+* calendar time support in timer, iCalendar semantics for the timer stuff (RFC2445)
-- mounts might take time, need to be executed asynchronously in a forked process?
+* complete dbus exposure
-- need gc for active jobs that nothing cares for
+* implicitly import "defaults" settings file into all types
-- need gc for units that are not referenced anymore
+* add #ifdefs for non-sysv builds
-- refreshing of units (i.e. reload config files)
+* reinvestigate random seed, hwclock
-- dbusification
+* "disabled" load state?
-- cgroupsification
+* ability to kill services? i.e. in contrast to stopping them, go directly
+ into killing mode?
-- udev hookup
+* port over to LISTEN_FDS/LISTEN_PID:
-- snapshot system
+ - uuidd DONE
+ - nscd DONE
+ - dbus DONE
+ - rsyslog DONE
+ - rpcbind (/var/run/rpcbind.sock!) DONE
+ - cups DONE
+ - avahi-daemon (/var/run/avahi-daemon/socket) DONE
+ - ssh CLASSIC
+ - postfix, saslauthd
+ - apache/samba
+ - libvirtd (/var/run/libvirt/libvirt-sock-ro)
+ - bluetoothd (/var/run/sdp! @/org/bluez/audio!)
+ - distccd
-- hookup with battery status, suspend/resume
+* write utmp record a la upstart for processes
-- add a job mode which shuts everything down but what is requested
+* follow property change dbus spec
-- implement priv dropping
+* selinux
-- allow to disable priv dropping for hook scripts from fragment
+* systemctl status $PID, systemctl stop $PID!
-- implement timer
+* sulogin in den single user mode, mit plymouth --hide davor
-- implement mount/automount
+External:
-- more process attributes: chroot, cpu affinity, scheduling
+* patch /etc/init.d/functions with:
-- create session/pgroup for child processes
+ if [ $PPID -ne 1 && mountpoint /cgroup/systemd ] ; then echo "You suck!" ; fi
-- fs namespaces
+Regularly:
-- templating/instances
+* look for close() vs. close_nointr() vs. close_nointr_nofail()
-- verify fragment data after loading: refuse cycles on yourself, service units contradicting, more than one Start executable, ...
+* check for strerror(r) instead of strerror(-r)
-- automatically delete stale unix sockets
+* Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
-- .socket needs to be notified not only by .service state changes, but also unsuccessful start jobs
-
-- we probably cannot use glibc's syslog() for logging, since it
- presumably uses the logging socket in blocking mode which might
- trigger a deadlock if syslog does not process the socket anymore
- (maybe because it is restarted) and the socket buffer is full.
+* %m in printf() instead of strerror();