> % systemctl status syslog-ng.service; dpkg-reconfigure syslog-ng-core > syslog-ng.service - System Logger Daemon > Loaded: loaded (/lib/systemd/system/syslog-ng.service; enabled) > Active: active (running) since Tue 2014-08-12 20:42:04 CEST; 1s ago > Docs: man:syslog-ng(8) > Main PID: 24163 (syslog-ng) > CGroup: /system.slice/syslog-ng.service > └─24163 /usr/sbin/syslog-ng -F > > Aug 12 20:42:04 mercury2 systemd[1]: Started System Logger Daemon. > Aug 12 20:42:04 mercury2 systemd[1]: Started System Logger Daemon. > + [ configure = triggered ] > + dpkg-trigger register-syslog-ng-plugin > + deb-systemd-helper unmask syslog-ng.service > + deb-systemd-helper --quiet was-enabled syslog-ng.service > + deb-systemd-helper enable syslog-ng.service > + [ -x /etc/init.d/syslog-ng ] > + update-rc.d syslog-ng defaults 10 90 > + exit 0 > Processing triggers for syslog-ng-core (3.5.6-1) ... > + [ triggered = triggered ] > + invoke-rc.d syslog-ng stop > Job for syslog-ng.service canceled. > invoke-rc.d: initscript syslog-ng, action "stop" failed. > + exit 1 > dpkg: error processing package syslog-ng-core (--configure): > subprocess installed post-installation script returned error exit status 1 > Errors were encountered while processing: > syslog-ng-core > > I was suspecting a systemd rate limit at first, but the upgrade that failed > today didn't include any other packages shipping anything systemd related. See > the attached history.log for the packages involved in the failing upgrade.
Turns out this is a known issue with systemd and socket activated services (such as syslog-ng), see #736258 and #751744. However, we do not use the recommended --restart-after-upgrade mechanism, and I'm unsure whether that would fix the issue. Nevertheless, I have a far better idea of what is happening, and will try to reproduce the issue locally. -- |8] -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org