[systemd-devel] Logind lingering processes despite KillUserProcess=yes
Tyson Whitehead
twhitehead at gmail.com
Tue Dec 17 09:20:17 PST 2013
I've got a system running Fedora 20 with systemd 208-9.
It doesn't seem to be honouring the KillUserProcess=yes setting in logind.conf.
$ grep KillUserProcesses /etc/systemd/logind.conf
#KillUserProcesses=no
KillUserProcesses=yes
$ loginctl
SESSION UID USER SEAT
24 0 root
c61 997 lightdm seat0
34 130018 tyson
3 sessions listed.
$ loginctl show-session 34
Id=34
Timestamp=Tue 2013-12-17 11:50:26 EST
TimestampMonotonic=66431256528
VTNr=0
Remote=yes
RemoteHost=whitehead.beowulf.uwo.ca
Service=sshd
Scope=session-34.scope
Leader=11543
Audit=34
Type=tty
Class=user
Active=yes
State=closing
IdleHint=no
IdleSinceHint=0
IdleSinceHintMonotonic=0
Name=tyson
$ loginctl session-status 34
34 - tyson (130018)
Since: Tue 2013-12-17 11:50:26 EST; 13min ago
Leader: 11543
Remote: whitehead.beowulf.uwo.ca
Service: sshd; type tty; class user
State: closing
Unit: session-34.scope
├─11551 ssh -x -a -oClearAllForwardings=yes -oNoneSwitch=yes -oNoneEnabled=yes -oNumberOfPasswordPrompts=1 ...
├─11552 sshfs tyson at exporter.sharcnet.ca:/home/tyson /home/tyson -o reconnect,nonempty,default_permissions,...
├─11559 ssh -x -a -oClearAllForwardings=yes -oNoneSwitch=yes -oNoneEnabled=yes -oNumberOfPasswordPrompts=1 ...
└─11560 sshfs tyson at exporter.sharcnet.ca:/work/tyson /work/tyson -o reconnect,nonempty,default_permissions,...
Shouldn't it be sending a kill signal to these process?
$ killall -u tyson
$ loginctl
SESSION UID USER SEAT
24 0 root
c61 997 lightdm seat0
Would this be related to (i.e., already fixed in git)
https://bugs.freedesktop.org/show_bug.cgi?id=71092
Thanks! -Tyson
More information about the systemd-devel
mailing list