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 * Dangling symlinks of .automount unit files in .wants/ directories, set up
7 automount points even when the original .automount file did not exist
8 anymore. Only the .mount unit was still around.
10 * make polkit checks async
12 * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point.
14 * add 'set -e' to scripts in test/
15 * make test in test/ work with separate output dir
17 * suppress log output on shutdown when "quiet" is used
20 * make anaconda write timeout=0 for encrypted devices
22 * drop no longer needed modprobe.d udlfb file (it does not belong in /etc anyway)
24 * Retest multi-seat (new kms driver for displaylink)
26 * make sure pkexec works fine with pam_systemd works fine with audit=0
28 * create /var/log/journal/
30 * logind: document new PrepareForSleep(false) semantics and UnlockSessions call in wiki
32 * external: maybe it is time to patch procps so that "ps" links to
33 libsystemd-logind to print a pretty service name, seat name, session
34 name in its output. Currently it only shows cgroup membership, but
35 that's sometimes kinda hard to parse for a human.
38 - don't unconditionally filter out duplicate settings
39 - support high-level cgroup setting syntax in systemctl. Example: "systemctl set-cgroup-attr MemoryLimit 5K"
40 - support writte string mapping even for non-high-level settings
41 - add man page for systemctl commands
42 - make sure we work fine with multi-line strings
44 * kernel cmdline switch to turn off predictable network interface names
46 * journal is not closed properly at shutdown when run in a container?
48 * introduce new "journal" group in place of adm?
51 - localectl: add listing support for X11 keymaps, by parsing /usr/share/X11/xkb/rules/xorg.lst
52 - localectl: support new converted x11→console keymaps
56 * nspawn: investigate whether we can support the same as LXC's
57 lxc.network.type=phys mode, and pass through entire network
58 interfaces to the container
60 * merge service_kill() logic into unit_kill_context() and unify the code paths
62 * when auto-detecting the main pid of a service, only look for those processes with ppid == 1
64 * introduce ExecCondition= in services
66 * if we have systemd-analyze in C "systemctl dot" should move there too
69 - fsck hookup for the ESP mount is missing
70 - write man page for efi boot generator
71 - add configure switch for enabling/disabling efi stuff
72 - honor language efi variables for default language selection (if there are any?)
73 - honor timezone efi variables for default timezone selection (if there are any?)
74 - introduce bootctl (backed by systemd-bootd) to control temporary and persistent default boot goal plus efi variables
76 * polkit: add configure switch to enable/disable polkit client side code
78 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
80 * re-enable "make check" for gtk-doc (broken for unknown reason)
82 * fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points
83 https://bugzilla.redhat.com/show_bug.cgi?id=812826
85 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
87 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
90 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
91 - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
92 - logind: wakelock/opportunistic suspend support
93 - Add pretty name for seats in logind
94 - logind: allow showing logout dialog from system?
95 - logind: spawn user@..service on login
96 - logind: non-local X11 server handling
97 - logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API
98 - pam_systemd: try to get old session id from cgroup, if audit sessionid cannot be determined
99 - pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree
101 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
103 * DeviceAllow/DeviceDeny: disallow everything by default, but whitelist /dev/zero, /dev/null and friends
105 * service: watchdog logic: for testing purposes allow ping, but do not require pong
108 - journald: also get thread ID from client, plus thread name
109 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
110 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
111 - fallback to /dev/log based logging in libsystemd-journal, if we can't log natively?
112 - declare the local journal protocol stable in the wiki interface chart
113 - journal: reuse XZ context
114 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
115 - journald: when dropping msgs due to ratelimit make sure to write
116 "dropped %u messages" not only when we are about to print the next
117 message that works, but alraedy after a short tiemout
118 - check if we can make journalctl by default use --follow mode inside of less if called without args?
119 - maybe add API to send pairs of iovecs via sd_journal_send
120 - journal: when writing journal auto-rotate if time jumps backwards
121 - gatewayd: should run under its own UID
122 - journal: add a setgid "adm" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
123 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
124 and !FOOBAR for events without FOOBAR.
125 - journal: when rotating, copy over old acls/access mode
126 - journal: document why we do not give ownership to journal files to the user that created them but use FS ACLs for that
127 - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs.
128 - 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
129 - 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.
130 - journal: find a way to allow dropping history early, based on priority, other rules
131 - journal: When used on NFS, check payload hashes
132 - journal: When used on NFS make sure wake up sd_journal_wait() every 2s, to handle missing inotify
133 - document that people can use file system ACLs to manage access to journal files, with example
134 - Introduce journalctl -b <nr> to show journal messages of a previous boot
135 - journald: check whether it is OK if the client can still modify delivered journal entries
136 - journal live copy, based on libneon (client) and libmicrohttpd
137 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
138 - refuse taking lower-case variable names in sd_journal_send() and friends.
139 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
140 - journal: deal nicely with byte-by-byte copied files, especially regards header
141 - journalctl: show multiline log messages sanely, expand tabs, and show all valid utf8 messages
142 - journal: store euid in journal if it differs from uid
143 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
146 - document unit_name_mangle()
147 - document that deps in [Unit] sections ignore Alias= fileds in
148 [Install] units of other units, unless those units are disabled
149 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
150 - Document word splitting syntax for ExecStart= and friends
151 - document that units from /etc override those from /usr and /run
152 - document the exit codes when services fail before they are exec()ed
153 - document that %% can be used to write % in a string that is specifier extended
154 - document that service reload may be implemented as service reexec
157 - systemctl list-jobs - show dependencies
158 - add systemctl switch to dump transaction without executing it
159 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
160 - "systemctl disable" on a static unit prints no message and does
161 nothing. "systemctl enable" does nothing, and gives a bad message
162 about it. Should fix both to print nice actionable messages.
163 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
164 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
165 - systemctl enable: improve the success messages (i.e. more human readable, less shell-like)
166 - systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards?
167 - systemctl: "Journal has been rotated since unit was started." message is misleading
168 - support "systemctl stop foobar@.service" to stop all units matching a certain template
169 - Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files"
171 * When shutdown.target is queued begin with an asynchronous sync()?
173 * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations
175 * deal with sendmail/postfix exclusivity
178 - configurable jitter for timer events
179 - timer events with system resume
180 - timer units should get the ability to trigger when:
181 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
184 * update the kernel's TZ (sys_tz) when DST changes
186 * sync down the system time to the RTC when:
187 - CLOCK_REALTIME makes jumps (the user explicitely requested a time set)
188 - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp)
189 This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime
190 mode, it will never touch the RTC if the no reliable time source is active or the
191 user did not request anything like it.
194 - implement conditional properties (dmi matches)
195 - hwdb --filter=ID_DRIVE_*
197 * if booted in "quiet" mode, and an error happens, turn on status
198 output again, so that the emergency mode isn't totally
199 surprising. Also, terminate plymouth.
201 * libunwind support for coredump pattern hook, and includes this in
202 the message for coredumps. After all, libunwind is now capable to
203 unwind coredumps since a few weeks ago. This probably requires that
204 we have nice support for multi-line messages on display in
205 logs-show.c. Alternatively: use libelfutil, which seems to be the
206 better supported alternative.
208 * figure out relation of --all and --full in the various tools
210 * add libsystemd-password or so to query passwords during boot using the password agent logic
212 * 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().
214 * fedup: add --unit to systemctl switch-root somehow
215 * fedup: don't delete initrd on switch-root
219 - timedated: export boolean that clarifies whether NTP is even available
220 - timedated: refuse time changes when NTP is on
222 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
224 * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields
226 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
228 * add "provisioning" instructions to setup an empty /etc + /var
229 - used to setup a new container from a shared /usr
230 - superset of tmpfiles model
231 - instructions shipped by packages and stored in /usr/lib/
232 - compose /etc/passwd and /etc/group, copy files
233 - able to create uid + gid used by packages, for file ownership
235 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
236 waiting for the timeout
238 * high level net_prio setting in execution context
240 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
242 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't
244 * WorkingDirectory: support env var replacements like in ExecStart= so that people can use $HOME
246 * refuse boot if /etc/machine-id is not useful (or set taint?)
249 - nspawn: consider changing users for -u with su, so that NSS resolving works correctly
250 - nspawn: implement personality changes a la linux32(8)
251 - nspawn: --read-only is not applied recursively to submounts
252 - nspawn: make use of device cgroup controller by default
253 - bind mount read-only the cgroup tree higher than nspawn
256 - cryptsetup-generator: warn if the password files are world-readable
257 - cryptsetup-generator: add RequiresMountsFor= to cryptseup service files referencing a file, similar for devices
258 - cryptsetup-generator: allow specification of passwords in crypttab itself
259 - move cryptsetup key caching into kernel keyctl?
260 https://bugs.freedesktop.org/show_bug.cgi?id=54982
261 - when key file cannot be found, read it from kbd in cryptsetup
263 * instantiated [Install] for target units
264 https://bugs.freedesktop.org/show_bug.cgi?id=54377
266 * move debug shell to tty6 and make sure this doesn't break the gettys on tty6
268 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
269 https://bugs.freedesktop.org/show_bug.cgi?id=54712
271 * after deserializing sockets in socket.c we should reapply sockopts and things
273 * make timer units go away after they elapsed
275 * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html
276 (network and remote-fs on shutdown)
278 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
280 * move PID 1 segfaults to /var/lib/systemd/coredump?
282 * create /sbin/init symlinks from the build system
284 * Query Paul Moore about relabelling socket fds while they are open
286 * system.conf should have controls for cgroups
288 * allow writing multiple conditions in unit files on one line
290 * explore multiple service instances per listening socket idea
292 * 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.
294 * shutdown: don't read-only mount anything when running in container
296 * MountFlags=shared acts as MountFlags=slave right now.
298 * ReadOnlyDirectories= is not applied recursively to submounts
300 * drop PID 1 reloading, only do reexecing (difficult: Reload()
301 currently is properly synchronous, Reexec() is weird, because we
302 can't delay the response properly until we are back, so instead of
303 being properly synchronous we just keep open the fd and close it
304 when done. That means clients don't get a successful method reply,
305 but much rather a disconnect on success.
307 * remember which condition failed for services, not just the fact that something failed
309 * use opterr = 0 for all getopt tools
311 * properly handle loop back mounts via fstab, especially regards to fsck/passno
313 * allow services with no ExecStart= but with an ExecStop=
315 * dracut-shutdown needs to be ordered before unmounting /boot
317 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
319 * rename "userspace" to "core-os"
322 - syscall filter: add knowledge about compat syscalls
323 - syscall filter: don't enforce no new privs?
324 - syscall filter: option to return EPERM rather than SIGSYS?
325 - syscall filter: port to libseccomp
326 - system-wide seccomp filter
328 * systemd-analyze: data collection tools should be lightweight (few dependencies); data analysis tools can be heavyweight
330 * .device aliases need to be implemented with the "following" logic, probably.
332 * load-fragment: when loading a unit file via a chain of symlinks
333 verify that it isn't masked via any of the names traversed.
335 * introduce Type=pid-file
337 * maybe allow services with ExecStop= set, but no ExecStart=?
339 * change Requires=basic.target to RequisiteOverride=basic.target
341 * support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
343 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
345 * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
347 * move passno parsing to fstab generator
349 * automount: implement expire:
350 - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD
351 - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD
352 - every timeout/4 (original autofs logic)
353 - blocking, needs a thread
355 - receive expire packet on pipe if kernel tells the timeout is over
357 - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD
358 - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns
360 * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel)
362 * ExecOnFailure=/usr/bin/foo
364 * ConditionSecurity= should learn about IMA and SMACK
367 - remove all (misguided from day 1) userspace firmware_class handling
369 - unify utf8 validator code with shared/
371 - add trigger --subsystem-match=usb/usb_device device
373 * cleanup syslog 'priority' vs. 'level' wording
375 * dbus upstream still refers to dbus.target and shouldn't
377 * 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
379 * support container_ttys=
381 * introduce mix of BindTo and Requisite
383 * add DeleteSocketsOnStop=yes|no option to socket units
385 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
387 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
389 * default unix qlen is too small (10). bump sysctl? add sockopt?
391 * figure out whether we should leave dbus around during shutdown
393 * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
395 * dbus: move dbus to early boot
397 * save coredump in Windows/Mozilla minidump format
399 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
401 * clean up session cgroups that remain after logout (think sshd), but eventually run empty
403 * when an instanced service exits, remove its parent cgroup too if possible.
405 * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max
407 * be able to specify a forced restart of service A where service B depends on, in case B
408 needs to be auto-respawned?
410 * when a bus name of a service disappears from the bus make sure to queue further activation requests
412 * tmpfiles: apply "x" on "D" too (see patch from William Douglas)
414 * for services: don't set $HOME in services unless requested
416 * hide PAM/TCPWrap options in fragment parser when compile time disabled
418 * when we automatically restart a service, ensure we restart its rdeps, too.
420 * allow Type=simple with PIDFile=
421 https://bugzilla.redhat.com/show_bug.cgi?id=723942
423 * move PAM code into its own binary
425 * implement Register= switch in .socket units to enable registration
426 in Avahi, RPC and other socket registration services.
428 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
430 * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination
432 * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1)
434 * verify that the AF_UNIX sockets of a service in the fs still exist
435 when we start a service in order to avoid confusion when a user
436 assumes starting a service is enough to make it accessible
438 * Make it possible to set the keymap independently from the font on
439 the kernel cmdline. Right now setting one resets also the other.
441 * and a dbus call to generate target from current state
444 - drop /.readahead on bigger upgrades with yum
445 - move readahead files into /var (look for them with .path units?)
446 - readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO
447 - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
448 - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
450 * add support for /bin/mount -s
452 * GC unreferenced jobs (such as .device jobs)
454 * write blog stories about:
455 - hwdb: what belongs into it, lsusb
456 - enabling dbus services
458 - how to make changes to sysctl and sysfs attributes
460 - how to pass throw-away units to systemd, or dynamically change properties of existing units
461 - how to integrate cgconfig and suchlike with systemd
462 - testing with Harald's awesome test kit
464 - how to develop against journal browsing APIs
465 - the journal HTTP iface
466 - non-cgroup resource management
467 - dynamic resource management with cgroups
468 - refreshed, longer missions statement
469 - celendar time events
471 * allow port=0 in .socket units
473 * support systemd.mask= on the kernel command line.
475 * reuse mkdtemp namespace dirs in /tmp?
477 * recreate systemd's D-Bus private socket file on SIGUSR2
479 * Support --test based on current system state
481 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
483 * maybe introduce ExecRestartPre=
485 * dot output for --test showing the 'initial transaction'
487 * port over to LISTEN_FDS/LISTEN_PID:
488 - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH
492 - libvirtd (/var/run/libvirt/libvirt-sock-ro)
493 - bluetoothd (/var/run/sdp! @/org/bluez/audio!)
496 * fingerprint.target, wireless.target, gps.target, netdevice.target
498 * io priority during initialization
500 * drop cap bounding set in readahead and other services
506 - natively watch for dbus-*.service symlinks (PENDING)
507 - allow specification of socket mode/umask when allocating DBusServer
508 - allow disabling of fd passing when connecting a AF_UNIX connection
509 - allow disabling of UID passing for AUTH EXTERNAL
510 - always pass cred data along each message
511 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
513 * fix alsa mixer restore to not print error when no config is stored
515 * make cryptsetup lower --iter-time
517 * patch kernel for xattr support in /dev, /proc/, /sys?
519 * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only
520 in an ~30 minutes window. It does not adjust larger differences. Find a way
521 to tell the kernel, to always do a full time sync when the RTC is in UTC and
522 we are in 11-minutes-mode. When we trust the system time to NTP we also want
525 * kernel: add device_type = "fb", "fbcon" to class "graphics"
529 * look for close() vs. close_nointr() vs. close_nointr_nofail()
531 * check for strerror(r) instead of strerror(-r)
533 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
535 * %m in printf() instead of strerror(errno);
539 * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()!
541 * use secure_getenv() instead of getenv() where appropriate
543 * link up selected blog stories from man pages and unit files Documentation= fields
545 Scheduled for removal or fixing:
547 * xxxOverridable dependencies (probably: fix)