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.
22 * man page for catalog APIs
24 * When shutdown.target is queued begin with an asynchronous sync()?
26 * Add ConditionBatteryPower= or ConditionACPower=? (but definitely not both)
28 * add API to close/reopen/get fd for journal client fd in libsystemd-journal.
30 * maybe add API to send pairs of iovecs via sd_journal_send
32 * fallback to /dev/log based logging in libsystemd-journal, if we can't log natively?
34 * declare the local journal protocol stable in the wiki interface chart
36 * sd-journal: don't return fields > a threshold by default
37 * journal: reuse XZ context
38 * sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
40 * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations
42 * timer units should get the ability to trigger when:
43 - CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
46 * update the kernel's TZ (sys_tz) when DST changes
48 * sync down the system time to the RTC when:
49 - CLOCK_REALTIME makes jumps (the user explicitely requested a time set)
50 - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp)
51 This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime
52 mode, it will never touch the RTC if the no reliable time source is active or the
53 user did not request anything like it.
55 * When we begin with system shutdown all kind of suspend/hibernation should be prohibited until shutdown/reboot
57 * When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
60 - implement conditional properties (dmi matches)
61 - hwdb --filter=ID_DRIVE_*
62 - find out what to do for blockdevs and skipping scsi modaliases
63 - move writing code to src/libudev/libudev-hwdb-private.c
65 * if booted in "quiet" mode, and an error happens, turn on status output again, so that the emergency mode isn't totally surprising
67 * localectl: add listing support for X11 keymaps, by parsing /usr/share/X11/xkb/rules/xorg.lst
69 * libunwind support for coredump pattern hook, and includes this in
70 the message for coredumps. After all, libunwind is now capable to
71 unwind coredumps since a few weeks ago. This probably requires that
72 we have nice support for multi-line messages on display in logs-show.c.
74 * figure out relation of --all and --full in the various tools
76 * journal: when writing journal auto-rotate if time jumps backwards
78 * introduce new "journal" group in place of adm? introduce groups for the various mini daemons?
80 * journal: add a setgid "adm" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
82 * link up selected blog stories from man pages?
84 * journactl: support negative filtering, i.e. FOOBAR!="waldo",
85 and !FOOBAR for events without FOOBAR.
87 * print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
89 * add libsystemd-password or so to query passwords during boot using the password agent logic
91 * journal: when rotating, copy over old acls/access mode
93 * journal: document why we do not give ownership to journal files to the user that created them but use FS ACLs for that
95 * journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs.
97 * 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
99 * 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.
101 * 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().
103 * fedup: add --unit to systemctl switch-root somehow
104 * fedup: don't delete initrd on switch-root
107 * journal: find a way to allow dropping history early, based on priority, other rules
109 * journal: When used on NFS, check payload hashes
111 * journal: When used on NFS make sure wake up sd_journal_wait() every 2s, to handle missing inotify
113 * document that people can use file system ACLs to manage access to journal files, with example
115 * don't show cgroup in "systemctl status" if empty/non-existent, especially for foreign .mount units
117 * timedated: export boolean that clarifies whether NTP is even available
119 * timedated: refuse time changes when NTP is on
121 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
123 * document unit_name_mangle()
125 * add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
127 * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields
129 * deal with sendmail/postfix exclusivity
131 * systemctl enable: improve the success messages (i.e. more human readable, less shell-like)
133 * systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards?
135 * on shutdown: move utmp, wall, audit logic all into PID 1 itself, get rid of systemd-update-utmp-runlevel
137 * add "provisioning" instructions to setup an empty /etc + /var
138 - used to setup a new container from a shared /usr
139 - superset of tmpfiles model
140 - instructions shipped by packages and stored in /usr/lib/
141 - compose /etc/passwd and /etc/group, copy files
142 - able to create uid + gid used by packages, for file ownership
144 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
145 waiting for the timeout
147 * high level net_prio setting in execution context
149 * Introduce journalctl -b <nr> to show journal messages of a previous boot
151 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
153 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't
155 * WorkingDirectory: support env var replacements like in ExecStart= so that people can use $HOME
157 * refuse boot if /etc/machine-id is not useful
159 * nspawn: consider changing users for -u with su, so that NSS resolving works correctly
161 * nspawn: implement personality changes a la linux32(8)
163 * nspawn: reset all aux groups
165 * cryptsetup-generator: warn if the password files are world-readable
167 * cryptsetup-generator: add RequiresMountsFor= to cryptseup service files referencing a file, similar for devices
169 * cryptsetup-generator: allow specification of passwords in crypttab itself
171 * document that deps in [Unit] sections ignore Alias= fileds in
172 [Install] units of other units, unless those units are disabled
174 * need to update LGPL2.1 text to newest version (with updated FSF address)
176 * systemctl: when powering down/suspending check for inhibitors, and warn.
178 * instantiated [Install] for target units
179 https://bugs.freedesktop.org/show_bug.cgi?id=54377
181 * move debug shell to tty6 and make sure this doesn't break the gettys on tty6
183 * move cryptsetup key caching into kernel keyctl?
184 https://bugs.freedesktop.org/show_bug.cgi?id=54982
186 * make nspawn work without terminal
188 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
189 https://bugs.freedesktop.org/show_bug.cgi?id=54712
191 * after deserializing sockets in socket.c we should reapply sockopts and things
193 * make timer units go away after they elapsed
195 * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html
197 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
199 * journald: add kernel cmdline option to disable ratelimiting for debug purposes
201 * move PID 1 segfaults to /var/lib/systemd/coredump?
203 * Document word splitting syntax for ExecStart= and friends
205 * create /sbin/init symlinks from the build system
207 * Query Paul Moore about relabelling socket fds while they are open
209 * move keymaps to /usr/lib/... rather than /usr/lib/udev/...
211 * journald: check whether it is OK if the client can still modify delivered journal entries
213 * journal live copy, based on libneon (client) and libmicrohttpd
215 * system-wide seccomp filter
217 * ability to pass fds into systemd
219 * system.conf should have controls for cgroups
221 * bind mount read-only the cgroup tree higher than nspawn
223 * allow writing multiple conditions in unit files on one line
225 * explore multiple service instances per listening socket idea
227 * 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.
229 * shutdown: don't read-only mount anything when running in container
231 * nspawn: --read-only is not applied recursively to submounts
233 * MountFlags=shared acts as MountFlags=slave right now.
235 * ReadOnlyDirectories= is not applied recursively to submounts
237 * drop PID 1 reloading, only do reexecing (difficult: Reload()
238 currently is properly synchronous, Reexec() is weird, because we
239 can't delay the response properly until we are back, so instead of
240 being properly synchronous we just keep open the fd and close it
241 when done. That means clients don't get a successful method reply,
242 but much rather a disconnect on success.
244 * document that service reload may be implemented as service reexec
246 * remember which condition failed for services, not just the fact that something failed
248 * use opterr = 0 for all getopt tools
250 * properly handle loop back mounts via fstab, especially regards to fsck/passno
252 * allow services with no ExecStart= but with an ExecStop=
254 * add proper journal support to "systemctl --user status ..."
256 * add _SYSTEMD_USER_UNIT= field to journal entries
258 * dracut-shutdown needs to be ordered before unmounting /boot
260 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
262 * rename "userspace" to "core-os"
264 * systemctl: "Journal has been rotated since unit was started." message is misleading
266 * syscall filter: add knowledge about compat syscalls
268 * syscall filter: don't enforce no new privs?
270 * syscall filter: option to return EPERM rather than SIGSYS?
272 * syscall filter: port to libseccomp
274 * logind: wakelock/opportunistic suspend support
276 * systemd-analyze post-boot is broken for initrd
278 * man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
280 * .device aliases need to be implemented with the "following" logic, probably.
282 * refuse taking lower-case variable names in sd_journal_send() and friends.
284 * load-fragment: when loading a unit file via a chain of symlinks
285 verify that it isn't masked via any of the names traversed.
287 * journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
292 * introduce Type=pid-file
294 * maybe allow services with ExecStop= set, but no ExecStart=?
296 * efi: implement /forcefsck as uefi variables thus not requiring file system altering to trigger a file system check
298 * efi: honor language efi variables for default language selection
300 * efi: honor timezone efi variables for default timezone selection
302 * efi: automatically mount EFI partition to /boot if no such entry exists in /etc/fstab and /boot is empty
303 gummiboot exports the EFI system partion (ESP) device:
304 /sys/firmware/efi/vars/LoaderDeviceIdentifier-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
305 Acpi(PNP0A03,0)/Pci(1F|2)/?/HD(Part1,Sig1FCBC57F-4BFC-4C2B-91A3-9C84FBCD9AF1)
306 '/' is the separator for the device path list
307 HD(Part1,Sig1FCBC57F-4BFC-4C2B-91A3-9C84FBCD9AF1) contains the GPT UUID of the ESP
309 * read the bootloader performance data (raw TSC) in systemd-analyze
310 /sys/firmware/efi/vars/LoaderTicksExec-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
312 /sys/firmware/efi/vars/LoaderTicksInit-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
314 /sys/firmware/efi/vars/LoaderTicksStartMenu-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
315 (only set if the menu was active)
317 * change Requires=basic.target to RequisiteOverride=basic.target
319 * support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
321 * nspawn: make use of device cgroup contrller by default
323 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
325 * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
327 * readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
329 * (attempt to) make Debianites happy:
330 - implement .d/ auto includes for unit files
331 - add syntax to reset ExecStart= lists (and similar)
333 * move passno parsing to fstab generator
335 * improve !/proc/*/loginuid situation: make /proc/*/loginuid less dependent on CONFIG_AUDIT,
336 or use the users cgroup information when /proc/*/loginuid is not available.
338 * pam_systemd: try to get old session id from cgroup, if audit sessionid cannot be determined
340 * pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree
342 * maybe introduce ~/.config/locale.conf and apply it within PAM
344 * readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
346 * automount: implement expire
348 * logind: auto-suspend, auto-shutdown:
349 IdleAction=(none|suspend|opportunistic|hibernate|poweroff)
351 SessionIdleMode=(explicit|ignore|login)
352 ForceShutdown=(yes|no)
354 * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel)
356 * don't delete /tmp/systemd-namespace-* before a process is gone down
358 * vconsole: implement setterm -store -foreground xxx --background zzz
360 * ExecOnFailure=/usr/bin/foo
362 * fedora: make sshd and pam_loginuid work in nspawn containers
364 * fix utmp for console logins in containers
366 * Add pretty name for seats in logind
368 * ConditionSecurity= should learn about IMA
370 * Auke: merge Auke's bootchart
374 * udev systemd unify:
375 - strpcpy(), strpcpyl(), strscpy(), strscpyl()
376 - utf8 validator code
378 * udev: scsi_id -> sg3_utils -> kill scsi_id
380 * udev: add trigger --subsystem-match=usb/usb_device device
382 * allow configuration of console width/height in vconsole.conf
384 * cleanup syslog 'priority' vs. 'level' wording
386 * dbus upstream still refers to dbus.target and shouldn't
388 * 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
390 * support container_ttys=
392 * introduce mix of BindTo and Requisite
394 * journalctl: show multiline log messages sanely, expand tabs, and show all valid utf8 messages
396 * add DeleteSocketsOnStop=yes|no option to socket units
398 * journal: store euid in journal if it differs from uid
400 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
402 * journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
404 * add command to systemctl to plot dependency graph as tree (see rhbz 795365)
406 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
408 * default unix qlen is too small (10). bump sysctl? add sockopt?
410 * figure out whether we should leave dbus around during shutdown
412 * dbus: in fedora, make the machine a symlink to /etc/machine-id
414 * dbus: move dbus to early boot
416 * logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API
418 * journal: deal nicely with byte-by-byte copied files, especially regards header
420 * journal: local deserializer of export mode, http server
422 * document the exit codes when services fail before they are exec()ed
424 * save coredump in Windows/Mozilla minidump format
426 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
428 * clean up session cgroups that remain after logout (think sshd), but eventually run empty
430 * support "systemctl stop foobar@.service" to stop all units matching a certain template
432 * logind: allow showing logout dialog from system
434 * document that %% can be used to write % in a string that is specifier extended
436 * when an instanced service exits, remove its parent cgroup too if possible.
438 * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max
440 * be able to specify a forced restart of service A where service B depends on, in case B
441 needs to be auto-respawned?
443 * Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files"
445 * when a bus name of a service disappears from the bus make sure to queue further activation requests
447 * something like ConditionExec= or ExecStartPre= without failure state
449 * tmpfiles: apply "x" on "D" too (see patch from William Douglas)
451 * don't set $HOME in services unless requested
453 * hide PAM/TCPWrap options in fragment parser when compile time disabled
455 * when we automatically restart a service, ensure we restart its rdeps, too.
457 * allow Type=simple with PIDFile=
458 https://bugzilla.redhat.com/show_bug.cgi?id=723942
460 * move PAM code into its own binary
462 * logind: spawn user@..service on login
464 * logind: non-local X11 server handling
466 * implement Register= switch in .socket units to enable registration
467 in Avahi, RPC and other socket registration services.
469 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
471 * readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO
473 * readahead: check whether a btrfs volume includes ssd by checking mount flag "ssd"
475 * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination
477 * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1)
479 * verify that the AF_UNIX sockets of a service in the fs still exist
480 when we start a service in order to avoid confusion when a user
481 assumes starting a service is enough to make it accessible
483 * Make it possible to set the keymap independently from the font on
484 the kernel cmdline. Right now setting one resets also the other.
486 * move nss-myhostname into systemd
488 * and a dbus call to generate target from current state
490 * drop /.readahead on bigger upgrades with yum
492 * add support for /bin/mount -s
494 * GC unreferenced jobs (such as .device jobs)
496 * write blog stories about:
497 - hwdb: what belongs into it, lsusb
498 - enabling dbus services
500 - how to make changes to sysctl and sysfs attributes
502 - how to pass throw-away units to systemd, or dynamically change properties of existing units
503 - how to integrate cgconfig and suchlike with systemd
504 - testing with Harald's awesome test kit
506 - how to develop against journal browsing APIs
507 - the journal HTTP iface
508 - non-cgroup resource management
510 * allow port=0 in .socket units
512 * move readahead files into /var (look for them with .path units?)
514 * teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
516 * support systemd.mask= on the kernel command line.
518 * when key file cannot be found, read it from kbd in cryptsetup
520 * reuse mkdtemp namespace dirs in /tmp?
522 * recreate systemd's D-Bus private socket file on SIGUSR2
524 * Support --test based on current system state
526 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
528 * maybe introduce ExecRestartPre=
530 * configurable jitter for timer events
532 * timer events with system resume
534 * dot output for --test showing the 'initial transaction'
536 * calendar time support in timer:
537 https://docs.google.com/document/pub?id=1bAMyFAjWLpzR3GTDYdgj5FWRMxoZiWw5zmUHEtvdHKA
539 * writable cgroups dbus properties for live changes
541 * port over to LISTEN_FDS/LISTEN_PID:
542 - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH
546 - libvirtd (/var/run/libvirt/libvirt-sock-ro)
547 - bluetoothd (/var/run/sdp! @/org/bluez/audio!)
550 * fingerprint.target, wireless.target, gps.target, netdevice.target
552 * io priority during initialization
554 * systemctl list-jobs - show dependencies
556 * add systemctl switch to dump transaction without executing it
558 * drop cap bounding set in readahead and other services
564 - natively watch for dbus-*.service symlinks (PENDING)
565 - allow specification of socket mode/umask when allocating DBusServer
566 - allow disabling of fd passing when connecting a AF_UNIX connection
567 - allow disabling of UID passing for AUTH EXTERNAL
568 - always pass cred data along each message
570 * fix alsa mixer restore to not print error when no config is stored
572 * gnome-shell python script/glxinfo/is-accelerated must die
574 * make cryptsetup lower --iter-time
576 * patch kernel for xattr support in /dev, /proc/, /sys and /sys/fs/cgroup?
578 * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only
579 in an ~30 minutes window. It does not adjust larger differences. Find a way
580 to tell the kernel, to always do a full time sync when the RTC is in UTC and
581 we are in 11-minutes-mode. When we trust the system time to NTP we also want
584 * kernel: add device_type = "fb", "fbcon" to class "graphics"
588 * look for close() vs. close_nointr() vs. close_nointr_nofail()
590 * check for strerror(r) instead of strerror(-r)
592 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
594 * %m in printf() instead of strerror(errno);
598 * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()!
600 * use secure_getenv() instead of getenv() where appropriate
602 Scheduled for removal or fixing:
604 * xxxOverridable dependencies (probably: fix)
606 * support for early-boot SysV services (definitely: remove)
608 * insserv support (definitely: remove)