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"
45 - don't filter out duplicate settings
46 - support high-level cgroup setting syntax in systemctl. Example: "systemctl set-cgroup-attr MemoryLimit 5K"
47 - support writte string mapping even for non-high-level settings
48 - add man page for systemctl commands
49 - make sure we work fine with multi-line strings
51 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
53 * kernel cmdline switch to turn off predictable network interface names
55 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
57 * logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
59 * service: when killing a service with SIGKILL always kill all processes, even if for SIGTERM we only killed the main process
61 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
63 * DeviceAllow/DeviceDeny: disallow everything by default, but whitelist /dev/zero, /dev/null and friends
65 * service: watchdog logic: for testing purposes allow ping, but do not require pong
67 * journald: when dropping msgs due to ratelimit make sure to write
68 "dropped %u messages" not only when we are about to print the next
69 message that works, but alraedy after a short tiemout
71 * journald: also get thread ID from client, plus thread name
73 * check if we can make journalctl by default use --follow mode inside of less if called without args?
75 * Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
77 * journal is not closed properly at shutdown when run in a container?
79 * journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
81 * When shutdown.target is queued begin with an asynchronous sync()?
83 * add API to close/reopen/get fd for journal client fd in libsystemd-journal.
85 * maybe add API to send pairs of iovecs via sd_journal_send
87 * fallback to /dev/log based logging in libsystemd-journal, if we can't log natively?
89 * declare the local journal protocol stable in the wiki interface chart
91 * journal: reuse XZ context
93 * sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
95 * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations
97 * timer units should get the ability to trigger when:
98 - CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
101 * update the kernel's TZ (sys_tz) when DST changes
103 * sync down the system time to the RTC when:
104 - CLOCK_REALTIME makes jumps (the user explicitely requested a time set)
105 - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp)
106 This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime
107 mode, it will never touch the RTC if the no reliable time source is active or the
108 user did not request anything like it.
110 * When we begin with system shutdown all kind of suspend/hibernation should be prohibited until shutdown/reboot
112 * When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
115 - implement conditional properties (dmi matches)
116 - hwdb --filter=ID_DRIVE_*
118 * if booted in "quiet" mode, and an error happens, turn on status output again, so that the emergency mode isn't totally surprising
120 * localectl: add listing support for X11 keymaps, by parsing /usr/share/X11/xkb/rules/xorg.lst
122 * libunwind support for coredump pattern hook, and includes this in
123 the message for coredumps. After all, libunwind is now capable to
124 unwind coredumps since a few weeks ago. This probably requires that
125 we have nice support for multi-line messages on display in
126 logs-show.c. Alternatively: use libelfutil, which seems to be the
127 better supported alternative.
129 * figure out relation of --all and --full in the various tools
131 * journal: when writing journal auto-rotate if time jumps backwards
133 * introduce new "journal" group in place of adm? introduce groups for the various mini daemons?
135 * journal: add a setgid "adm" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
137 * journactl: support negative filtering, i.e. FOOBAR!="waldo",
138 and !FOOBAR for events without FOOBAR.
140 * print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
142 * add libsystemd-password or so to query passwords during boot using the password agent logic
144 * journal: when rotating, copy over old acls/access mode
146 * journal: document why we do not give ownership to journal files to the user that created them but use FS ACLs for that
148 * journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs.
150 * 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
152 * 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.
154 * 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().
156 * fedup: add --unit to systemctl switch-root somehow
157 * fedup: don't delete initrd on switch-root
160 * journal: find a way to allow dropping history early, based on priority, other rules
162 * journal: When used on NFS, check payload hashes
164 * journal: When used on NFS make sure wake up sd_journal_wait() every 2s, to handle missing inotify
166 * document that people can use file system ACLs to manage access to journal files, with example
168 * timedated: export boolean that clarifies whether NTP is even available
170 * timedated: refuse time changes when NTP is on
172 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
174 * document unit_name_mangle()
176 * add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
178 * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields
180 * deal with sendmail/postfix exclusivity
182 * systemctl enable: improve the success messages (i.e. more human readable, less shell-like)
184 * systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards?
186 * on shutdown: move utmp, wall, audit logic all into PID 1 itself, get rid of systemd-update-utmp-runlevel
188 * add "provisioning" instructions to setup an empty /etc + /var
189 - used to setup a new container from a shared /usr
190 - superset of tmpfiles model
191 - instructions shipped by packages and stored in /usr/lib/
192 - compose /etc/passwd and /etc/group, copy files
193 - able to create uid + gid used by packages, for file ownership
195 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
196 waiting for the timeout
198 * high level net_prio setting in execution context
200 * Introduce journalctl -b <nr> to show journal messages of a previous boot
202 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
204 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't
206 * WorkingDirectory: support env var replacements like in ExecStart= so that people can use $HOME
208 * refuse boot if /etc/machine-id is not useful (or set taint?)
210 * nspawn: consider changing users for -u with su, so that NSS resolving works correctly
212 * nspawn: implement personality changes a la linux32(8)
214 * cryptsetup-generator: warn if the password files are world-readable
216 * cryptsetup-generator: add RequiresMountsFor= to cryptseup service files referencing a file, similar for devices
218 * cryptsetup-generator: allow specification of passwords in crypttab itself
220 * document that deps in [Unit] sections ignore Alias= fileds in
221 [Install] units of other units, unless those units are disabled
223 * instantiated [Install] for target units
224 https://bugs.freedesktop.org/show_bug.cgi?id=54377
226 * move debug shell to tty6 and make sure this doesn't break the gettys on tty6
228 * move cryptsetup key caching into kernel keyctl?
229 https://bugs.freedesktop.org/show_bug.cgi?id=54982
231 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
232 https://bugs.freedesktop.org/show_bug.cgi?id=54712
234 * after deserializing sockets in socket.c we should reapply sockopts and things
236 * make timer units go away after they elapsed
238 * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html
239 (network and remote-fs on shutdown)
241 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
243 * journald: add kernel cmdline option to disable ratelimiting for debug purposes
245 * move PID 1 segfaults to /var/lib/systemd/coredump?
247 * Document word splitting syntax for ExecStart= and friends
249 * create /sbin/init symlinks from the build system
251 * Query Paul Moore about relabelling socket fds while they are open
253 * journald: check whether it is OK if the client can still modify delivered journal entries
255 * journal live copy, based on libneon (client) and libmicrohttpd
257 * system-wide seccomp filter
259 * system.conf should have controls for cgroups
261 * bind mount read-only the cgroup tree higher than nspawn
263 * allow writing multiple conditions in unit files on one line
265 * explore multiple service instances per listening socket idea
267 * 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.
269 * shutdown: don't read-only mount anything when running in container
271 * nspawn: --read-only is not applied recursively to submounts
273 * MountFlags=shared acts as MountFlags=slave right now.
275 * ReadOnlyDirectories= is not applied recursively to submounts
277 * drop PID 1 reloading, only do reexecing (difficult: Reload()
278 currently is properly synchronous, Reexec() is weird, because we
279 can't delay the response properly until we are back, so instead of
280 being properly synchronous we just keep open the fd and close it
281 when done. That means clients don't get a successful method reply,
282 but much rather a disconnect on success.
284 * document that service reload may be implemented as service reexec
286 * remember which condition failed for services, not just the fact that something failed
288 * use opterr = 0 for all getopt tools
290 * properly handle loop back mounts via fstab, especially regards to fsck/passno
292 * allow services with no ExecStart= but with an ExecStop=
294 * add proper journal support to "systemctl --user status ..."
296 * add _SYSTEMD_USER_UNIT= field to journal entries
298 * dracut-shutdown needs to be ordered before unmounting /boot
300 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
302 * rename "userspace" to "core-os"
304 * systemctl: "Journal has been rotated since unit was started." message is misleading
306 * syscall filter: add knowledge about compat syscalls
308 * syscall filter: don't enforce no new privs?
310 * syscall filter: option to return EPERM rather than SIGSYS?
312 * syscall filter: port to libseccomp
314 * logind: wakelock/opportunistic suspend support
316 * systemd-analyze post-boot is broken for initrd
318 * systemd-analyze: data collection tools should be lightweight (few dependencies); data analysis tools can be heavyweight
320 * man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
322 * .device aliases need to be implemented with the "following" logic, probably.
324 * refuse taking lower-case variable names in sd_journal_send() and friends.
326 * load-fragment: when loading a unit file via a chain of symlinks
327 verify that it isn't masked via any of the names traversed.
329 * journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
331 * introduce Type=pid-file
333 * maybe allow services with ExecStop= set, but no ExecStart=?
335 * efi: implement /forcefsck as uefi variables thus not requiring file system altering to trigger a file system check
337 * efi: honor language efi variables for default language selection
339 * efi: honor timezone efi variables for default timezone selection
341 * efi: automatically mount EFI partition to /boot if no such entry exists in /etc/fstab and /boot is empty
342 gummiboot exports the EFI system partion (ESP) device:
343 /sys/firmware/efi/vars/LoaderDeviceIdentifier-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
344 Acpi(PNP0A03,0)/Pci(1F|2)/?/HD(Part1,Sig1FCBC57F-4BFC-4C2B-91A3-9C84FBCD9AF1)
345 '/' is the separator for the device path list
346 HD(Part1,Sig1FCBC57F-4BFC-4C2B-91A3-9C84FBCD9AF1) contains the GPT UUID of the ESP
348 * read the bootloader performance data (raw TSC) in systemd-analyze
349 /sys/firmware/efi/vars/LoaderTicksExec-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
351 /sys/firmware/efi/vars/LoaderTicksInit-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
353 /sys/firmware/efi/vars/LoaderTicksStartMenu-4a67b082-0a4c-41cf-b6c7-440b29bb8c4f/data
354 (only set if the menu was active)
356 * change Requires=basic.target to RequisiteOverride=basic.target
358 * support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
360 * nspawn: make use of device cgroup controller by default
362 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
364 * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
366 * readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
368 * move passno parsing to fstab generator
370 * improve !/proc/*/loginuid situation: make /proc/*/loginuid less dependent on CONFIG_AUDIT,
371 or use the users cgroup information when /proc/*/loginuid is not available.
373 * pam_systemd: try to get old session id from cgroup, if audit sessionid cannot be determined
375 * pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree
377 * maybe introduce ~/.config/locale.conf and apply it within PAM
379 * readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
381 * automount: implement expire:
382 - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD
383 - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD
384 - every timeout/4 (original autofs logic)
385 - blocking, needs a thread
387 - receive expire packet on pipe if kernel tells the timeout is over
389 - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD
390 - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns
392 * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel)
394 * don't delete /tmp/systemd-namespace-* before a process is gone down
396 * ExecOnFailure=/usr/bin/foo
398 * fedora: make sshd and pam_loginuid work in nspawn containers
400 * fix utmp for console logins in containers
402 * Add pretty name for seats in logind
404 * ConditionSecurity= should learn about IMA and SMACK
407 - remove all (misguided from day 1) userspace firmware_class handling
409 - unify utf8 validator code with shared/
411 - add trigger --subsystem-match=usb/usb_device device
413 * cleanup syslog 'priority' vs. 'level' wording
415 * dbus upstream still refers to dbus.target and shouldn't
417 * 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
419 * support container_ttys=
421 * introduce mix of BindTo and Requisite
423 * journalctl: show multiline log messages sanely, expand tabs, and show all valid utf8 messages
425 * add DeleteSocketsOnStop=yes|no option to socket units
427 * journal: store euid in journal if it differs from uid
429 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
431 * journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
433 * add command to systemctl to plot dependency graph as tree (see rhbz 795365)
435 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
437 * default unix qlen is too small (10). bump sysctl? add sockopt?
439 * figure out whether we should leave dbus around during shutdown
441 * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
443 * dbus: move dbus to early boot
445 * logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API
447 * journal: deal nicely with byte-by-byte copied files, especially regards header
449 * journal: local deserializer of export mode, http server
451 * document the exit codes when services fail before they are exec()ed
453 * save coredump in Windows/Mozilla minidump format
455 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
457 * clean up session cgroups that remain after logout (think sshd), but eventually run empty
459 * support "systemctl stop foobar@.service" to stop all units matching a certain template
461 * logind: allow showing logout dialog from system
463 * document that %% can be used to write % in a string that is specifier extended
465 * when an instanced service exits, remove its parent cgroup too if possible.
467 * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max
469 * be able to specify a forced restart of service A where service B depends on, in case B
470 needs to be auto-respawned?
472 * Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files"
474 * when a bus name of a service disappears from the bus make sure to queue further activation requests
476 * something like ConditionExec= or ExecCondition= without failure state
478 * tmpfiles: apply "x" on "D" too (see patch from William Douglas)
480 * don't set $HOME in services unless requested
482 * hide PAM/TCPWrap options in fragment parser when compile time disabled
484 * when we automatically restart a service, ensure we restart its rdeps, too.
486 * allow Type=simple with PIDFile=
487 https://bugzilla.redhat.com/show_bug.cgi?id=723942
489 * move PAM code into its own binary
491 * logind: spawn user@..service on login
493 * logind: non-local X11 server handling
495 * implement Register= switch in .socket units to enable registration
496 in Avahi, RPC and other socket registration services.
498 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
500 * readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO
502 * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination
504 * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1)
506 * verify that the AF_UNIX sockets of a service in the fs still exist
507 when we start a service in order to avoid confusion when a user
508 assumes starting a service is enough to make it accessible
510 * Make it possible to set the keymap independently from the font on
511 the kernel cmdline. Right now setting one resets also the other.
513 * and a dbus call to generate target from current state
515 * drop /.readahead on bigger upgrades with yum
517 * add support for /bin/mount -s
519 * GC unreferenced jobs (such as .device jobs)
521 * write blog stories about:
522 - hwdb: what belongs into it, lsusb
523 - enabling dbus services
525 - how to make changes to sysctl and sysfs attributes
527 - how to pass throw-away units to systemd, or dynamically change properties of existing units
528 - how to integrate cgconfig and suchlike with systemd
529 - testing with Harald's awesome test kit
531 - how to develop against journal browsing APIs
532 - the journal HTTP iface
533 - non-cgroup resource management
534 - dynamic resource management with cgroups
535 - refreshed, longer missions statement
536 - celendar time events
538 * allow port=0 in .socket units
540 * move readahead files into /var (look for them with .path units?)
542 * teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
544 * support systemd.mask= on the kernel command line.
546 * when key file cannot be found, read it from kbd in cryptsetup
548 * reuse mkdtemp namespace dirs in /tmp?
550 * recreate systemd's D-Bus private socket file on SIGUSR2
552 * Support --test based on current system state
554 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
556 * maybe introduce ExecRestartPre=
558 * configurable jitter for timer events
560 * timer events with system resume
562 * dot output for --test showing the 'initial transaction'
564 * writable cgroups dbus properties for live changes
566 * port over to LISTEN_FDS/LISTEN_PID:
567 - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH
571 - libvirtd (/var/run/libvirt/libvirt-sock-ro)
572 - bluetoothd (/var/run/sdp! @/org/bluez/audio!)
575 * fingerprint.target, wireless.target, gps.target, netdevice.target
577 * io priority during initialization
579 * systemctl list-jobs - show dependencies
581 * add systemctl switch to dump transaction without executing it
583 * drop cap bounding set in readahead and other services
589 - natively watch for dbus-*.service symlinks (PENDING)
590 - allow specification of socket mode/umask when allocating DBusServer
591 - allow disabling of fd passing when connecting a AF_UNIX connection
592 - allow disabling of UID passing for AUTH EXTERNAL
593 - always pass cred data along each message
595 * fix alsa mixer restore to not print error when no config is stored
597 * gnome-shell python script/glxinfo/is-accelerated must die
599 * make cryptsetup lower --iter-time
601 * patch kernel for xattr support in /dev, /proc/, /sys?
603 * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only
604 in an ~30 minutes window. It does not adjust larger differences. Find a way
605 to tell the kernel, to always do a full time sync when the RTC is in UTC and
606 we are in 11-minutes-mode. When we trust the system time to NTP we also want
609 * kernel: add device_type = "fb", "fbcon" to class "graphics"
613 * look for close() vs. close_nointr() vs. close_nointr_nofail()
615 * check for strerror(r) instead of strerror(-r)
617 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
619 * %m in printf() instead of strerror(errno);
623 * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()!
625 * use secure_getenv() instead of getenv() where appropriate
627 * link up selected blog stories from man pages and unit files Documentation= fields
629 Scheduled for removal or fixing:
631 * xxxOverridable dependencies (probably: fix)