On May 21, Marc Lehmann <debian-report...@plan9.de> wrote: > this unexpectedly not only killed inetd, but also all remote shells (which > means the cp that was labouring for hours to copy a big faile form a disk > was also interrupted...) and other services started via inetd. This is a > serious regression compared to pre-systemd behaviour. It is different... I am not sure yet about which semantic is better as the default. Can you better describe your setup? Which kind of shells were you spawning from inetd?
If you want to restore the old behaviour right now then you can use: mkdir /etc/systemd/system/cron.service.d/ echo KillMode=process > /etc/systemd/system/cron.service.d/local.conf -- ciao, Marco
signature.asc
Description: PGP signature
