Upgrades between yesterday and today

Something really strange is happening, something failed when the system tried to create a new snapshot. Now sudo is gone, can't open anything, looks like the snapshot I was working in got deleted or something.

Just trying to open a new terminal window gives me this error in a KDE popup.

'Unable to create io-slave. Can not create socket for launching io-slave for protocol "applications"'

With no logs I can't imagine what happened.
If sudo is not working, use su to login as root and do a complete update.


Ok, I'm finally back. I went through all the snapshots and finally got one that would work from the 16th. I couldn't get to any logs earlier.

I'm pretty sure what I need to do first is go into timeshift and restore this snapshot, but what logs should I paste in here first?

╰─λ sudo journalctl -b -p3
[sudo] password for dale:     
-- Journal begins at Tue 2021-02-02 10:48:55 CST, ends at Thu 2021-02-18 21:06:10 CST. --
Feb 18 20:53:22 dalegaruda systemd-udevd[312]: could not read from '/sys/module/pcc_cpufreq/initstate': No such device
Feb 18 20:53:32 dalegaruda kernel: 
Feb 18 20:53:33 dalegaruda systemd-tmpfiles[489]: Failed to write file "/sys/module/pcie_aspm/parameters/policy": Operation not permitted
Feb 18 20:54:01 dalegaruda nmbd[924]: [2021/02/18 20:54:01.193390,  0] ../../lib/util/become_daemon.c:147(daemon_status)
Feb 18 20:54:01 dalegaruda nmbd[924]:   daemon_status: daemon 'nmbd' : No local IPv4 non-loopback interfaces available, waiting for interface ...
Feb 18 20:54:01 dalegaruda nmbd[924]: [2021/02/18 20:54:01.217320,  0] ../../source3/nmbd/nmbd_subnetdb.c:253(create_subnets)
Feb 18 20:54:01 dalegaruda nmbd[924]:   NOTE: NetBIOS name resolution is not supported for Internet Protocol Version 6 (IPv6).
Feb 18 20:54:20 dalegaruda sddm-helper[1146]: gkr-pam: unable to locate daemon control file
Feb 18 20:54:37 dalegaruda kwin_x11[1389]: kwin_core: Compositing is not possible
Feb 18 20:55:06 dalegaruda nmbd[924]: [2021/02/18 20:55:06.220424,  0] ../../lib/util/become_daemon.c:135(daemon_ready)
Feb 18 20:55:06 dalegaruda nmbd[924]:   daemon_ready: daemon 'nmbd' finished starting up and ready to serve connections
Feb 18 20:55:07 dalegaruda smbd[2007]: [2021/02/18 20:55:07.387958,  0] ../../lib/util/become_daemon.c:135(daemon_ready)
Feb 18 20:55:07 dalegaruda smbd[2007]:   daemon_ready: daemon 'smbd' finished starting up and ready to serve connections
Feb 18 20:55:07 dalegaruda bluetoothd[498]: src/profile.c:record_cb() Unable to get Headset Voice gateway SDP record: Host is down

All the versions are back to what they were two days ago.

╰─λ sudo pacman -Qs kwin
local/kwin-effects-sliding-notifications 1.0.3-1
    Sliding animation for notification windows
local/kwin-lowlatency 5.20.5-1 (plasma)
    the compositor, with added stutter/latency reductions
local/kwin-scripts-forceblur 0.5-1
    Force-enable blur effect to user-specified windows
local/kwindowsystem 5.79.0-1 (kf5)
    Access to the windowing system

It's getting funny. :grinning_face_with_smiling_eyes:

1 Like

I think I don't understand enough about how timeshift works to not cause problems for myself.

Here are some scenarios.

Update or add software, pamac says need a reboot for changes to take effect, installs more software before doing reboot. Reboot and now machine hangs as described in earlier posts.

Goes back to previous snapshot by selecting it from grub menu, previous one hangs too.

Goes back to the other previous version, now that one works. Does not go into timeshift and "restore" the snapshot because I foolishly thought that by going back to the snapshot in grub I was restoring, didn't look closely enough at the directions.

Does new pacmac -Syu still without having restored properly. pacman -Qs shows kwin being incorrect version.

Does pacman -Syu kwin which updates that version of kwin-lowlatency is now correct 5.21-1.

Reboot takes me back to being hung as previously described, probably because I fixed something in a snapsnot that was never restored properly.

I finally realize I was doing something wrong and not really restoring, so I go back to the snapshot that worked before, load it up, and open timeshift and restore. I still am too thick headed to realize I need to reboot after the restore and before doing anything else, and also even worse, while timeshift is still open I try to do pacman -Sy extra/kwin which makes the new pacman command fail, but the automated snapshot taker has already blown up since the old snapshot I restored was missing and because there was another instance of timeshift running.

Now, no windows can be opened as the snapshot I was logged into is actually deleted so there are no executables remaining, I can't open anything, but I still have a browser window open on this page so I am able to post about something strange going on. Anything I try to open gets an error.

Try reboot to default, which theoretically would have been fixed if I hadn't run the extra pacman command at the end, of course it couldn't even be found by grub since it got deleted.

Try each snapshot in turn, all except for the very oldest from 2/16 either can't be found at all (3 others) or has the original hanging problem after entering password to log in and while being asked for wallet password as previously described.

Now I'm sitting on the last snapshot that works, thinking I need to go into timeshift and restore this snapshot, then reboot, then start over with pacman -Syu. What I'm afraid of is if I do the restore and reboot that I'll be getting hung.

Any last minute advice on changes I should make to my plan would be appreciated. I will gladly listen to your laughter at my noobish mistakes too, because honestly I think it's pretty darn funny how badly I messed up my system with my lack of understanding.

1 Like

I am not laughing :rofl:

Do one task at a time, when it is about your system administration.

If you have just used a snapshot from grub menu, use timeshift to restore the same snapshot and reboot. At the grub menu, choose the default entry (no snapshot). This should lead you to the same place.
Do your update as normal.

About kwin-lowlatency, don't worry. The issue is (kind-of) fixed. The newest package is a dummy package, which just installs the normal kwin. Versions prove this.


Success! Thanks for all your help @petsam .


This thread is a nice example on how troubleshooting should be done. :grin: :clap:


This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.