I may view them more easily.
>>>>>>>>>>>>> What clues should I look for in these, or other files ?
>>>>>>>>>>>>> Regards,
>>>>>>>>>>>>> Alistair
>>>>>>>>>>>>> On Tuesday, 13 September 2022 18:15:05 BST Ken Robson wrote:
>>>>>>>>>>>>> I had a similar issue a few years ago.
>>>>>>>>>>>>> Try renaming the ~/.xauth file (sudo mv ~/.xauth ~/xauth) and
>>>>>> see if
>>>>>>
>>>>>>>>>>>>> x will load then.
>>>>>>>>>>>>> ⁣Ken
>>>>>>>>>>>>> Get BlueMail for Android ​
>>>>>>>>>>>>> On 13 Sep 2022, 17:56, at 17:56, Alistair
>>>>>>>>>>>>> <alistair at archerry.plus.com>  wrote:
>>>>>>>>>>>>> azmodie,
>>>>>>>>>>>>> Thanks for the prompt reply.
>>>>>>>>>>>>> "sudo dpkg --configure -a"
>>>>>>>>>>>>> Reports no actions.
>>>>>>>>>>>>> "sudo apt-get update", followed by
>>>>>>>>>>>>> "sudo apt-get dist-upgrade" or "sudo apt-get install -f":
>>>>>>>>>>>>> Both Reports
>>>>>>>>>>>>> "0 to upgrade 0 to newly install 0 to remove and 13 not to
>>>>>> upgrade"
>>>>>>
>>>>>>>>>>>>> So I need to dig deeper...
>>>>>>>>>>>>> Regards,
>>>>>>>>>>>>> Alistair
>>>>>>>>>>>>> On Tuesday, 13 September 2022 15:26:18 BST azmodie wrote:
>>>>>>>>>>>>> Hi Alistar,
>>>>>>>>>>>>> I'd try the usual recovery commands.
>>>>>>>>>>>>> sudo apt-get update
>>>>>>>>>>>>> sudo apt-get upgrade
>>>>>>>>>>>>> You may need to run
>>>>>>>>>>>>> sudo dpkg --configure -a
>>>>>>>>>>>>> or
>>>>>>>>>>>>> sudo apt-get install -f
>>>>>>>>>>>>> Regards,
>>>>>>>>>>>>> azmodie
>>>>>>>>>>>>> --
>>>>>>>>>>>>> http://gplus.to/azmodie
>>>>>>>>>>>>> "Since light travels faster than sound, people appear bright
>>>>>> until
>>>>>>
>>>>>>>>>>>>> you hear
>>>>>>>>>>>>> them speak." -- some bright spark
>>>>>>>>>>>>> On Tue, 13 Sept 2022 at 14:23, Alistair
>>>>>>>>>>>>> <alistair at archerry.plus.com>
>>>>>>>>>>>>> wrote:
>>>>>>>>>>>>> Last week I updated my Thinkpad T480 (running Kubuntu 22.04)
>>>>>>>>>>>>> using
>>>>>>>>>>>>> the
>>>>>>>>>>>>> command line. During the process the laptop crashed, and I am no
>>>>>>>>>>>>> longer
>>>>>>>>>>>>> able to access the desktop.
>>>>>>>>>>>>> I can login to console after boot up, but when I try to start
>>>>>>>>>>>>> the
>>>>>>>>>>>>> desktop
>>>>>>>>>>>>> with 'startx' I get a black screen. To get there it asks for a
>>>>>>>>>>>>> Kwallet
>>>>>>>>>>>>> password - (never previously used) and a wireless network
>>>>>>>>>>>>> password.
>>>>>>>>>>>>> The
>>>>>>>>>>>>> mouse pointer is a black arrow with a white outline, which does
>>>>>>>>>>>>> follow
>>>>>>>>>>>>> movements of the trackpad but does not respond to right or left
>>>>>>>>>>>>> buttons.
>>>>>>>>>>>>> I tried 'apt-get dist-upgrade' with the '-f' option to no
>>>>>>>>>>>>> avail.
>>>>>>>>>>>>> I think it stated that several packages would be held back
>>>>>>>>>>>>> before
>>>>>>>>>>>>> the
>>>>>>>>>>>>> update. I booted into recovery mode, ran 'fsck', 'clean',
>>>>>>>>>>>>> enabled
>>>>>>>>>>>>> network,
>>>>>>>>>>>>> then ran 'dpkg' which updated 41 packages, but still no proper
>>>>>>>>>>>>> desktop.
>>>>>>>>>>>>> How do I confirm the current state of the system, and then
>>>>>>>>>>>>> recover
>>>>>>>>>>>>> /
>>>>>>>>>>>>> repair the Kubuntu desktop?
>>>>>>>>>>>>> Regards,
>>>>>>>>>>>>> Alistair
>>>>>>>>>>>>> --
>>>>>>>>>>>>> EdLUG mailing list
>>>>>>>>>>>>> EdLUG at mailman.lug.org.uk
>>>>>>>>>>>>> https://lists.edlug.org.uk/mailman/listinfo/edlug
>
>
>
>

More information about the EdLUG mailing list