Summary: | systemd: exit user session broken | ||
---|---|---|---|
Product: | systemd | Reporter: | Oleksii Shevchuk <public.avatar> |
Component: | general | Assignee: | systemd-bugs |
Status: | RESOLVED FIXED | QA Contact: | systemd-bugs |
Severity: | blocker | ||
Priority: | medium | ||
Version: | unspecified | ||
Hardware: | Other | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Looks like system exit broken tor Hmm, I think only recent util-linux has a kill tool that can handle RTSIG as signal correctly. We could specify the numeric signal number instead of the symbolic name on the command line which would work in all cases. Numeric argument should be ok, but units must be generated in that case. There is tiny possibility, that RTMAX-6 code representation will be different for different versions of headers. (In reply to comment #2) > Hmm, I think only recent util-linux has a kill tool that can handle RTSIG as > signal correctly. > > We could specify the numeric signal number instead of the symbolic name on > the command line which would work in all cases. Oh.. > equery b /bin/kill * Searching for /bin/kill ... sys-process/procps-3.3.3 (/bin/kill) Looks like numeric args to kill is the One Right Way :) Fixed in 195. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.
As for 20f59e42cd64c0c7fd9386b0ca93842c6cdbc387 systemctl --user exit broken: ExecStart={ path=/bin/kill ; argv[]=/bin/kill -s KILL $MANAGERPID ; ignore_errors=no ; start_time=[n/a] ; stop_time=[n/a] ; pid=0 ; code=(null) ; status=0/0 } > strace -e kill -f /bin/kill -s RTMAX-6 $(pidof systemd) kill(2750, -1) = -1 EINVAL (Invalid argument) kill(1, -1) = -1 EINVAL (Invalid argument) +++ exited with 1 +++