* if a service fails too often, make the service enter failed mode, and the socket, too.
-* bash completion a la gdbus
-
* systemctl list-jobs - show dependencies
* accountsservice is borked
* start getty 2-6 only when actual vt switch happens
(same model as socket on-demand activation)
-* ask-password: find way to disable wall agent where tty agent is running
-
* readahead: btrfs/LVM SSD detection
-* document locale.conf, vconsole.conf and possibly the tempfiles.d and modules-load.d mechanism.
+* man pages for: locale.conf, vconsole.conf modules-load.d
* when processes remain in a service even though the start command failed enter active
* parse early boot time env var from dracut RD_TIMEOUT, drop RD_xxx
-* cryptsetup
-
* add seperate man page for [Install] settings
* only add quotacheck deps to .mount units which mention grpquota/usrquota in the mount flags
-* we now order plymouth after udev-trigger. Can we optimize this?
+* systemctl condrestart should return 0 if service isn't running
-* drop tmp.mount
+* tmpfiles: allow specification of .conf files on cmdline
-* systemctl condrestart should return 0 if service isn't running
+* allow runtime changing of log level and target
+
+* in the PAM module rely on loginuid to figure out XDG_RUNTIME_DIR
+
+Pre v12:
* fsck-root.service/start gets queued twice
+* pull in 'following' units in transactions
+
+* fix hotplug transactions
+
+* plymouth agent start loop
+
External:
* patch kernel for xattr support in /dev, /proc/, /sys and /sys/fs/cgroup.