home.social

#timjangrumpsoncomputers — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #timjangrumpsoncomputers, aggregated by home.social.

  1. >The following OpenPGP key (0x74C35BC8) is about to be removed:
    Reason : Expired on 2026-02-11 11:44:18

    It will be successfully removed.

    Tomorrow it will be back.

    Why I think so? Because it has returned, and I have re-deleted it, practically every day since February 11.

    #fedora #timjanGrumpsOnComputers

  2. Joplin update woes:

    I don't use Joplin on the desktop as much as on the phone, which means that I had neglected updating the desktop version (AppImage; yes, Linux) for some time.

    Today I finally did open it, and saw several notes being un-encryptable. Okay, time to update, I guess.

    Only problem turned out to be that version 3.5.11 and 3.5.13 (the two I tried) don't load properly: the app only shows a grey rectangle with a half menubar that doesn't respond to button presses.

    So I tried to get back to version 3.4.x. Nope, can't do, the profile is too new. So version 3.5.x did do something: update the profile and prevent me from opening an older version.

    What finally solved it was to restore $HOME/.config/Joplin and $HOME/.config/joplin-desktop from backup, and run version 3.4.10. The app opens, and all notes decrypt properly.

    I still don't dare trying to run 3.5.x again - it would probably require another restore from backup in order to get back to 3.4.x again.

    As an aside, I did learn that backintime backups of softlinks can get you to the live version of a file rather than the backed up version. Which is why my 2 week old backup looked like it contained files that had been updated 43 minutes ago - those files weren't the backed up files, but the live ones.

    #TimjanGrumpsOnComputers #Joplin

  3. "wINdoWs jUst wOrkS"

    Windows suddenly can't find the external monitor, and the settings dialog freezes.
    #TimjanGrumpsOnComputers

  4. CW: signal desktop grumping.

    But of course Linux = Debian.
    #TimjanGrumpsOnComputers

  5. CW: System-upgrade

    #TimjanGrumpsOnComputers

    One hour later.

    "Upgrading system. Do not turn ff your computer. 0% complete"

  6. CW: Bellyaching about Linux

    Wee, I got my main system to hibernate and wake up again...

    Except that the network interface didn't come up.

    And it doesn't want to poweroff or reboot: it just sits there doing nothing.

    #TimjanGrumpsOnComputers

  7. CW: no sleep in Fedora (sometimes)

    #TimjanGrumpsOnComputers

    Yet again, #fedora freezes hard instead of going to sleep.
    Any recommendations on how to figure out what is causing this?

    Comparing two journalctl printouts, for a successful and unsuccessful sleep respectively, and it appears to fail to reach this stage:

    `
    Sep 03 09:22:44 <hostname> kernel: sd 1:0:0:0: [sdb] Synchronizing SCSI cache
    Sep 03 09:22:44 <hostname> kernel: sd 0:0:0:0: [sda] Synchronizing SCSI cache
    Sep 03 09:22:44 <hostname> kernel: ata1.00: Entering standby power mode
    Sep 03 09:22:44 <hostname> kernel: sd 2:0:0:0: [sdc] Synchronizing SCSI cache
    Sep 03 09:22:44 <hostname> kernel: ata3.00: Entering standby power mode
    Sep 03 09:22:44 <hostname> kernel: ata2.00: Entering standby power mode
    `

    (Forgot to post this after composing...)

  8. I shouldn't only write about computers when they don't work.
    This morning, for example, the upgrade from #fedora39 to #fedora40 went smoothly and the system could boot afterwards.

    #TimjanGrumpsOnComputers (<-- not today)

  9. CW: Bellyaching about computers (Wayland (KDE?))

    XWayland crashes twice during a one hour Skype call.

    Someone reported a very similar bug 6 months ago; another half dozen people made "me too" reports.

    If someone is working on it, that's not visible from the bug report.

    Giving up on Wayland, so trying to log out to switch to Xorg. After clicking on "leave", everything froze. Mouse, keyboard, status bar clock... hard reset, sigh.

    #TimjanGrumpsOnComputers

  10. CW: Bellyaching about linux (gentoo)

    Okay, that's enough of Gentoo for me.

    No updates have succeeded for the last month, and I just wasted a full Saturday on trying to coax it to continue.
    I'm still stuck on compilation failures:

    #timjangrumpsoncomputers

  11. CW: Bellyaching on Li... Windows.

    #TimjanGrumpsOnComputers

    "No Timjan, you may not delete this folder. To do so, you need permission from Timjan. [Skip] [Cancel]".

  12. CW: Bellyaching about linux (gentoo)

    Alright, that's it.

    Either I'll have to upgrade the desktop, or I'll have to leave Gentoo.

    It's been compiling more or less constantly for the last week!
    Yes, in many cases, it's had to restart large compilations, because they tend to crash after 5 hours .

    #timjangrumpsoncomputers

  13. CW: Bellyaching about Linux (fedora)

    And the next step was to upgrade the *desktop* from 36 to #Fedora38. (36 was a fresh install).

    So far it's stuck in a loop right after 'system-upgrade reboot':
    "systemd[1]: looping too fast. Throttling execution a little"

    Second image: how far things has progressed after about 9 hours.

    The first image is from after I rebooted to let it try again... It doesn't seem to go any further just because the reboot.

    #TimjanGrumpsOnComputers
    #TooDumbForLinux

  14. CW: Third party apps versus apps from repos.

    #TimjanGrumpsOnComputers
    Reminder to self to not use apps that saves data in binary formats.

    =====

    Apparently I had two versions of app 'foo' installed - an older from the repos, a newer downloaded from the website ("portable").
    (Did I do that when the repo version was unavailable / broken a while back?)

  15. CW: Bellyaching on Linux (gnome)

    #TimjanGrumpsOnComputers

    I'm sorry, but I still don't get the #gnome-derived file interaction dialogs that certain applications insist on, despite being in a #KDE environment.

    If I don't want to open from/save to the default location that the dialog presents, my mental model of what I'm doing comes crashing down until I rearrange myself.

  16. CW: Bellyaching on Linux (snap)

    #TimjanGrumpsOnComputers

    What the ... snap?

    `snap refresh` ?

    It's enough to try to remember which tools use 'update' and which use 'upgrade' - and you come here with 'refresh'?

  17. CW: Bellyaching about Linux (gentoo, qtwebengine)

    F*ck, a reboot didn't fix the pam spew.

    And now sddm doesn't start properly either. So no graphics beyond one mouse pointer on a black screen.

    F*ck.

    I don't want to reinstall Gentoo, too.

    Well, I guess I really have to get on that fedora reinstallation, then

    #TimjanGrumpsOnComputers

  18. CW: Bellyaching about Linux (gentoo, qtwebengine)

    That was another episode of #TimjanGrumpsOnComputers

  19. CW: Bellyaching about Linux

    #TimjanGrumpsOnComputers

    "Oh, you want to accomplish <blah>? The nice, powerful piece of software FooBaz will do that for you"

    $ sudo emerge --ask FooBaz
    (FooBaz is installed)
    $ FooBaz
    bash: FooBaz: command not found
    $ foobaz
    bash: foobaz: command not found
    $ man foobaz
    No manual entry for foobaz

    $ ddg "foobaz get started"
    "Just double click on the icon"

  20. CW: Bellyaching about Linux

    #TimjanGrumpsOnComputers

    Trying to wake up desktop from sleep.

    Mouse works, keyboard does not. Can't unlock.

    Ssh in from phone. Tries to suspend again. Wake up a second time.

    Bright white screen, no indication of mouse or keyboard.

    Duck this, I'm too tired to deal with this ship.

  21. CW: bellyaching about linux (gentoo)

    #TimjanGrumpsOnComputers

    Oh, those are USE flags, not packages

    /me is too dumb for Gentoo

  22. CW: Bellyaching about Linux (kde)

    #TimjanGrumpsOnComputers

    Oh, there's a workaround:

    don't update the flatbpaks, only the themes

    only select ONE theme at a time (else it'll crash)

    some will fail for "not finding <some file>"... and crash.

    After certain crashes, it'll be unable to start up all way again. I'll probably see it again after the next reboot and/or gentoo reinstallation.

    :shrug:

  23. CW: Bellyaching about Linux (kde)

    #TimjanGrumpsOnComputers

    But I guess I should have known what I should have installed to get the backtrace for the tool to generate the backtrace for what broke in the first place.

  24. CW: Bellyaching about Linux (kde)

    #TimjanGrumpsOnComputers

    ...
    Okay, this is repeatable, yadda yadda.
    "Fetching backtrace"..
    "The debugger has quit unexpectedly".
    "The backtrace could not be generated".

    Uhm, okay?

    "This report does not contain enough information for the developers".

    "Finish".

    So no report generated, this time either. Good thing I can update anything important through CLI, then.

  25. CW: Bellyaching about Linux (kde)

    #TimjanGrumpsOnComputers

    #KDE wants to use its tool ("Discover") to update the system. Today, I let it.
    It found the flatpaks, and the various themes I've installed.
    Okay, let's install them...
    First, it takes 5 minutes to "fetch updates" (why not already in the background?)
    Installing...
    "We're sorry, discover closed unexpectedly", and please report the error. Sure, let's report the bug.
    "The assistant will analyze the crash information and guide you..."

    ...

  26. CW: Bellyaching about Linux

    #TimjanGrumpsOnComputers

    Went back to AwesomeWM.
    Tried to suspend to disk ('loginctl suspend'); didn't notice until next morning that it didn't do anything.
    Tried to suspend again. 4 attempts; some did nothing, some 'suspended' for about 5 seconds after which the system woke up by itself. 5th attempt successful. Or?

    Now, trying to wake up the system. Power indicator is on but the system doesn't respond to any inputs. (Keyboard, mouse, ssh).

    Hard power cycle, then.

  27. CW: bellyaching about linux (gentoo)

    #TimjanGrumpsOnComputers
    > The following REQUIRED_USE flag constraints are unsatisfied:
    > foo? ( any-of ( bar baz ) )

    > no ebuilds to satisfy "bar"
    > no ebuilds to satisfy "baz"

    Okay, what's next?