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.
20 * create /var/log/journal/
22 * logind: document new PrepareForSleep(false) semantics and UnlockSessions call in wiki
24 * external: maybe it is time to patch procps so that "ps" links to
25 libsystemd-logind to print a pretty service name, seat name, session
26 name in its output. Currently it only shows cgroup membership, but
27 that's sometimes kinda hard to parse for a human.
30 - don't unconditionally filter out duplicate settings
31 - support high-level cgroup setting syntax in systemctl. Example: "systemctl set-cgroup-attr MemoryLimit 5K"
32 - support writte string mapping even for non-high-level settings
33 - add man page for systemctl commands
34 - make sure we work fine with multi-line strings
36 * kernel cmdline switch to turn off predictable network interface names
38 * journal is not closed properly at shutdown when run in a container?
40 * introduce new "journal" group in place of adm?
44 * introduce ExecCondition= in services
46 * unify killing logic of service, socket, mount, swap units
48 * if we have systemd-analyze in C "systemctl dot" should move there too
51 - fsck hookup for the ESP mount is missing
52 - write man page for efi boot generator
53 - add configure switch for enabling/disabling efi stuff
54 - honor language efi variables for default language selection (if there are any?)
55 - honor timezone efi variables for default timezone selection (if there are any?)
56 - introduce bootctl (backed by systemd-bootd) to control temporary and persistent default boot goal plus efi variables
58 * maybe not install getty@tty1.service symlink in /etc but in /usr?
60 * re-enable "make check" for gtk-doc (broken for unknown reason)
62 * fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points
63 https://bugzilla.redhat.com/show_bug.cgi?id=812826
65 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
67 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
70 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
71 - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
72 - logind: wakelock/opportunistic suspend support
73 - Add pretty name for seats in logind
74 - logind: allow showing logout dialog from system?
75 - logind: spawn user@..service on login
76 - logind: non-local X11 server handling
77 - logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API
78 - pam_systemd: try to get old session id from cgroup, if audit sessionid cannot be determined
79 - pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree
81 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
83 * DeviceAllow/DeviceDeny: disallow everything by default, but whitelist /dev/zero, /dev/null and friends
85 * service: watchdog logic: for testing purposes allow ping, but do not require pong
88 - journald: also get thread ID from client, plus thread name
89 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
90 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
91 - fallback to /dev/log based logging in libsystemd-journal, if we can't log natively?
92 - declare the local journal protocol stable in the wiki interface chart
93 - journal: reuse XZ context
94 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
95 - journald: when dropping msgs due to ratelimit make sure to write
96 "dropped %u messages" not only when we are about to print the next
97 message that works, but alraedy after a short tiemout
98 - check if we can make journalctl by default use --follow mode inside of less if called without args?
99 - maybe add API to send pairs of iovecs via sd_journal_send
100 - journal: when writing journal auto-rotate if time jumps backwards
101 - gatewayd: should run under its own UID
102 - journal: add a setgid "adm" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
103 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
104 and !FOOBAR for events without FOOBAR.
105 - journal: when rotating, copy over old acls/access mode
106 - journal: document why we do not give ownership to journal files to the user that created them but use FS ACLs for that
107 - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs.
108 - 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
109 - 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.
110 - journal: find a way to allow dropping history early, based on priority, other rules
111 - journal: When used on NFS, check payload hashes
112 - 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
114 - Introduce journalctl -b <nr> to show journal messages of a previous boot
115 - journald: check whether it is OK if the client can still modify delivered journal entries
116 - journal live copy, based on libneon (client) and libmicrohttpd
117 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
118 - refuse taking lower-case variable names in sd_journal_send() and friends.
119 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
120 - journal: deal nicely with byte-by-byte copied files, especially regards header
121 - journalctl: show multiline log messages sanely, expand tabs, and show all valid utf8 messages
122 - journal: store euid in journal if it differs from uid
123 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
126 - document unit_name_mangle()
127 - document that deps in [Unit] sections ignore Alias= fileds in
128 [Install] units of other units, unless those units are disabled
129 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
130 - Document word splitting syntax for ExecStart= and friends
131 - document that units from /etc override those from /usr and /run
132 - document the exit codes when services fail before they are exec()ed
133 - document that %% can be used to write % in a string that is specifier extended
134 - document that service reload may be implemented as service reexec
137 - systemctl list-jobs - show dependencies
138 - add systemctl switch to dump transaction without executing it
139 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
140 - "systemctl disable" on a static unit prints no message and does
141 nothing. "systemctl enable" does nothing, and gives a bad message
142 about it. Should fix both to print nice actionable messages.
143 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
144 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
145 - systemctl enable: improve the success messages (i.e. more human readable, less shell-like)
146 - systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards?
147 - systemctl: "Journal has been rotated since unit was started." message is misleading
148 - support "systemctl stop foobar@.service" to stop all units matching a certain template
149 - Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files"
151 * When shutdown.target is queued begin with an asynchronous sync()?
153 * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations
155 * deal with sendmail/postfix exclusivity
158 - configurable jitter for timer events
159 - timer events with system resume
160 - timer units should get the ability to trigger when:
161 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
164 * update the kernel's TZ (sys_tz) when DST changes
166 * sync down the system time to the RTC when:
167 - CLOCK_REALTIME makes jumps (the user explicitely requested a time set)
168 - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp)
169 This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime
170 mode, it will never touch the RTC if the no reliable time source is active or the
171 user did not request anything like it.
174 - implement conditional properties (dmi matches)
175 - hwdb --filter=ID_DRIVE_*
177 * if booted in "quiet" mode, and an error happens, turn on status
178 output again, so that the emergency mode isn't totally
179 surprising. Also, terminate plymouth.
182 - localectl: add listing support for X11 keymaps, by parsing /usr/share/X11/xkb/rules/xorg.lst
183 - localectl: support new converted x11→console keymaps
185 * libunwind support for coredump pattern hook, and includes this in
186 the message for coredumps. After all, libunwind is now capable to
187 unwind coredumps since a few weeks ago. This probably requires that
188 we have nice support for multi-line messages on display in
189 logs-show.c. Alternatively: use libelfutil, which seems to be the
190 better supported alternative.
192 * figure out relation of --all and --full in the various tools
194 * add libsystemd-password or so to query passwords during boot using the password agent logic
196 * 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().
198 * fedup: add --unit to systemctl switch-root somehow
199 * fedup: don't delete initrd on switch-root
203 - timedated: export boolean that clarifies whether NTP is even available
204 - timedated: refuse time changes when NTP is on
206 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
208 * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields
210 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
212 * add "provisioning" instructions to setup an empty /etc + /var
213 - used to setup a new container from a shared /usr
214 - superset of tmpfiles model
215 - instructions shipped by packages and stored in /usr/lib/
216 - compose /etc/passwd and /etc/group, copy files
217 - able to create uid + gid used by packages, for file ownership
219 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
220 waiting for the timeout
222 * high level net_prio setting in execution context
224 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
226 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't
228 * WorkingDirectory: support env var replacements like in ExecStart= so that people can use $HOME
230 * refuse boot if /etc/machine-id is not useful (or set taint?)
233 - nspawn: consider changing users for -u with su, so that NSS resolving works correctly
234 - nspawn: implement personality changes a la linux32(8)
235 - nspawn: --read-only is not applied recursively to submounts
236 - nspawn: make use of device cgroup controller by default
237 - bind mount read-only the cgroup tree higher than nspawn
240 - cryptsetup-generator: warn if the password files are world-readable
241 - cryptsetup-generator: add RequiresMountsFor= to cryptseup service files referencing a file, similar for devices
242 - cryptsetup-generator: allow specification of passwords in crypttab itself
243 - move cryptsetup key caching into kernel keyctl?
244 https://bugs.freedesktop.org/show_bug.cgi?id=54982
245 - when key file cannot be found, read it from kbd in cryptsetup
247 * instantiated [Install] for target units
248 https://bugs.freedesktop.org/show_bug.cgi?id=54377
250 * move debug shell to tty6 and make sure this doesn't break the gettys on tty6
252 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
253 https://bugs.freedesktop.org/show_bug.cgi?id=54712
255 * after deserializing sockets in socket.c we should reapply sockopts and things
257 * make timer units go away after they elapsed
259 * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html
260 (network and remote-fs on shutdown)
262 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
264 * move PID 1 segfaults to /var/lib/systemd/coredump?
266 * create /sbin/init symlinks from the build system
268 * Query Paul Moore about relabelling socket fds while they are open
270 * system.conf should have controls for cgroups
272 * allow writing multiple conditions in unit files on one line
274 * explore multiple service instances per listening socket idea
276 * 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.
278 * shutdown: don't read-only mount anything when running in container
280 * MountFlags=shared acts as MountFlags=slave right now.
282 * ReadOnlyDirectories= is not applied recursively to submounts
284 * drop PID 1 reloading, only do reexecing (difficult: Reload()
285 currently is properly synchronous, Reexec() is weird, because we
286 can't delay the response properly until we are back, so instead of
287 being properly synchronous we just keep open the fd and close it
288 when done. That means clients don't get a successful method reply,
289 but much rather a disconnect on success.
291 * remember which condition failed for services, not just the fact that something failed
293 * use opterr = 0 for all getopt tools
295 * properly handle loop back mounts via fstab, especially regards to fsck/passno
297 * allow services with no ExecStart= but with an ExecStop=
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"
306 - syscall filter: add knowledge about compat syscalls
307 - syscall filter: don't enforce no new privs?
308 - syscall filter: option to return EPERM rather than SIGSYS?
309 - syscall filter: port to libseccomp
310 - system-wide seccomp filter
312 * systemd-analyze: data collection tools should be lightweight (few dependencies); data analysis tools can be heavyweight
314 * .device aliases need to be implemented with the "following" logic, probably.
316 * load-fragment: when loading a unit file via a chain of symlinks
317 verify that it isn't masked via any of the names traversed.
319 * introduce Type=pid-file
321 * maybe allow services with ExecStop= set, but no ExecStart=?
323 * change Requires=basic.target to RequisiteOverride=basic.target
325 * support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
327 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
329 * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
331 * move passno parsing to fstab generator
333 * automount: implement expire:
334 - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD
335 - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD
336 - every timeout/4 (original autofs logic)
337 - blocking, needs a thread
339 - receive expire packet on pipe if kernel tells the timeout is over
341 - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD
342 - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns
344 * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel)
346 * ExecOnFailure=/usr/bin/foo
348 * ConditionSecurity= should learn about IMA and SMACK
351 - remove all (misguided from day 1) userspace firmware_class handling
353 - unify utf8 validator code with shared/
355 - add trigger --subsystem-match=usb/usb_device device
357 * cleanup syslog 'priority' vs. 'level' wording
359 * dbus upstream still refers to dbus.target and shouldn't
361 * 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
363 * support container_ttys=
365 * introduce mix of BindTo and Requisite
367 * add DeleteSocketsOnStop=yes|no option to socket units
369 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
371 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
373 * default unix qlen is too small (10). bump sysctl? add sockopt?
375 * figure out whether we should leave dbus around during shutdown
377 * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
379 * dbus: move dbus to early boot
381 * save coredump in Windows/Mozilla minidump format
383 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
385 * clean up session cgroups that remain after logout (think sshd), but eventually run empty
387 * when an instanced service exits, remove its parent cgroup too if possible.
389 * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max
391 * be able to specify a forced restart of service A where service B depends on, in case B
392 needs to be auto-respawned?
394 * when a bus name of a service disappears from the bus make sure to queue further activation requests
396 * tmpfiles: apply "x" on "D" too (see patch from William Douglas)
398 * for services: don't set $HOME in services unless requested
400 * hide PAM/TCPWrap options in fragment parser when compile time disabled
402 * when we automatically restart a service, ensure we restart its rdeps, too.
404 * allow Type=simple with PIDFile=
405 https://bugzilla.redhat.com/show_bug.cgi?id=723942
407 * move PAM code into its own binary
409 * implement Register= switch in .socket units to enable registration
410 in Avahi, RPC and other socket registration services.
412 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
414 * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination
416 * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1)
418 * verify that the AF_UNIX sockets of a service in the fs still exist
419 when we start a service in order to avoid confusion when a user
420 assumes starting a service is enough to make it accessible
422 * Make it possible to set the keymap independently from the font on
423 the kernel cmdline. Right now setting one resets also the other.
425 * and a dbus call to generate target from current state
428 - drop /.readahead on bigger upgrades with yum
429 - move readahead files into /var (look for them with .path units?)
430 - readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO
431 - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
432 - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
434 * add support for /bin/mount -s
436 * GC unreferenced jobs (such as .device jobs)
438 * write blog stories about:
439 - hwdb: what belongs into it, lsusb
440 - enabling dbus services
442 - how to make changes to sysctl and sysfs attributes
444 - how to pass throw-away units to systemd, or dynamically change properties of existing units
445 - how to integrate cgconfig and suchlike with systemd
446 - testing with Harald's awesome test kit
448 - how to develop against journal browsing APIs
449 - the journal HTTP iface
450 - non-cgroup resource management
451 - dynamic resource management with cgroups
452 - refreshed, longer missions statement
453 - celendar time events
455 * allow port=0 in .socket units
457 * support systemd.mask= on the kernel command line.
459 * reuse mkdtemp namespace dirs in /tmp?
461 * recreate systemd's D-Bus private socket file on SIGUSR2
463 * Support --test based on current system state
465 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
467 * maybe introduce ExecRestartPre=
469 * dot output for --test showing the 'initial transaction'
471 * port over to LISTEN_FDS/LISTEN_PID:
472 - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH
476 - libvirtd (/var/run/libvirt/libvirt-sock-ro)
477 - bluetoothd (/var/run/sdp! @/org/bluez/audio!)
480 * fingerprint.target, wireless.target, gps.target, netdevice.target
482 * io priority during initialization
484 * drop cap bounding set in readahead and other services
490 - natively watch for dbus-*.service symlinks (PENDING)
491 - allow specification of socket mode/umask when allocating DBusServer
492 - allow disabling of fd passing when connecting a AF_UNIX connection
493 - allow disabling of UID passing for AUTH EXTERNAL
494 - always pass cred data along each message
495 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
497 * fix alsa mixer restore to not print error when no config is stored
499 * make cryptsetup lower --iter-time
501 * patch kernel for xattr support in /dev, /proc/, /sys?
503 * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only
504 in an ~30 minutes window. It does not adjust larger differences. Find a way
505 to tell the kernel, to always do a full time sync when the RTC is in UTC and
506 we are in 11-minutes-mode. When we trust the system time to NTP we also want
509 * kernel: add device_type = "fb", "fbcon" to class "graphics"
513 * look for close() vs. close_nointr() vs. close_nointr_nofail()
515 * check for strerror(r) instead of strerror(-r)
517 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
519 * %m in printf() instead of strerror(errno);
523 * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()!
525 * use secure_getenv() instead of getenv() where appropriate
527 * link up selected blog stories from man pages and unit files Documentation= fields
529 Scheduled for removal or fixing:
531 * xxxOverridable dependencies (probably: fix)