1 * enabling an instance unit creates pointless link, and
2 the unit will be started with getty@getty.service:
3 $ systemctl enable getty@.service
4 ln -s '/usr/lib/systemd/system/getty@.service' '/etc/systemd/system/getty.target.wants/getty@.service'
6 * check systemd-tmpfiles for selinux context hookup for mknod(), symlink() and similar
8 * swap units that are activated by one name but shown in the kernel under another are semi-broken
10 * Dangling symlinks of .automount unit files in .wants/ directories, set up
11 automount points even when the original .automount file did not exist
12 anymore. Only the .mount unit was still around.
14 * make polkit checks async
16 * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point.
18 * ellipsize_mem must take into account multi-byte unicode characters, and
19 - make the resulting line the requested number of *characters*, not *bytes*,
20 - avoid truncuating multi-byte sequences in the middle.
22 * When we detect invalid UTF-8, we cant't use it in an error message:
23 log...("Path is not UTF-8 clean, ignoring assignment: %s", rvalue);
25 * shorten the message to sane length:
27 Cannot add dependency job for unit display-manager.service, ignoring: Unit display-manager.service failed to load: No such file or directory. See system logs and 'systemctl status display-manager.service' for details.
31 * external: ps should gain colums for slice and machine
34 - localectl: support new converted x11→console keymaps
36 * when installing fedora with yum --installroot /var/run is a directory, not a symlink
37 https://bugzilla.redhat.com/show_bug.cgi?id=975864
39 CGroup Rework Completion:
41 * introduce "mainpid" for scopes (or maybe not?)
43 * implement system-wide DefaultCPUAccounting=1 switch (and similar for blockio, memory?)
45 * implement per-slice CPUFairScheduling=1 switch
47 * handle jointly mounted controllers correctly
49 * make BlockIODeviceWeight=, BlockIODeviceBandwidth= runtime settable
51 * introduce high-level settings for RT budget, swappiness
53 * wiki: document new bus APIs of PID 1 (transient units, Reloading signal)
57 * libdsystemd-bus should expose utf8 validation calls
59 * When using "systemd status" on a slice unit also show all messages
60 matching _SYSTEMD_SLICE= not just _SYSTEMD_UNIT=
62 * always set memory.user_hierarchy for all cgroups we create
64 * After coming back from hibernation reset hibernation swap partition
66 * mounts: do not test each mount unit against each other mount unit to
67 determine prefixes. Instead generated list of all prefixes and
68 interate through that to bring down complexity from O(n^2) to O(n)
71 * If we try to find a unit via a dangling symlink generate a clean
72 error. Currently we just ignore it and read the unit from the search
75 * When a Type=forking service fails and needed another service that
76 service is not cleaned up again when it has StopWhenUnneeded=yes
77 http://lists.freedesktop.org/archives/systemd-devel/2013-July/012141.html
79 * backlight: properly handle multiple backlight devices for the same
80 hardware: at shutdown we should only save the backlight setting for
81 the "best" way to access the backlight. Strategy should be: at
82 shutdown, ignore all backlights that are connected to a non-eDP or
83 non-LVDS port, and then prefer the firmware device over platform
84 device over raw device per-PCI card. Delete all old data. At boot
85 simply apply whatever data we find. Also see
86 http://cgit.freedesktop.org/libbacklight/tree/libbacklight.c#n194
88 * rfkill: save/restore soft rfkill status across reboots
90 * refuse boot if /etc/os-release is missing or /etc/machine-id cannot be set up
92 * ensure scope units may be started only a single time
94 * document that in instead of FsckPassNo= people should just add a manual dep to systemd-fsck@.service to their mount units.
96 * better error message if you run systemctl without systemd running
98 * systemctl status output should should include list of triggering units and their status
100 * for transient units, instead of writing out drop-ins for all properties consider serializing them in the normal serialization stream
102 * logind: when logging out, remove user-owned sysv and posix IPC objects
104 * session scopes/user unit: add RequiresMountsFor for the home directory of the user
106 * add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
108 * journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
110 * given that logind/machined now let PID 1 do all nasty work we can
111 probably reduce the capability set they retain substantially.
113 * btrfs raid assembly: some .device jobs stay stuck in the queue
115 * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
117 * Fedora: post FPC ticket to move add %tmpfiles_create to the packaging guidelines
119 * make sure gdm doesn't use multi-user-x but the new default X configuration file, and then remove multi-user-x from systemd
121 * when parsing calendar timestamps support the UTC timezone (even if we won't support arbitrary timezone specs, support UTC itself certainly makes sense), also support syntaxes such as +0200
123 * when a kernel driver logs in a tight loop we should ratelimit that too.
125 * "systemctl disable" of a unit instance removes all symlinks, should
126 only remove the instance symlink (systemctl disable of a template
127 unit however should remove them all).
129 * journald: optionally, log debug messages to /run but everything else to /var
131 * systemctl list-unit-files should list generated files (and probably with a new state "generated" for them, or so)
133 * journald: when we drop syslog messages because the syslog socket is
134 full, make sure to write how many messages are lost as first thing
135 to syslog when it works again.
137 * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
139 * load .d/*.conf dropins for device units
141 * add a fixed dbus path for "my own unit", "my own session", ... to PID1, logind, ...
143 * service_coldplug() appears to reinstall the wrong stop timeout watch?
145 * transient units: allow creating auxiliary units with the same call
147 * how to reset dynamically changed attributes sanely?
149 * when reloading configuration, apply new cgroup configuration
151 * journald: make sure ratelimit is actually really per-service with the new cgroup changes
153 * gparted needs to disable auto-activation of mount units somehow, or
154 maybe we should stop doing auto-activation of this after boot
155 entirely. https://bugzilla.gnome.org/show_bug.cgi?id=701676
156 Maybe take a BSD lock at the disk device node and teach udev to
157 check for that and suppress event handling.
159 * when recursively showing the cgroup hierarchy, optionally also show
160 the hierarchies of child processes
162 * document logic of auto/noauto and fail/nofail in fstab in systemd.mount or systemd-fstab-generator man page
164 * we should properly escape ssh hostnames we add into dbus server strings
166 * something pulls in pcre as so dep into our daemons such as hostnamed.
168 * document systemd-journal-flush.service properly
170 * change systemd-journal-flush into a service that stays around during
171 boot, and causes the journal to be moved back to /run on shutdown,
172 so that we don't keep /var busy. This needs to happen synchronously,
173 hence doing this via signals is not going to work.
175 * allow implementation of InaccessibleDirectories=/ plus
176 ReadOnlyDirectories=... for whitelisting files for a service.
179 - default policy (allow uid == 0 and our own uid)
180 - enforce alignment of pointers passed in
181 - when kdbus doesn't take our message without memfds, try again with memfds
182 - implement translator service
183 - port systemd to new library
184 - implement busname unit type in systemd
186 - merge busctl into systemctl or so?
187 - synthesize sd_bus_message objects from kernel messages
188 - properly implement name registry ioctls for kdbus
189 - get rid of object hash table, use decision tree everyhwere instead?
190 - implement monitor logic
191 - object vtable logic
194 * priority inheritance
196 * in the final killing spree, detect processes from the root directory, and
197 complain loudly if they have argv[0][0] == '@' set.
198 https://bugzilla.redhat.com/show_bug.cgi?id=961044
200 * add an option to nspawn that uses seccomp to make socket(AF_NETLINK,
201 SOCK_RAW, NETLINK_AUDIT) fail the the appropriate error code that
202 makes the audit userspace to think auditing is not available in the
205 * Introduce a way how we can kill the main process of a service with KillSignal, but all processes with SIGKILL later on
206 https://bugzilla.redhat.com/show_bug.cgi?id=952634
208 * maybe add a warning to the unit file parses where the access mode of unit files is nonsensical.
210 * investigate endianness issues of UUID vs. GUID
212 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
213 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
215 * if pam_systemd is invoked by su from a process that is outside of a
216 any session we should probably just become a NOP, since that's
217 usually not a real user session but just some system code that just
220 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
222 * add a pam module that on password changes updates any LUKS slot where the password matches
224 * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases...
226 * timedatctl, localectl: possibly make some commands work without the daemon, for chroot situations...
228 * logind: add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
230 * we need dynamic units
232 * cgtop: make cgtop useful in a container
235 - add 'set -e' to scripts in test/
236 - make stuff in test/ work with separate output dir
237 - remove all the duplicated code in test/
239 * systemctl delete x.snapshot leaves no trace in logs (at least at default level).
241 * make the coredump collector tool move itself into the user's cgroup
242 so that the coredump is properly written to the user's own journal
245 * seems that when we follow symlinks to units we prefer the symlink
246 destination path over /etc and /usr. We shouldn't do that. Instead
247 /etc should always override /run+/usr and also any symlink
250 * when isolating, try to figure out a way how we implicitly can order
251 all units we stop before the isolating unit...
253 * add ConditionArchitecture= or so
255 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
257 * Support SO_REUSEPORT with socket activation:
258 - Let systemd maintain a pool of servers.
259 - Use for seamless upgrades, by running the new server before stopping the
262 * after all byte-wise realloc() is slow, even on glibc, so i guess we
263 need manual exponential loops after all
265 * BootLoaderSpec: drop allowing ext234 for $BOOT. Clarify that the
266 kernel has to be in $BOOT. Clarify that the boot loader should be
267 installed to the ESP. Define a way how an installer can figure out
268 whether a BLS compliant boot loader is installed.
270 * man: remove .include documentation, and instead push people to use .d/*.conf
272 * think about requeuing jobs when daemon-reload is issued? usecase:
273 the initrd issues a reload after fstab from the host is accessible
274 and we might want to requeue the mounts local-fs acquired through
277 * rework specifier logic so that we can distinguish OOM errors from other errors
279 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
281 * journal-or-kmsg is currently broken? See reverted commit 4a01181e460686d8b4a543b1dfa7f77c9e3c5ab8.
283 * remove any syslog support from log.c -- we probably can't do this before split-off udev is gone for good
285 * documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
287 * add a tool that lists active timer units plus their next elapse and the time the units ran last
289 * man: document the very specific env the shutdown drop-in tools live in
291 * shutdown logging: store to EFI var, and store to USB stick?
293 * man: extend runlevel(8) to mention that runlevels suck, and are dead. Maybe add runlevel(7) with a note about that too
295 * systemctl: maybe add "systemctl add-wants" or so...
297 * man: add more examples to man pages
299 * man: maybe sort directives in man pages, and take sections from --help and apply them to man too
301 * write UI tool that pops up emergency messages from the journal as notification
303 * think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service
305 * msgcatalog determine default language of entries from the catalog
306 file name. i.e. foobar.de_DE.catalog would set the default entry
307 name for the entries to de_DE if they aren't explicitly suffixed
310 * use "log level" rather than "log priority" everywhere
312 * timedate: have global on/off switches for auto-time (NTP), and auto-timezone that connman can subscribe to.
314 * dev-setup.c: when running in a container, create a tiny stub udev
315 database with the systemd tag set for all network interfaces found,
316 so that libudev reports them as present, and systemd's .device units
319 * merge unit_kill_common() and unit_kill_context()
321 * introduce ExecCondition= in services
324 - fsck hookup for the ESP mount is missing
325 - write man page for efi boot generator
326 - honor language efi variables for default language selection (if there are any?)
327 - honor timezone efi variables for default timezone selection (if there are any?)
328 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
330 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
332 * re-enable "make check" for gtk-doc (broken for unknown reason)
334 * fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points
335 https://bugzilla.redhat.com/show_bug.cgi?id=812826
337 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
339 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
342 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
343 - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
344 - logind: wakelock/opportunistic suspend support
345 - Add pretty name for seats in logind
346 - logind: allow showing logout dialog from system?
347 - logind: spawn user@..service on login
348 - logind: non-local X11 server handling
349 - logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API
350 - pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree
351 - logind: GetSessionByPID() should accept 0 as PID value
352 - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case
354 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
356 * DeviceAllow/DeviceDeny: disallow everything by default, but whitelist /dev/zero, /dev/null and friends
358 * service: watchdog logic: for testing purposes allow ping, but do not require pong
361 - do not use magic msync() in src/journal/journal-file.c, just call fsync()
362 <alxchk> poettering: looks like msync just calls vfs_fsync
363 http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/mm/msync.c#n18
364 - import and delete pstore filesystem content at startup
365 - journald: also get thread ID from client, plus thread name
366 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
367 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
368 - fallback to /dev/log based logging in libsystemd-journal, if we can't log natively?
369 - declare the local journal protocol stable in the wiki interface chart
370 - journal: reuse XZ context
371 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
372 - journald: when dropping msgs due to ratelimit make sure to write
373 "dropped %u messages" not only when we are about to print the next
374 message that works, but alraedy after a short tiemout
375 - check if we can make journalctl by default use --follow mode inside of less if called without args?
376 - maybe add API to send pairs of iovecs via sd_journal_send
377 - journal: when writing journal auto-rotate if time jumps backwards
378 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
379 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
380 and !FOOBAR for events without FOOBAR.
381 - journal: when rotating, copy over old acls/access mode
382 - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs.
383 - journal-send.c, log.c: when the log socket is clogged, and we drop, count this and write a message about this when it gets unclogged again.
384 - journal: find a way to allow dropping history early, based on priority, other rules
385 - journal: When used on NFS, check payload hashes
386 - journald: check whether it is OK if the client can still modify delivered journal entries
387 - journal live copy, based on libneon (client) and libmicrohttpd (server)
388 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
389 - refuse taking lower-case variable names in sd_journal_send() and friends.
390 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
391 - journal: deal nicely with byte-by-byte copied files, especially regards header
392 - journalctl: expand tabs
393 - journal: store euid in journal if it differs from uid
394 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
395 - Replace utmp, wtmp, btmp, and lastlog completely with journal
396 - Port upower to use the journal for historical power information used in future calculations
399 - document unit_name_mangle()
400 - document that deps in [Unit] sections ignore Alias= fileds in
401 [Install] units of other units, unless those units are disabled
402 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
403 - Document word splitting syntax for ExecStart= and friends
404 - document that units from /etc override those from /usr and /run
405 - document the exit codes when services fail before they are exec()ed
406 - document that service reload may be implemented as service reexec
407 - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications
410 - systemctl list-jobs - show dependencies
411 - add systemctl switch to dump transaction without executing it
412 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
413 - "systemctl disable" on a static unit prints no message and does
414 nothing. "systemctl enable" does nothing, and gives a bad message
415 about it. Should fix both to print nice actionable messages.
416 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
417 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
418 - systemctl enable: improve the success messages (i.e. more human readable, less shell-like)
419 - systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards?
420 - systemctl: "Journal has been rotated since unit was started." message is misleading
421 - support "systemctl stop foobar@.service" to stop all units matching a certain template
422 - Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files"
424 * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations
426 * deal with sendmail/postfix exclusivity
429 - configurable jitter for timer events
430 - timer events with system resume
431 - timer units should get the ability to trigger when:
432 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
434 - Support 2012-02~4 as syntax for specifying the fourth to last day of the month.
436 * update the kernel's TZ (sys_tz) when DST changes
438 * sync down the system time to the RTC when:
439 - CLOCK_REALTIME makes jumps (the user explicitely requested a time set)
440 - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp)
441 This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime
442 mode, it will never touch the RTC if the no reliable time source is active or the
443 user did not request anything like it.
445 * if booted in "quiet" mode, and an error happens, turn on status
446 output again, so that the emergency mode isn't totally
447 surprising. Also, terminate plymouth.
449 * libunwind support for coredump pattern hook, and includes this in
450 the message for coredumps. After all, libunwind is now capable to
451 unwind coredumps since a few weeks ago. This probably requires that
452 we have nice support for multi-line messages on display in
453 logs-show.c. Alternatively: use libelfutil, which seems to be the
454 better supported alternative.
456 * figure out relation of --all and --full in the various tools
458 * add libsystemd-password or so to query passwords during boot using the password agent logic
460 * If we show an error about a unit (such as not showing up) and it has no Description string, then show a description string generated form the reverse of unit_name_mangle().
462 * fedup: add --unit to systemctl switch-root somehow
463 * fedup: don't delete initrd on switch-root
467 - timedated: refuse time changes when NTP is on
469 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
471 * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields
473 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
475 * add "provisioning" instructions to setup an empty /etc + /var
476 - used to setup a new container from a shared /usr
477 - superset of tmpfiles model
478 - instructions shipped by packages and stored in /usr/lib/
479 - compose /etc/passwd and /etc/group, copy files
480 - able to create uid + gid used by packages, for file ownership
482 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
483 waiting for the timeout
485 * high level net_prio setting in execution context
487 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
489 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't
491 * refuse boot if /etc/machine-id is not useful (or set taint?)
494 - nspawn: consider changing users for -u with su, so that NSS resolving works correctly
495 - nspawn: implement personality changes a la linux32(8)
496 - nspawn: --read-only is not applied recursively to submounts
497 - nspawn: make use of device cgroup controller by default
498 - bind mount read-only the cgroup tree higher than nspawn
499 - nspawn: investigate whether we can support the same as LXC's
500 lxc.network.type=phys mode, and pass through entire network
501 interfaces to the container
502 - nspawn: maybe add a way to drop additional caps, in addition to add additional caps
503 - nspawn: maybe explicitly reset loginuid?
504 - nspawn: make it work for dwalsh and shared /usr containers -- tmpfs mounts as command line parameters, selinux exec context
507 - cryptsetup-generator: allow specification of passwords in crypttab itself
508 - move cryptsetup key caching into kernel keyctl?
509 https://bugs.freedesktop.org/show_bug.cgi?id=54982
510 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
512 * move debug shell to tty6 and make sure this doesn't break the gettys on tty6
514 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
515 https://bugs.freedesktop.org/show_bug.cgi?id=54712
517 * after deserializing sockets in socket.c we should reapply sockopts and things
519 * make timer units go away after they elapsed
521 * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html
522 (network and remote-fs on shutdown)
524 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
526 * move PID 1 segfaults to /var/lib/systemd/coredump?
528 * create /sbin/init symlinks from the build system
530 * Query Paul Moore about relabelling socket fds while they are open
532 * system.conf should have controls for cgroups
534 * allow writing multiple conditions in unit files on one line
536 * explore multiple service instances per listening socket idea
538 * shutdown: don't read-only mount anything when running in container
540 * MountFlags=shared acts as MountFlags=slave right now.
542 * ReadOnlyDirectories= is not applied recursively to submounts
544 * drop PID 1 reloading, only do reexecing (difficult: Reload()
545 currently is properly synchronous, Reexec() is weird, because we
546 can't delay the response properly until we are back, so instead of
547 being properly synchronous we just keep open the fd and close it
548 when done. That means clients don't get a successful method reply,
549 but much rather a disconnect on success.
551 * use opterr = 0 for all getopt tools
553 * properly handle loop back mounts via fstab, especially regards to fsck/passno
555 * allow services with no ExecStart= but with an ExecStop=
557 * dracut-shutdown needs to be ordered before unmounting /boot
559 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
561 * rename "userspace" to "core-os"
564 - syscall filter: add knowledge about compat syscalls
565 - syscall filter: don't enforce no new privs?
566 - syscall filter: option to return EPERM rather than SIGSYS?
567 - syscall filter: port to libseccomp
568 - system-wide seccomp filter
570 * .device aliases need to be implemented with the "following" logic, probably.
572 * load-fragment: when loading a unit file via a chain of symlinks
573 verify that it isn't masked via any of the names traversed.
575 * introduce Type=pid-file
577 * change Requires=basic.target to RequisiteOverride=basic.target
579 * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
581 * move passno parsing to fstab generator
583 * automount: implement expire:
584 - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD
585 - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD
586 - every timeout/4 (original autofs logic)
587 - blocking, needs a thread
589 - receive expire packet on pipe if kernel tells the timeout is over
591 - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD
592 - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns
594 * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel)
596 * ExecOnFailure=/usr/bin/foo
599 - remove src/udev/udev-builtin-firmware.c (CONFIG_FW_LOADER_USER_HELPER=n)
602 - add trigger --subsystem-match=usb/usb_device device
604 * cleanup syslog 'priority' vs. 'level' wording
606 * when a service has the same env var set twice we actually store it twice and return that in systemctl show -p... We should only show the last setting
608 * support container_ttys=
610 * introduce mix of BindTo and Requisite
612 * add DeleteSocketsOnStop=yes|no option to socket units
614 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
616 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
618 * default unix qlen is too small (10). bump sysctl? add sockopt?
620 * save coredump in Windows/Mozilla minidump format
622 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
624 * clean up session cgroups that remain after logout (think sshd), but eventually run empty
626 * when an instanced service exits, remove its parent cgroup too if possible.
628 * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max
630 * be able to specify a forced restart of service A where service B depends on, in case B
631 needs to be auto-respawned?
633 * when a bus name of a service disappears from the bus make sure to queue further activation requests
635 * tmpfiles: apply "x" on "D" too (see patch from William Douglas)
637 * for services: don't set $HOME in services unless requested
639 * hide PAM/TCPWrap options in fragment parser when compile time disabled
641 * when we automatically restart a service, ensure we restart its rdeps, too.
643 * allow Type=simple with PIDFile=
644 https://bugzilla.redhat.com/show_bug.cgi?id=723942
646 * move PAM code into its own binary
648 * implement Register= switch in .socket units to enable registration
649 in Avahi, RPC and other socket registration services.
651 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
653 * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination
655 * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1)
657 * verify that the AF_UNIX sockets of a service in the fs still exist
658 when we start a service in order to avoid confusion when a user
659 assumes starting a service is enough to make it accessible
661 * support User= and Group= attributes for AF_UNIX sockets.
663 * Make it possible to set the keymap independently from the font on
664 the kernel cmdline. Right now setting one resets also the other.
666 * and a dbus call to generate target from current state
669 - drop /.readahead on bigger upgrades with yum
670 - move readahead files into /var (look for them with .path units?)
671 - readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO
672 - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
673 - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
675 * add support for /bin/mount -s
677 * GC unreferenced jobs (such as .device jobs)
679 * write blog stories about:
680 - hwdb: what belongs into it, lsusb
681 - enabling dbus services
683 - how to make changes to sysctl and sysfs attributes
685 - how to pass throw-away units to systemd, or dynamically change properties of existing units
686 - how to integrate cgconfig and suchlike with systemd
687 - testing with Harald's awesome test kit
689 - how to develop against journal browsing APIs
690 - the journal HTTP iface
691 - non-cgroup resource management
692 - dynamic resource management with cgroups
693 - refreshed, longer missions statement
694 - calendar time events
695 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
696 - how to create your own target
697 - instantiated apache, dovecot and so on
698 - hooking a script into various stages of shutdown/rearly booot
700 * allow port=0 in .socket units
702 * support systemd.mask= on the kernel command line.
704 * recreate systemd's D-Bus private socket file on SIGUSR2
706 * Support --test based on current system state
708 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
710 * maybe introduce ExecRestartPre=
712 * dot output for --test showing the 'initial transaction'
714 * port over to LISTEN_FDS/LISTEN_PID:
715 - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH
719 - libvirtd (/var/run/libvirt/libvirt-sock-ro)
720 - bluetoothd (/var/run/sdp! @/org/bluez/audio!)
723 * fingerprint.target, wireless.target, gps.target, netdevice.target
725 * io priority during initialization
727 * drop cap bounding set in readahead and other services
730 - figure out a simple way to wait for journal events in a way that
732 - add documentation to systemd.daemon
735 - plot per-process IO utilization
736 - group processes based on service association (cgroups)
737 - document initcall_debug
738 - kernel cmdline "bootchart" option for simplicity?
740 * systemd-run is missing zsh completion scripts
746 - natively watch for dbus-*.service symlinks (PENDING)
747 - allow specification of socket mode/umask when allocating DBusServer
748 - allow disabling of fd passing when connecting a AF_UNIX connection
749 - allow disabling of UID passing for AUTH EXTERNAL
750 - always pass cred data along each message
751 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
753 * fix alsa mixer restore to not print error when no config is stored
755 * make cryptsetup lower --iter-time
757 * patch kernel for xattr support in /dev, /proc/, /sys?
759 * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only
760 in an ~30 minutes window. It does not adjust larger differences. Find a way
761 to tell the kernel, to always do a full time sync when the RTC is in UTC and
762 we are in 11-minutes-mode. When we trust the system time to NTP we also want
765 * kernel: add device_type = "fb", "fbcon" to class "graphics"
767 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
769 * dbus upstream still refers to dbus.target and shouldn't
771 * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
773 * add "# export SYSTEMD_PAGER=" to bash login
775 * /usr/bin/service should actually show the new command line
777 * fedora: suggest auto-restart on failure, but not on success and not on coredump. also, ask people to think about changing the start limit logic. Also point people to RestartPreventExitStatus=, SuccessExitStatus=
779 * fedora: F20: go timer units all the way, leave cron.daily for cron
781 * neither pkexec nor sudo initialize environ[] from the PAM environment?
783 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
787 * look for close() vs. close_nointr() vs. close_nointr_nofail()
789 * check for strerror(r) instead of strerror(-r)
791 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
793 * %m in printf() instead of strerror(errno);
797 * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()!
799 * use secure_getenv() instead of getenv() where appropriate
801 * link up selected blog stories from man pages and unit files Documentation= fields
803 Scheduled for removal or fixing:
805 * xxxOverridable dependencies (probably: fix)