2 * enabling an instance unit creates a pointless link, and
3 the unit will be started with getty@getty.service:
4 $ systemctl enable getty@.service
5 ln -s '/usr/lib/systemd/system/getty@.service' '/etc/systemd/system/getty.target.wants/getty@.service'
7 * check systemd-tmpfiles for selinux context hookup for mknod(), symlink() and similar
9 * swap units that are activated by one name but shown in the kernel under another are semi-broken
11 * Dangling symlinks of .automount unit files in .wants/ directories, set up
12 automount points even when the original .automount file did not exist
13 anymore. Only the .mount unit was still around.
15 * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point.
17 * When we detect invalid UTF-8, we cannot use it in an error message:
18 log...("Path is not UTF-8 clean, ignoring assignment: %s", rvalue);
20 * shorten the message to sane length:
22 Cannot add dependency job for unit display-manager.service, ignoring: Unit display-manager.service failed to load: No such file or directory. See system logs and 'systemctl status display-manager.service' for details.
26 * external: ps should gain colums for slice and machine
29 - localectl: support new converted x11→console keymaps
31 * when installing fedora with yum --installroot /var/run is a directory, not a symlink
32 https://bugzilla.redhat.com/show_bug.cgi?id=975864
34 CGroup Rework Completion:
36 * implement system-wide DefaultCPUAccounting=1 switch (and similar for blockio, memory?)
38 * implement per-slice CPUFairScheduling=1 switch
40 * handle jointly mounted controllers correctly
42 * introduce high-level settings for RT budget, swappiness
46 * sd-bus: synthesized messages should get serial number (uint32_t) -1
48 * sd-event: allow multiple signal handlers per signal
50 * timer: expose accuracy as unit setting
52 * when we detect low battery and no AC on boot, show pretty splash and refuse boot
54 * move libasyncns into systemd as libsystemd-asyncns
56 * calendarspec: support value ranges with ".." notation. Example: 2013-4..8-1
59 - restore selinux access control on properties
61 * sd-bus: when triggering property change events, allow a NULL strv indicate that all properties listed as such are send out as changed
63 * sd-bus: enforce signatures on response messages
65 * sd-bus: make message handlers take an sd_bus_error and generate error replies automatically if they are set
67 * sd-bus: see if we can drop more message validation on the sending side
69 * sd-bus: introduce sd_bus_creds object and attach it to messages as well as allow querying it for names
71 * sd-bus: support "const" properties as flag
73 * sd-event: when a handler returns an error, just turn off its event
74 source, but do not return anything up to the event loop
75 caller. Instead add parameter to sd_event_request_quit() to take
76 retval. This way errors rippling upwards are the option, not the
79 * sd-event: child pid handling: first invoke waitid(WNOHANG) and call event handler, only afterwards reap the process
81 * sd-event: native support for watchdog stuff
83 * machined, localed: when we try to kill an empty cgroup, generate an ESRCH call over the bus
85 * sd-bus: SD_BUS_COMMENT() macro for inclusion in vtables, syntax inspired by gdbus
87 * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
89 * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
91 * check :no-sender logic after PID 1 conversion
93 * increase journal files by a few MB each time, instead of piecemeal
95 * add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
97 * systemctl: rework wait filter to not require match callback
99 * unify dispatch table in systemctl_main() and friends
101 * bus: access policy as vtable flag
103 * journalctl: support -M to read journal of containers and determine journal directory from root directory of container
105 * tmpfiles: to make sure we don't delete unpacked tarballs with old timestamps right-away never delete stuff that is inside a directory with a new mtime
107 * "systemctl mask" should find all names by which a unit is accessible
108 (i.e. by scanning for symlinks to it) and link them all to /dev/null
110 * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk?
112 * remove NSS usage from PID 1 (notably the specifiers)
115 - Support multiple inherited sockets mapped to different remote hosts
116 - Use a nonblocking alternative to getaddrinfo
117 - Until we can start daemons directly, find a less ugly, less racy alternative than shell scripts for the second man page example.
118 - Support starting daemons directly without requiring a shell script; update man pages
120 * optimize the cgroup propagation bits, especially unit_get_members_mask(), cgroup_context_get_mask()
122 * "systemctl cat" or "systemctl view" command or or so, that cats the backing unit file of a service, plus its drop-ins and shows them in a pager
124 * rfkill,backlight: we probably should run the load tools inside of the udev rules so that the state is properly initialized by the time other software sees it
126 * Add a new Distribute=$NUMBER key to socket units that makes use of SO_REUSEPORT to distribute network traffic on $NUMBER instances
128 * tmpfiles: when applying ownership to /run/log/journal, also do this for the journal fails contained in it
130 * we probably should replace the left-over uses of strv_append() and replace them by strv_push() or strv_extend()
132 * move config_parse_path_strv() out of conf-parser.c
134 * libdsystemd-bus should expose utf8 validation calls
136 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
138 * If we try to find a unit via a dangling symlink, generate a clean
139 error. Currently, we just ignore it and read the unit from the search
142 * When a Type=forking service fails and needed another service, that
143 service is not cleaned up again when it has StopWhenUnneeded=yes
144 http://lists.freedesktop.org/archives/systemd-devel/2013-July/012141.html
146 * refuse boot if /etc/os-release is missing or /etc/machine-id cannot be set up
148 * ensure scope units may be started only a single time
150 * better error message if you run systemctl without systemd running
152 * systemctl status output should should include list of triggering units and their status
154 * for transient units, instead of writing out drop-ins for all properties consider serializing them in the normal serialization stream
156 * logind: when logging out, remove user-owned sysv and posix IPC objects
158 * session scopes/user unit: add RequiresMountsFor for the home directory of the user
160 * add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
162 * journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
164 * given that logind/machined now let PID 1 do all nasty work, we can
165 probably reduce the capability set they retain substantially.
167 * btrfs raid assembly: some .device jobs stay stuck in the queue
169 * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
171 * Fedora: post FPC ticket to move add %tmpfiles_create to the packaging guidelines
173 * make sure gdm doesn't use multi-user-x but the new default X configuration file, and then remove multi-user-x from systemd
175 * when parsing calendar timestamps support the UTC timezone (even if we won't support arbitrary timezone specs, support UTC itself certainly makes sense), also support syntaxes such as +0200
177 * when a kernel driver logs in a tight loop, we should ratelimit that too.
179 * "systemctl disable" of a unit instance removes all symlinks, but should
180 only remove the instance symlink (systemctl disable of a template
181 unit however should remove them all).
183 * journald: optionally, log debug messages to /run but everything else to /var
185 * systemctl list-unit-files should list generated files (and probably with a new state "generated" for them, or so)
187 * journald: when we drop syslog messages because the syslog socket is
188 full, make sure to write how many messages are lost as first thing
189 to syslog when it works again.
191 * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
193 * load .d/*.conf dropins for device units
195 * service_coldplug() appears to reinstall the wrong stop timeout watch.
197 * transient units: allow creating auxiliary units with the same call
199 * how to reset dynamically changed attributes sanely?
201 * when reloading configuration, apply new cgroup configuration
203 * journald: make sure ratelimit is actually really per-service with the new cgroup changes
205 * gparted needs to disable auto-activation of mount units somehow, or
206 maybe we should stop doing auto-activation of this after boot
207 entirely. https://bugzilla.gnome.org/show_bug.cgi?id=701676
208 Maybe take a BSD lock at the disk device node and teach udev to
209 check for that and suppress event handling.
211 * when recursively showing the cgroup hierarchy, optionally also show
212 the hierarchies of child processes
214 * document logic of auto/noauto and fail/nofail in fstab in systemd.mount or systemd-fstab-generator man page
216 * something pulls in pcre as shared object dependency into our daemons such as hostnamed.
218 * document systemd-journal-flush.service properly
220 * change systemd-journal-flush into a service that stays around during
221 boot, and causes the journal to be moved back to /run on shutdown,
222 so that we don't keep /var busy. This needs to happen synchronously,
223 hence doing this via signals is not going to work.
225 * allow implementation of InaccessibleDirectories=/ plus
226 ReadOnlyDirectories=... for whitelisting files for a service.
229 - default policy (allow uid == 0 and our own uid)
230 - enforce alignment of pointers passed in
231 - when kdbus doesn't take our message without memfds, try again with memfds
232 - implement translator service
233 - port systemd to new library
234 - implement busname unit type in systemd
236 - merge busctl into systemctl or so?
237 - synthesize sd_bus_message objects from kernel messages
238 - properly implement name registry ioctls for kdbus
239 - implement monitor logic
240 - properly map matches with well-known names against messages with unique names
243 * priority inheritance
245 * in the final killing spree, detect processes from the root directory, and
246 complain loudly if they have argv[0][0] == '@' set.
247 https://bugzilla.redhat.com/show_bug.cgi?id=961044
249 * add an option to nspawn that uses seccomp to make socket(AF_NETLINK,
250 SOCK_RAW, NETLINK_AUDIT) fail the the appropriate error code that
251 makes the audit userspace to think auditing is not available in the
254 * Introduce a way how we can kill the main process of a service with KillSignal, but all processes with SIGKILL later on
255 https://bugzilla.redhat.com/show_bug.cgi?id=952634
257 * maybe add a warning to the unit file parses where the access mode of unit files is nonsensical.
259 * investigate endianness issues of UUID vs. GUID
261 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
262 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
264 * if pam_systemd is invoked by su from a process that is outside of a
265 any session we should probably just become a NOP, since that's
266 usually not a real user session but just some system code that just
269 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
271 * add a pam module that on password changes updates any LUKS slot where the password matches
273 * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases...
275 * timedatctl, localectl: possibly make some commands work without the daemon, for chroot situations...
277 * logind: add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
279 * cgtop: make cgtop useful in a container
282 - add 'set -e' to scripts in test/
283 - make stuff in test/ work with separate output dir
284 - qemu wrapper script: http://www.spinics.net/lists/kvm/msg72389.html
286 * systemctl delete x.snapshot leaves no trace in logs (at least at default level).
288 * make the coredump collector tool move itself into the user's cgroup
289 so that the coredump is properly written to the user's own journal
292 * seems that when we follow symlinks to units we prefer the symlink
293 destination path over /etc and /usr. We shouldn't do that. Instead
294 /etc should always override /run+/usr and also any symlink
297 * when isolating, try to figure out a way how we implicitly can order
298 all units we stop before the isolating unit...
300 * add ConditionArchitecture= or so
302 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
304 * Support SO_REUSEPORT with socket activation:
305 - Let systemd maintain a pool of servers.
306 - Use for seamless upgrades, by running the new server before stopping the
309 * after all byte-wise realloc() is slow, even on glibc, so i guess we
310 need manual exponential loops after all
312 * BootLoaderSpec: drop allowing ext234 for $BOOT. Clarify that the
313 kernel has to be in $BOOT. Clarify that the boot loader should be
314 installed to the ESP. Define a way how an installer can figure out
315 whether a BLS compliant boot loader is installed.
317 * man: remove .include documentation, and instead push people to use .d/*.conf
319 * think about requeuing jobs when daemon-reload is issued? usecase:
320 the initrd issues a reload after fstab from the host is accessible
321 and we might want to requeue the mounts local-fs acquired through
324 * rework specifier logic so that we can distinguish OOM errors from other errors
326 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
328 * journal-or-kmsg is currently broken? See reverted commit 4a01181e460686d8b4a543b1dfa7f77c9e3c5ab8.
330 * remove any syslog support from log.c -- we probably can't do this before split-off udev is gone for good
332 * documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
334 * add a tool that lists active timer units plus their next elapse and the time the units ran last
336 * man: document the very specific env the shutdown drop-in tools live in
338 * shutdown logging: store to EFI var, and store to USB stick?
340 * man: extend runlevel(8) to mention that runlevels suck, and are dead. Maybe add runlevel(7) with a note about that too
342 * systemctl: maybe add "systemctl add-wants" or so...
344 * man: add more examples to man pages
346 * man: maybe sort directives in man pages, and take sections from --help and apply them to man too
348 * write UI tool that pops up emergency messages from the journal as notification
350 * think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service
352 * msgcatalog determine default language of entries from the catalog
353 file name. i.e. foobar.de_DE.catalog would set the default entry
354 name for the entries to de_DE if they aren't explicitly suffixed
357 * use "log level" rather than "log priority" everywhere
359 * timedate: have global on/off switches for auto-time (NTP), and auto-timezone that connman can subscribe to.
361 * dev-setup.c: when running in a container, create a tiny stub udev
362 database with the systemd tag set for all network interfaces found,
363 so that libudev reports them as present, and systemd's .device units
366 * merge unit_kill_common() and unit_kill_context()
368 * introduce ExecCondition= in services
371 - write man page for efi boot generator
372 - honor language efi variables for default language selection (if there are any?)
373 - honor timezone efi variables for default timezone selection (if there are any?)
374 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
376 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
378 * re-enable "make check" for gtk-doc (broken for unknown reason)
380 * fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points
381 https://bugzilla.redhat.com/show_bug.cgi?id=812826
383 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
385 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
388 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
389 - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
390 - logind: wakelock/opportunistic suspend support
391 - Add pretty name for seats in logind
392 - logind: allow showing logout dialog from system?
393 - logind: non-local X11 server handling
394 - logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API
395 - pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree
396 - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case
398 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
400 * service: watchdog logic: for testing purposes allow ping, but do not require pong
403 - do not use magic msync() in src/journal/journal-file.c, just call fsync()
404 <alxchk> poettering: looks like msync just calls vfs_fsync
405 http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/mm/msync.c#n18
406 - import and delete pstore filesystem content at startup
407 - journald: also get thread ID from client, plus thread name
408 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
409 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
410 - fallback to /dev/log based logging in libsystemd-journal, if we can't log natively?
411 - declare the local journal protocol stable in the wiki interface chart
412 - journal: reuse XZ context
413 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
414 - journald: when dropping msgs due to ratelimit make sure to write
415 "dropped %u messages" not only when we are about to print the next
416 message that works, but alraedy after a short tiemout
417 - check if we can make journalctl by default use --follow mode inside of less if called without args?
418 - maybe add API to send pairs of iovecs via sd_journal_send
419 - journal: when writing journal auto-rotate if time jumps backwards
420 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
421 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
422 and !FOOBAR for events without FOOBAR.
423 - journal: when rotating, copy over old acls/access mode
424 - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs.
425 - 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.
426 - journal: find a way to allow dropping history early, based on priority, other rules
427 - journal: When used on NFS, check payload hashes
428 - journald: check whether it is OK if the client can still modify delivered journal entries
429 - journal live copy, based on libneon (client) and libmicrohttpd (server)
430 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
431 - refuse taking lower-case variable names in sd_journal_send() and friends.
432 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
433 - journal: deal nicely with byte-by-byte copied files, especially regards header
434 - journalctl: expand tabs
435 - journal: store euid in journal if it differs from uid
436 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
437 - Replace utmp, wtmp, btmp, and lastlog completely with journal
438 - Port upower to use the journal for historical power information used in future calculations
441 - document unit_name_mangle()
442 - document that deps in [Unit] sections ignore Alias= fileds in
443 [Install] units of other units, unless those units are disabled
444 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
445 - Document word splitting syntax for ExecStart= and friends
446 - document that units from /etc override those from /usr and /run
447 - document the exit codes when services fail before they are exec()ed
448 - document that service reload may be implemented as service reexec
449 - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications
452 - systemctl list-jobs - show dependencies
453 - add systemctl switch to dump transaction without executing it
454 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
455 - "systemctl disable" on a static unit prints no message and does
456 nothing. "systemctl enable" does nothing, and gives a bad message
457 about it. Should fix both to print nice actionable messages.
458 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
459 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
460 - systemctl enable: improve the success messages (i.e. more human readable, less shell-like)
461 - systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards?
462 - systemctl: "Journal has been rotated since unit was started." message is misleading
463 - support "systemctl stop foobar@.service" to stop all units matching a certain template
464 - Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files"
466 * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations
468 * deal with sendmail/postfix exclusivity
471 - configurable jitter for timer events
472 - Adjust timers to be triggered at the same time as sd-event timers
473 - timer events with system resume
474 - timer units should get the ability to trigger when:
475 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
477 - Support 2012-02~4 as syntax for specifying the fourth to last day of the month.
478 - Modulate timer frequency based on battery state
479 - anacron-like feature
481 * update the kernel's TZ (sys_tz) when DST changes
483 * sync down the system time to the RTC when:
484 - CLOCK_REALTIME makes jumps (the user explicitely requested a time set)
485 - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp)
486 This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime
487 mode, it will never touch the RTC if the no reliable time source is active or the
488 user did not request anything like it.
490 * if booted in "quiet" mode, and an error happens, turn on status
491 output again, so that the emergency mode isn't totally
492 surprising. Also, terminate plymouth.
494 * libunwind support for coredump pattern hook, and includes this in
495 the message for coredumps. After all, libunwind is now capable to
496 unwind coredumps since a few weeks ago. This probably requires that
497 we have nice support for multi-line messages on display in
498 logs-show.c. Alternatively: use libelfutil, which seems to be the
499 better supported alternative.
501 * figure out relation of --all and --full in the various tools
503 * add libsystemd-password or so to query passwords during boot using the password agent logic
505 * 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().
507 * fedup: add --unit to systemctl switch-root somehow
508 * fedup: don't delete initrd on switch-root
512 - timedated: refuse time changes when NTP is on
514 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
516 * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields
518 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
520 * add "provisioning" instructions to setup an empty /etc + /var
521 - used to setup a new container from a shared /usr
522 - superset of tmpfiles model
523 - instructions shipped by packages and stored in /usr/lib/
524 - compose /etc/passwd and /etc/group, copy files
525 - able to create uid + gid used by packages, for file ownership
527 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
528 waiting for the timeout
530 * high level net_prio setting in execution context
532 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
534 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't
536 * refuse boot if /etc/machine-id is not useful (or set taint?)
539 - nspawn: consider changing users for -u with su, so that NSS resolving works correctly
540 - nspawn: implement personality changes a la linux32(8)
541 - nspawn: --read-only is not applied recursively to submounts
542 - nspawn: make use of device cgroup controller by default
543 - bind mount read-only the cgroup tree higher than nspawn
544 - nspawn: investigate whether we can support the same as LXC's
545 lxc.network.type=phys mode, and pass through entire network
546 interfaces to the container
547 - nspawn: maybe add a way to drop additional caps, in addition to add additional caps
548 - nspawn: maybe explicitly reset loginuid?
549 - nspawn: make it work for dwalsh and shared /usr containers -- tmpfs mounts as command line parameters, selinux exec context
550 - refuses to boot containers without /etc/machine-id (OK?), and with empty /etc/machine-id (not OK).
553 - cryptsetup-generator: allow specification of passwords in crypttab itself
554 - move cryptsetup key caching into kernel keyctl?
555 https://bugs.freedesktop.org/show_bug.cgi?id=54982
556 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
558 * move debug shell to tty6 and make sure this doesn't break the gettys on tty6
560 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
561 https://bugs.freedesktop.org/show_bug.cgi?id=54712
563 * after deserializing sockets in socket.c we should reapply sockopts and things
565 * make timer units go away after they elapsed
567 * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html
568 (network and remote-fs on shutdown)
570 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
572 * move PID 1 segfaults to /var/lib/systemd/coredump?
574 * create /sbin/init symlinks from the build system
576 * Query Paul Moore about relabelling socket fds while they are open
578 * allow writing multiple conditions in unit files on one line
580 * explore multiple service instances per listening socket idea
582 * shutdown: don't read-only mount anything when running in container
584 * MountFlags=shared acts as MountFlags=slave right now.
586 * ReadOnlyDirectories= is not applied recursively to submounts
588 * drop PID 1 reloading, only do reexecing (difficult: Reload()
589 currently is properly synchronous, Reexec() is weird, because we
590 can't delay the response properly until we are back, so instead of
591 being properly synchronous we just keep open the fd and close it
592 when done. That means clients don't get a successful method reply,
593 but much rather a disconnect on success.
595 * use opterr = 0 for all getopt tools
597 * properly handle loop back mounts via fstab, especially regards to fsck/passno
599 * allow services with no ExecStart= but with an ExecStop=
601 * dracut-shutdown needs to be ordered before unmounting /boot
603 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
605 * rename "userspace" to "core-os"
608 - syscall filter: add knowledge about compat syscalls
609 - syscall filter: don't enforce no new privs?
610 - syscall filter: option to return EPERM rather than SIGSYS?
611 - syscall filter: port to libseccomp
612 - system-wide seccomp filter
614 * .device aliases need to be implemented with the "following" logic, probably.
616 * load-fragment: when loading a unit file via a chain of symlinks
617 verify that it isn't masked via any of the names traversed.
619 * introduce Type=pid-file
621 * change Requires=basic.target to RequisiteOverride=basic.target
623 * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
625 * automount: implement expire:
626 - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD
627 - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD
628 - every timeout/4 (original autofs logic)
629 - blocking, needs a thread
631 - receive expire packet on pipe if kernel tells the timeout is over
633 - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD
634 - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns
636 * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel)
638 * ExecOnFailure=/usr/bin/foo
641 - remove src/udev/udev-builtin-firmware.c (CONFIG_FW_LOADER_USER_HELPER=n)
644 - add trigger --subsystem-match=usb/usb_device device
646 * cleanup syslog 'priority' vs. 'level' wording
648 * 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
650 * support container_ttys=
652 * introduce mix of BindTo and Requisite
654 * add DeleteSocketsOnStop=yes|no option to socket units
656 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
658 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
660 * default unix qlen is too small (10). bump sysctl? add sockopt?
662 * save coredump in Windows/Mozilla minidump format
664 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
666 * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max
668 * be able to specify a forced restart of service A where service B depends on, in case B
669 needs to be auto-respawned?
671 * when a bus name of a service disappears from the bus make sure to queue further activation requests
673 * tmpfiles: apply "x" on "D" too (see patch from William Douglas)
675 * for services: don't set $HOME in services unless requested
677 * hide PAM/TCPWrap options in fragment parser when compile time disabled
679 * when we automatically restart a service, ensure we restart its rdeps, too.
681 * allow Type=simple with PIDFile=
682 https://bugzilla.redhat.com/show_bug.cgi?id=723942
684 * move PAM code into its own binary
686 * implement Register= switch in .socket units to enable registration
687 in Avahi, RPC and other socket registration services.
689 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
691 * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination
693 * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1)
695 * verify that the AF_UNIX sockets of a service in the fs still exist
696 when we start a service in order to avoid confusion when a user
697 assumes starting a service is enough to make it accessible
699 * support User= and Group= attributes for AF_UNIX sockets.
701 * Make it possible to set the keymap independently from the font on
702 the kernel cmdline. Right now setting one resets also the other.
704 * and a dbus call to generate target from current state
707 - drop /.readahead on bigger upgrades with yum
708 - move readahead files into /var (look for them with .path units?)
709 - readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO
710 - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
711 - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
713 * add support for /bin/mount -s
715 * GC unreferenced jobs (such as .device jobs)
717 * write blog stories about:
718 - hwdb: what belongs into it, lsusb
719 - enabling dbus services
721 - how to make changes to sysctl and sysfs attributes
723 - how to pass throw-away units to systemd, or dynamically change properties of existing units
724 - how to integrate cgconfig and suchlike with systemd
725 - testing with Harald's awesome test kit
727 - how to develop against journal browsing APIs
728 - the journal HTTP iface
729 - non-cgroup resource management
730 - dynamic resource management with cgroups
731 - refreshed, longer missions statement
732 - calendar time events
733 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
734 - how to create your own target
735 - instantiated apache, dovecot and so on
736 - hooking a script into various stages of shutdown/rearly booot
738 * allow port=0 in .socket units
740 * support systemd.mask= on the kernel command line.
742 * recreate systemd's D-Bus private socket file on SIGUSR2
744 * Support --test based on current system state
746 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
748 * maybe introduce ExecRestartPre=
750 * dot output for --test showing the 'initial transaction'
752 * port over to LISTEN_FDS/LISTEN_PID:
753 - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH
757 - libvirtd (/var/run/libvirt/libvirt-sock-ro)
758 - bluetoothd (/var/run/sdp! @/org/bluez/audio!)
761 * fingerprint.target, wireless.target, gps.target, netdevice.target
763 * io priority during initialization
765 * drop cap bounding set in readahead and other services
768 - figure out a simple way to wait for journal events in a way that
770 - add documentation to systemd.daemon
773 - plot per-process IO utilization
774 - group processes based on service association (cgroups)
775 - document initcall_debug
776 - kernel cmdline "bootchart" option for simplicity?
779 - Make sure ID_PATH is always exported and complete for
780 network devices where possible, so we can safely rely
782 - NamePolicy= replace the current naming rules
783 - MACPolicy= support 'firmware', 'synthetic' and 'random'
784 - Check if Driver= is broken, or just my driver (bcma)
790 - natively watch for dbus-*.service symlinks (PENDING)
791 - allow specification of socket mode/umask when allocating DBusServer
792 - allow disabling of fd passing when connecting a AF_UNIX connection
793 - allow disabling of UID passing for AUTH EXTERNAL
794 - always pass cred data along each message
795 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
797 * fix alsa mixer restore to not print error when no config is stored
799 * make cryptsetup lower --iter-time
801 * patch kernel for xattr support in /dev, /proc/, /sys?
803 * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only
804 in an ~30 minutes window. It does not adjust larger differences. Find a way
805 to tell the kernel, to always do a full time sync when the RTC is in UTC and
806 we are in 11-minutes-mode. When we trust the system time to NTP we also want
809 * kernel: add device_type = "fb", "fbcon" to class "graphics"
811 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
813 * dbus upstream still refers to dbus.target and shouldn't
815 * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
817 * add "# export SYSTEMD_PAGER=" to bash login
819 * /usr/bin/service should actually show the new command line
821 * fedora: suggest auto-restart on failure, but not on success and not on coredump. also, ask people to think about changing the start limit logic. Also point people to RestartPreventExitStatus=, SuccessExitStatus=
823 * fedora: F20: go timer units all the way, leave cron.daily for cron
825 * neither pkexec nor sudo initialize environ[] from the PAM environment?
827 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
831 * look for close() vs. close_nointr() vs. close_nointr_nofail()
833 * check for strerror(r) instead of strerror(-r)
835 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
837 * %m in printf() instead of strerror(errno);
841 * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()!
843 * use secure_getenv() instead of getenv() where appropriate
845 * link up selected blog stories from man pages and unit files Documentation= fields
847 Scheduled for removal or fixing:
849 * xxxOverridable dependencies (probably: fix)