[kde] [Bug 477327] startplasma-x11 fails - no usage possible - no log - no sign of debugging possible

J bugzilla_noreply at kde.org
Wed Nov 22 10:15:16 GMT 2023


https://bugs.kde.org/show_bug.cgi?id=477327

--- Comment #2 from J <jojo at automatix.de> ---
A clean .config works, but is not an option, due a huge amount of kwallet
(gnupg) keys and forms data for chromium and saving/restoring them is
nessesary.

I  tried out to remove anything not nessesary out of the way in .cache (deleted
completely) .config (deleted partially, execpt license files of commercial
licenses, kwallet files, chromium (arch version), gnupg (linked from
$HOME/.gnupg -> .config/gnupg.

startplasma-x11 was triggered by the sddm standard session startup.
Unfortunately no logging in systemd journal or dmesg or .xsession-errors are
done.
Changing the Xsession start profile to "xterm" and starting "strace
/usr/bin/startplasma-x11" in there gives no clue. I'll try to  catch the strace
out, Last line is IMHO "brk" line, repeating endless, 
and console top shows up no forks, but 100% cpu and expotentially raising
memory usage, each second 200kiB RSS segement.

User auth is done by ldap/nslcd/pam_ldap from central ldap server.

This Filesystem setup was nessesary, due performance prooblems with kde/plasma
users, if a nfs-server home was used:
server:/home        734002176   198294528  532041728   28% /home
/dev/sdb3          1447541220  1324448464   84672400   94% /home/jojo/.cache
/dev/sdb3          1447541220  1324448464   84672400   94% /home/jojo/.config
/dev/sdb3          1447541220  1324448464   84672400   94% /home/jojo/.fonts
/dev/sdb3          1447541220  1324448464   84672400   94% /home/jojo/.local
/dev/sdb3          1447541220  1324448464   84672400   94%
/home/jojo/.thunderbird
tmpfs                13184980         156   13184824    1% /run/user/10000

i.E. /home/jojo/.*  is mounted as btrfs subvolume, to keep buggy IO-intensive
tasks on the local machine.

We are a small company, having 20 employee logging in each morning at 8am. 
The startup of plasma sessions was without "local mounted" helps a DDOS against
the NFS Server.

standby, I try to grep more logs

-- 
You are receiving this mail because:
You are the assignee for the bug.


More information about the Unassigned-bugs mailing list