2 * check systemd-tmpfiles for selinux context hookup for mknod(), symlink() and similar
4 * swap units that are activated by one name but shown in the kernel under another are semi-broken
6 * make anaconda write timeout=0 for encrypted devices
8 * Dangling symlinks of .automount unit files in .wants/ directories, set up
9 automount points even when the original .automount file did not exist
10 anymore. Only the .mount unit was still around.
12 * make polkit checks async
14 * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point.
19 * create /var/log/journal/
23 * link runlevel targets on install only if sysv compat is enabled
25 * logind: when executing sleep/shutdown job, keep track of its id, and
26 as long as the job is running refuse any further operation.
28 * maybe not install getty@tty1.service symlink in /etc but in /usr?
30 * re-enable "make check" for gtk-doc (broken for unknown reason)
32 * logind: make PrepareForSuspend(false) an official api for notification of resumes
34 * "systemctl disable" on a static unit prints no message and does
35 nothing. "systemctl enable" does nothing, and gives a bad message
36 about it. Should fix both to print nice actionable messages.
38 * fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points
39 https://bugzilla.redhat.com/show_bug.cgi?id=812826
41 * add "loginctl unlock-sessions" as counterpart for "loginctl lock-sessions"
44 - don't filter out duplicate settings
45 - support high-level cgroup setting syntax in systemctl. Example: "systemctl set-cgroup-attr MemoryLimit 5K"
46 - support writte string mapping even for non-high-level settings
47 - add man page for systemctl commands
48 - make sure we work fine with multi-line strings
50 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
52 * kernel cmdline switch to turn off predictable network interface names
54 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
56 * logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
58 * service: when killing a service with SIGKILL always kill all processes, even if for SIGTERM we only killed the main process
60 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
62 * DeviceAllow/DeviceDeny: disallow everything by default, but whitelist /dev/zero, /dev/null and friends
64 * service: watchdog logic: for testing purposes allow ping, but do not require pong
66 * journald: when dropping msgs due to ratelimit make sure to write
67 "dropped %u messages" not only when we are about to print the next
68 message that works, but alraedy after a short tiemout
70 * journald: also get thread ID from client, plus thread name
72 * check if we can make journalctl by default use --follow mode inside of less if called without args?
74 * Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
76 * journal is not closed properly at shutdown when run in a container?
78 * journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
80 * When shutdown.target is queued begin with an asynchronous sync()?
82 * add API to close/reopen/get fd for journal client fd in libsystemd-journal.
84 * maybe add API to send pairs of iovecs via sd_journal_send
86 * fallback to /dev/log based logging in libsystemd-journal, if we can't log natively?
88 * declare the local journal protocol stable in the wiki interface chart
90 * journal: reuse XZ context
92 * sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
94 * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations
96 * timer units should get the ability to trigger when:
97 - CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
100 * update the kernel's TZ (sys_tz) when DST changes
102 * sync down the system time to the RTC when:
103 - CLOCK_REALTIME makes jumps (the user explicitely requested a time set)
104 - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp)
105 This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime
106 mode, it will never touch the RTC if the no reliable time source is active or the
107 user did not request anything like it.
109 * When we begin with system shutdown all kind of suspend/hibernation should be prohibited until shutdown/reboot
111 * When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
114 - implement conditional properties (dmi matches)
115 - hwdb --filter=ID_DRIVE_*
117 * if booted in "quiet" mode, and an error happens, turn on status
118 output again, so that the emergency mode isn't totally
119 surprising. Also, terminate plymouth.
121 * localectl: add listing support for X11 keymaps, by parsing /usr/share/X11/xkb/rules/xorg.lst
123 * libunwind support for coredump pattern hook, and includes this in
124 the message for coredumps. After all, libunwind is now capable to
125 unwind coredumps since a few weeks ago. This probably requires that
126 we have nice support for multi-line messages on display in
127 logs-show.c. Alternatively: use libelfutil, which seems to be the
128 better supported alternative.
130 * figure out relation of --all and --full in the various tools
132 * journal: when writing journal auto-rotate if time jumps backwards
134 * introduce new "journal" group in place of adm? introduce groups for the various mini daemons?
136 * journal: add a setgid "adm" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
138 * journactl: support negative filtering, i.e. FOOBAR!="waldo",
139 and !FOOBAR for events without FOOBAR.
141 * print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
143 * add libsystemd-password or so to query passwords during boot using the password agent logic
145 * journal: when rotating, copy over old acls/access mode
147 * journal: document why we do not give ownership to journal files to the user that created them but use FS ACLs for that
149 * journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs.
151 * journal: when we haven't written anything in a while, sync to disk and mark file as offline, in order to be more often than not in a clean state
153 * 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.
155 * 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().
157 * fedup: add --unit to systemctl switch-root somehow
158 * fedup: don't delete initrd on switch-root
161 * journal: find a way to allow dropping history early, based on priority, other rules
163 * journal: When used on NFS, check payload hashes
165 * journal: When used on NFS make sure wake up sd_journal_wait() every 2s, to handle missing inotify
167 * document that people can use file system ACLs to manage access to journal files, with example
169 * timedated: export boolean that clarifies whether NTP is even available
171 * timedated: refuse time changes when NTP is on
173 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
175 * document unit_name_mangle()
177 * add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
179 * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields
181 * deal with sendmail/postfix exclusivity
183 * systemctl enable: improve the success messages (i.e. more human readable, less shell-like)
185 * systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards?
187 * on shutdown: move utmp, wall, audit logic all into PID 1 itself, get rid of systemd-update-utmp-runlevel
189 * add "provisioning" instructions to setup an empty /etc + /var
190 - used to setup a new container from a shared /usr
191 - superset of tmpfiles model
192 - instructions shipped by packages and stored in /usr/lib/
193 - compose /etc/passwd and /etc/group, copy files
194 - able to create uid + gid used by packages, for file ownership
196 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
197 waiting for the timeout
199 * high level net_prio setting in execution context
201 * Introduce journalctl -b <nr> to show journal messages of a previous boot
203 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
205 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't
207 * WorkingDirectory: support env var replacements like in ExecStart= so that people can use $HOME
209 * refuse boot if /etc/machine-id is not useful (or set taint?)
211 * nspawn: consider changing users for -u with su, so that NSS resolving works correctly
213 * nspawn: implement personality changes a la linux32(8)
215 * cryptsetup-generator: warn if the password files are world-readable
217 * cryptsetup-generator: add RequiresMountsFor= to cryptseup service files referencing a file, similar for devices
219 * cryptsetup-generator: allow specification of passwords in crypttab itself
221 * document that deps in [Unit] sections ignore Alias= fileds in
222 [Install] units of other units, unless those units are disabled
224 * instantiated [Install] for target units
225 https://bugs.freedesktop.org/show_bug.cgi?id=54377
227 * move debug shell to tty6 and make sure this doesn't break the gettys on tty6
229 * move cryptsetup key caching into kernel keyctl?
230 https://bugs.freedesktop.org/show_bug.cgi?id=54982
232 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
233 https://bugs.freedesktop.org/show_bug.cgi?id=54712
235 * after deserializing sockets in socket.c we should reapply sockopts and things
237 * make timer units go away after they elapsed
239 * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html
240 (network and remote-fs on shutdown)
242 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
244 * journald: add kernel cmdline option to disable ratelimiting for debug purposes
246 * move PID 1 segfaults to /var/lib/systemd/coredump?
248 * Document word splitting syntax for ExecStart= and friends
250 * create /sbin/init symlinks from the build system
252 * Query Paul Moore about relabelling socket fds while they are open
254 * journald: check whether it is OK if the client can still modify delivered journal entries
256 * journal live copy, based on libneon (client) and libmicrohttpd
258 * system-wide seccomp filter
260 * system.conf should have controls for cgroups
262 * bind mount read-only the cgroup tree higher than nspawn
264 * allow writing multiple conditions in unit files on one line
266 * explore multiple service instances per listening socket idea
268 * testing tool for socket activation: some binary that listens on a socket and passes it on using the usual socket activation protocol to some server.
270 * shutdown: don't read-only mount anything when running in container
272 * nspawn: --read-only is not applied recursively to submounts
274 * MountFlags=shared acts as MountFlags=slave right now.
276 * ReadOnlyDirectories= is not applied recursively to submounts
278 * drop PID 1 reloading, only do reexecing (difficult: Reload()
279 currently is properly synchronous, Reexec() is weird, because we
280 can't delay the response properly until we are back, so instead of
281 being properly synchronous we just keep open the fd and close it
282 when done. That means clients don't get a successful method reply,
283 but much rather a disconnect on success.
285 * document that service reload may be implemented as service reexec
287 * remember which condition failed for services, not just the fact that something failed
289 * use opterr = 0 for all getopt tools
291 * properly handle loop back mounts via fstab, especially regards to fsck/passno
293 * allow services with no ExecStart= but with an ExecStop=
295 * add proper journal support to "systemctl --user status ..."
297 * add _SYSTEMD_USER_UNIT= field to journal entries
299 * dracut-shutdown needs to be ordered before unmounting /boot
301 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
303 * rename "userspace" to "core-os"
305 * systemctl: "Journal has been rotated since unit was started." message is misleading
307 * syscall filter: add knowledge about compat syscalls
309 * syscall filter: don't enforce no new privs?
311 * syscall filter: option to return EPERM rather than SIGSYS?
313 * syscall filter: port to libseccomp
315 * logind: wakelock/opportunistic suspend support
317 * systemd-analyze post-boot is broken for initrd
319 * systemd-analyze: data collection tools should be lightweight (few dependencies); data analysis tools can be heavyweight
321 * man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
323 * .device aliases need to be implemented with the "following" logic, probably.
325 * refuse taking lower-case variable names in sd_journal_send() and friends.
327 * load-fragment: when loading a unit file via a chain of symlinks
328 verify that it isn't masked via any of the names traversed.
330 * journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
332 * introduce Type=pid-file
334 * maybe allow services with ExecStop= set, but no ExecStart=?
336 * efi: implement /forcefsck as uefi variables thus not requiring file system altering to trigger a file system check
338 * efi: honor language efi variables for default language selection
340 * efi: honor timezone efi variables for default timezone selection
342 * efi: automatically mount EFI partition to /boot if no such entry exists in /etc/fstab and /boot is empty
343 gummiboot exports the EFI system partion (ESP) device:
344 /sys/firmware/efi/vars/LoaderDeviceIdentifier-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
345 Acpi(PNP0A03,0)/Pci(1F|2)/?/HD(Part1,Sig1FCBC57F-4BFC-4C2B-91A3-9C84FBCD9AF1)
346 '/' is the separator for the device path list
347 HD(Part1,Sig1FCBC57F-4BFC-4C2B-91A3-9C84FBCD9AF1) contains the GPT UUID of the ESP
349 * read the bootloader performance data (raw TSC) in systemd-analyze
350 /sys/firmware/efi/vars/LoaderTicksExec-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
352 /sys/firmware/efi/vars/LoaderTicksInit-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
354 /sys/firmware/efi/vars/LoaderTicksStartMenu-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
355 (only set if the menu was active)
357 * change Requires=basic.target to RequisiteOverride=basic.target
359 * support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
361 * nspawn: make use of device cgroup controller by default
363 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
365 * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
367 * readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
369 * move passno parsing to fstab generator
371 * improve !/proc/*/loginuid situation: make /proc/*/loginuid less dependent on CONFIG_AUDIT,
372 or use the users cgroup information when /proc/*/loginuid is not available.
374 * pam_systemd: try to get old session id from cgroup, if audit sessionid cannot be determined
376 * pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree
378 * maybe introduce ~/.config/locale.conf and apply it within PAM
380 * readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
382 * automount: implement expire:
383 - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD
384 - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD
385 - every timeout/4 (original autofs logic)
386 - blocking, needs a thread
388 - receive expire packet on pipe if kernel tells the timeout is over
390 - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD
391 - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns
393 * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel)
395 * don't delete /tmp/systemd-namespace-* before a process is gone down
397 * ExecOnFailure=/usr/bin/foo
399 * fedora: make sshd and pam_loginuid work in nspawn containers
401 * fix utmp for console logins in containers
403 * Add pretty name for seats in logind
405 * ConditionSecurity= should learn about IMA and SMACK
408 - remove all (misguided from day 1) userspace firmware_class handling
410 - unify utf8 validator code with shared/
412 - add trigger --subsystem-match=usb/usb_device device
414 * cleanup syslog 'priority' vs. 'level' wording
416 * dbus upstream still refers to dbus.target and shouldn't
418 * 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
420 * support container_ttys=
422 * introduce mix of BindTo and Requisite
424 * journalctl: show multiline log messages sanely, expand tabs, and show all valid utf8 messages
426 * add DeleteSocketsOnStop=yes|no option to socket units
428 * journal: store euid in journal if it differs from uid
430 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
432 * journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
434 * add command to systemctl to plot dependency graph as tree (see rhbz 795365)
436 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
438 * default unix qlen is too small (10). bump sysctl? add sockopt?
440 * figure out whether we should leave dbus around during shutdown
442 * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
444 * dbus: move dbus to early boot
446 * logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API
448 * journal: deal nicely with byte-by-byte copied files, especially regards header
450 * journal: local deserializer of export mode, http server
452 * document the exit codes when services fail before they are exec()ed
454 * save coredump in Windows/Mozilla minidump format
456 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
458 * clean up session cgroups that remain after logout (think sshd), but eventually run empty
460 * support "systemctl stop foobar@.service" to stop all units matching a certain template
462 * logind: allow showing logout dialog from system
464 * document that %% can be used to write % in a string that is specifier extended
466 * when an instanced service exits, remove its parent cgroup too if possible.
468 * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max
470 * be able to specify a forced restart of service A where service B depends on, in case B
471 needs to be auto-respawned?
473 * Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files"
475 * when a bus name of a service disappears from the bus make sure to queue further activation requests
477 * something like ConditionExec= or ExecCondition= without failure state
479 * tmpfiles: apply "x" on "D" too (see patch from William Douglas)
481 * don't set $HOME in services unless requested
483 * hide PAM/TCPWrap options in fragment parser when compile time disabled
485 * when we automatically restart a service, ensure we restart its rdeps, too.
487 * allow Type=simple with PIDFile=
488 https://bugzilla.redhat.com/show_bug.cgi?id=723942
490 * move PAM code into its own binary
492 * logind: spawn user@..service on login
494 * logind: non-local X11 server handling
496 * implement Register= switch in .socket units to enable registration
497 in Avahi, RPC and other socket registration services.
499 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
501 * readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO
503 * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination
505 * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1)
507 * verify that the AF_UNIX sockets of a service in the fs still exist
508 when we start a service in order to avoid confusion when a user
509 assumes starting a service is enough to make it accessible
511 * Make it possible to set the keymap independently from the font on
512 the kernel cmdline. Right now setting one resets also the other.
514 * and a dbus call to generate target from current state
516 * drop /.readahead on bigger upgrades with yum
518 * add support for /bin/mount -s
520 * GC unreferenced jobs (such as .device jobs)
522 * write blog stories about:
523 - hwdb: what belongs into it, lsusb
524 - enabling dbus services
526 - how to make changes to sysctl and sysfs attributes
528 - how to pass throw-away units to systemd, or dynamically change properties of existing units
529 - how to integrate cgconfig and suchlike with systemd
530 - testing with Harald's awesome test kit
532 - how to develop against journal browsing APIs
533 - the journal HTTP iface
534 - non-cgroup resource management
535 - dynamic resource management with cgroups
536 - refreshed, longer missions statement
537 - celendar time events
539 * allow port=0 in .socket units
541 * move readahead files into /var (look for them with .path units?)
543 * teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
545 * support systemd.mask= on the kernel command line.
547 * when key file cannot be found, read it from kbd in cryptsetup
549 * reuse mkdtemp namespace dirs in /tmp?
551 * recreate systemd's D-Bus private socket file on SIGUSR2
553 * Support --test based on current system state
555 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
557 * maybe introduce ExecRestartPre=
559 * configurable jitter for timer events
561 * timer events with system resume
563 * dot output for --test showing the 'initial transaction'
565 * writable cgroups dbus properties for live changes
567 * port over to LISTEN_FDS/LISTEN_PID:
568 - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH
572 - libvirtd (/var/run/libvirt/libvirt-sock-ro)
573 - bluetoothd (/var/run/sdp! @/org/bluez/audio!)
576 * fingerprint.target, wireless.target, gps.target, netdevice.target
578 * io priority during initialization
580 * systemctl list-jobs - show dependencies
582 * add systemctl switch to dump transaction without executing it
584 * drop cap bounding set in readahead and other services
590 - natively watch for dbus-*.service symlinks (PENDING)
591 - allow specification of socket mode/umask when allocating DBusServer
592 - allow disabling of fd passing when connecting a AF_UNIX connection
593 - allow disabling of UID passing for AUTH EXTERNAL
594 - always pass cred data along each message
596 * fix alsa mixer restore to not print error when no config is stored
598 * gnome-shell python script/glxinfo/is-accelerated must die
600 * make cryptsetup lower --iter-time
602 * patch kernel for xattr support in /dev, /proc/, /sys?
604 * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only
605 in an ~30 minutes window. It does not adjust larger differences. Find a way
606 to tell the kernel, to always do a full time sync when the RTC is in UTC and
607 we are in 11-minutes-mode. When we trust the system time to NTP we also want
610 * kernel: add device_type = "fb", "fbcon" to class "graphics"
614 * look for close() vs. close_nointr() vs. close_nointr_nofail()
616 * check for strerror(r) instead of strerror(-r)
618 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
620 * %m in printf() instead of strerror(errno);
624 * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()!
626 * use secure_getenv() instead of getenv() where appropriate
628 * link up selected blog stories from man pages and unit files Documentation= fields
630 Scheduled for removal or fixing:
632 * xxxOverridable dependencies (probably: fix)