3 * Should systemctl status \* work on all unit types, not just .service?
5 * Dangling symlinks of .automount unit files in .wants/ directories, set up
6 automount points even when the original .automount file did not exist
7 anymore. Only the .mount unit was still around.
9 * ExecStart with unicode characters fails in strv_split_extract:
12 Environment=ONE='one' "TWO='two two' too" THREE=
13 ExecStart=/bin/python3 -c 'import sys;print(sys.argv)' $ONE $TWO $THREE
17 * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
19 * wiki: update journal format documentation for lz4 additions
23 * Rearrange tests so that the various test-xyz.c match a specific src/basic/xyz.c again
27 * StateDirectory= and friends should imply RequiresMountsFor=
29 * set LockPersonality= on all our services
31 * Add AddUser= setting to unit files, similar to DynamicUser=1 which however
32 creates a static, persistent user rather than a dynamic, transient user. We
33 can leverage code from sysusers.d for this.
35 * add some optional flag to ReadWritePaths= and friends, that has the effect
36 that we create the dir in question when the service is started. Example:
38 ReadWritePaths=:/var/lib/foobar
40 * sort generated hwdb files alphabetically when we import them, so that git
41 diffs remain minimal (in particular: the OUI databases we import are not
42 sorted, and not stable)
44 * set SystemCallArchitectures=native on all our services
46 * maybe add call sd_journal_set_block_timeout() or so to set SO_SNDTIMEO for
47 the sd-journal logging socket, and, if the timeout is set to 0, sets
48 O_NONBLOCK on it. That way people can control if and when to block for
51 * tighten sd_notify() MAINPID= checks a bit: don't accept foreign PIDs (i.e.
52 PIDs not managed by the service manager)
54 * journald: when we recv a log datagram via the native or syslog transports,
55 search for the PID in the active stream connections, and let's make sure to
56 always process the datagrams before the streams. Then, cache client metadata
57 per stream in the stream object. This way we can somewhat fix the race with
58 quickly exiting processes which log as long as they had their own stream
61 * hostnamed: populate form factor data from a new hwdb database, so that old
62 yogas can be recognized as "convertible" too, even if they predate the DMI
63 "convertible" form factor
65 * Maybe add a small tool invoked early at boot, that adds in or resizes
66 partitions automatically, to be used when the media used is actually larger
67 than the image written onto it is.
69 * Maybe add PrivatePIDs= as new unit setting, and do minimal PID namespacing
70 after all. Be strict however, only support the equivalent of nspawn's
71 --as-pid2 switch, and sanely proxy sd_notify() messages dropping stuff such
74 * change the dependency Set* objects in Unit structures to become Hashmap*, and
75 then store a bit mask who created a specific dependency: the source unit via
76 fragment configuration, the destination unit via fragment configuration, or
77 the source unit via udev rules (in case of .device units), or any combination
78 thereof. This information can then be used to flush out old udev-created
79 dependencies when the udev properties change, and eventually to implement a
80 "systemctl refresh" operation for reloading the configuration of individual
81 units without reloading the whole set.
83 * Add ExecMonitor= setting. May be used multiple times. Forks off a process in
84 the service cgroup, which is supposed to monitor the service, and when it
85 exits the service is considered failed by its monitor.
87 * track the per-service PAM process properly (i.e. as an additional control
88 process), so that it may be queried on the bus and everything.
90 * add a new "debug" job mode, that is propagated to unit_start() and for
91 services results in two things: we raise SIGSTOP right before invoking
92 execve() and turn off watchdog support. Then, use that to implement
93 "systemd-gdb" for attaching to the start-up of any system service in its
96 * replace all canonicalize_file_name() invocations by chase_symlinks(), in
97 particulr those where a rootdir is relevant.
99 * maybe introduce gpt auto discovery for /var/tmp?
101 * set ProtectSystem=strict for all our usual services.
103 * fix PrivateNetwork= so that we fall back gracefully on kernels lacking
104 namespacing support (similar for the other namespacing options)
106 * maybe add gpt-partition-based user management: each user gets his own
107 LUKS-encrypted GPT partition with a new GPT type. A small nss module
108 enumerates users via udev partition enumeration. UIDs are assigned in a fixed
109 way: the partition index is added as offset to some fixed base uid. User name
110 is stored in GPT partition name. A PAM module authenticates the user via the
111 LUKS partition password. Benefits: strong per-user security, compatibility
112 with stateless/read-only/verity-enabled root. (other idea: do this based on
113 loopback files in /home, without GPT involvement)
115 * gpt-auto logic: introduce support for discovering /var matching an image. For
116 that, use a partition type UUID that is hashed from the OS name (as encoded
117 in /etc/os-release), the architecture, and 4 new bits from the gpt flags
118 field of the root partition. This way can easily support multiple OS
119 installations on the same GPT partition table, without problems with
120 unmatched /var partitions.
122 * gpt-auto logic: related to the above, maybe support a "secondary" root
123 partition, that is mounted to / and is writable, and where the actual root's
124 /usr is mounted into.
126 * machined: add apis to query /etc/machine-info data of a container
128 * .mount and .swap units: add Format=yes|no option that formats the partition before mounting/enabling it, implicitly
130 * gpt-auto logic: support encrypted swap, add kernel cmdline option to force it, and honour a gpt bit about it, plus maybe a configuration file
132 * drop nss-myhostname in favour of nss-resolve?
134 * drop internal dlopen() based nss-dns fallback in nss-resolve, and rely on the
135 external nsswitch.conf based one
137 * add a percentage syntax for TimeoutStopSec=, e.g. TimeoutStopSec=150%, and
138 then use that for the setting used in user@.service. It should be understood
139 relative to the configured default value.
141 * on cgroupsv2 add DelegateControllers=, to pick the precise cgroup controllers to delegate
143 * in networkd, when matching device types, fix up DEVTYPE rubbish the kernel passes to us
145 * enable LockMLOCK to take a percentage value relative to physical memory
147 * switch to ProtectSystem=strict for all our long-running services where that's possible
149 * Permit masking specific netlink APIs with RestrictAddressFamily=
151 * nspawn: start UID allocation loop from hash of container name
153 * nspawn: support that /proc, /sys/, /dev are pre-mounted
155 * define gpt header bits to select volatility mode
157 * ProtectKernelLogs= (drops CAP_SYSLOG, add seccomp for syslog() syscall, and DeviceAllow to /dev/kmsg) in service files
159 * ProtectClock= (drops CAP_SYS_TIMES, adds seecomp filters for settimeofday, adjtimex), sets DeviceAllow o /dev/rtc
161 * ProtectTracing= (drops CAP_SYS_PTRACE, blocks ptrace syscall, makes /sys/kernel/tracing go away)
163 * ProtectMount= (drop mount/umount/pivot_root from seccomp, disallow fuse via DeviceAllow, imply Mountflags=slave)
165 * ProtectKeyRing= to take keyring calls away
167 * RemoveKeyRing= to remove all keyring entries of the specified user
169 * ProtectReboot= that masks reboot() and kexec_load() syscalls, prohibits kill
170 on PID 1 with the relevant signals, and makes relevant files in /sys and
171 /proc (such as the sysrq stuff) unavailable
173 * DeviceAllow= should also generate seccomp filters for mknod()
175 * Add DataDirectory=, CacheDirectory= and LogDirectory= to match
176 RuntimeDirectory=, and create it as necessary when starting a service, owned by the right user.
178 * make sure the ratelimit object can deal with USEC_INFINITY as way to turn off things
180 * journalctl: make sure -f ends when the container indicated by -M terminates
182 * mount: automatically search for "main" partition of an image has multiple
185 * expose the "privileged" flag of ExecCommand on the bus, and open it up to
188 * in nss-systemd, if we run inside of RootDirectory= with PrivateUsers= set,
189 find a way to map the User=/Group= of the service to the right name. This way
190 a user/group for a service only has to exist on the host for the right
193 * allow attaching additional journald log fields to cgroups
195 * add bus API for creating unit files in /etc, reusing the code for transient units
197 * add bus API to remove unit files from /etc
199 * add bus API to retrieve current unit file contents (i.e. implement "systemctl cat" on the bus only)
201 * rework fopen_temporary() to make use of open_tmpfile_linkable() (problem: the
202 kernel doesn't support linkat() that replaces existing files, currently)
204 * check if DeviceAllow= should split first, resolve specifiers later
206 * transient units: don't bother with actually setting unit properties, we
207 reload the unit file anyway
209 * journald: sigbus API via a signal-handler safe function that people may call
210 from the SIGBUS handler
212 * optionally, also require WATCHDOG=1 notifications during service start-up and shutdown
214 * resolved: when routing queries, make sure only look for the *longest* suffix...
216 * delay activation of logind until somebody logs in, or when /dev/tty0 pulls it
217 in or lingering is on (so that containers don't bother with it until PAM is used). also exit-on-idle
219 * cache sd_event_now() result from before the first iteration...
221 * add systemctl stop --job-mode=triggering that follows TRIGGERED_BY deps and adds them to the same transaction
223 * PID1: find a way how we can reload unit file configuration for
224 specific units only, without reloading the whole of systemd
226 * add an explicit parser for LimitRTPRIO= that verifies
227 the specified range and generates sane error messages for incorrect
230 * do something about "/control" subcgroups in the unified cgroup hierarchy
232 * when we detect that there are waiting jobs but no running jobs, do something
234 * push CPUAffinity= also into the "cpuset" cgroup controller (only after the cpuset controller got ported to the unified hierarchy)
236 * PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn)
238 * there's probably something wrong with having user mounts below /sys,
239 as we have for debugfs. for exmaple, src/core/mount.c handles mounts
240 prefixed with /sys generally special.
241 http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html
243 * man: document that unless you use StandardError=null the shell >/dev/stderr won't work in shell scripts in services
245 * fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline
247 * docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date
249 * mounting and unmounting mount points manually with different source
250 devices will result in collected on all devices used.
251 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030225.html
253 * add a job mode that will fail if a transaction would mean stopping
254 running units. Use this in timedated to manage the NTP service
256 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html
258 * Maybe add support for the equivalent of "ethtool advertise" to .link files?
259 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030112.html
261 * The udev blkid built-in should expose a property that reflects
262 whether media was sensed in USB CF/SD card readers. This should then
263 be used to control SYSTEMD_READY=1/0 so that USB card readers aren't
264 picked up by systemd unless they contain a medium. This would mirror
265 the behaviour we already have for CD drives.
267 * networkd/udev: implement SR_IOV configuration in .link files:
268 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html
270 * Rework systemctl's GetAll property parsing to use the generic bus_map_all_properties() API
272 * implement a per-service firewall based on net_cls
274 * Port various tools to make use of verbs.[ch], where applicable: busctl,
275 coredumpctl, hostnamectl, localectl, systemd-analyze, timedatectl
277 * hostnamectl: show root image uuid
279 * sysfs set api in libudev is not const
281 * Find a solution for SMACK capabilities stuff:
282 http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html
284 * "systemctl preset-all" should probably order the unit files it
285 operates on lexicographically before starting to work, in order to
286 ensure deterministic behaviour if two unit files conflict (like DMs
289 * synchronize console access with BSD locks:
290 http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html
292 * as soon as we have sender timestamps, revisit coalescing multiple parallel daemon reloads:
293 http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
295 * in systemctl list-unit-files: show the install value the presets would suggest for a service in a third column
297 * figure out when we can use the coarse timers
299 * add "systemctl start -v foobar.service" that shows logs of a service
300 while the start command runs. This is non-trivial to do without
301 races though, since we should flush out all journal messages before
302 returning from the "systemctl stop".
304 * firstboot: make it useful to be run immediately after yum --installroot to set up a machine. (most specifically, make --copy-root-password work even if /etc/passwd already exists
306 * maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment=
308 * introduce systemd-timesync-wait.service or so to sync on an NTP fix?
310 * systemd --user should issue sd_notify() upon reaching basic.target, not on becoming idle
312 * consider showing the unit names during boot up in the status output, not just the unit descriptions
314 * maybe allow timer units with an empty Units= setting, so that they
315 can be used for resuming the system but nothing else.
317 * what to do about udev db binary stability for apps? (raw access is not an option)
319 * maybe provide an API to allow migration of foreign PIDs into existing scopes.
321 * man: maybe use the word "inspect" rather than "introspect"?
323 * systemctl: if some operation fails, show log output?
325 * systemctl edit: use equvalent of cat() to insert existing config as a comment, prepended with #.
326 Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc.
328 * exponential backoff in timesyncd when we cannot reach a server
330 * timesyncd: add ugly bus calls to set NTP servers per-interface, for usage by NM
332 * merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share....
334 * systemd.show_status= should probably have a mode where only failed
337 * add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL
338 (throughout the codebase, not only PID1)
342 - service registration
343 - service/domain/types browsing
345 - DNS-SD service registration from socket units
346 - resolved should optionally register additional per-interface LLMNR
347 names, so that for the container case we can establish the same name
348 (maybe "host") for referencing the server, everywhere.
349 - allow clients to request DNSSEC for a single lookup even if DNSSEC is off (?)
350 - hook up resolved with machined-based address resolution
352 * refcounting in sd-resolve is borked
354 * Add a new verb "systemctl top"
356 * add new gpt type for btrfs volumes
358 * support empty /etc boots nicely:
359 - nspawn/gpt-generator: introduce new gpt partition type for /usr
361 * generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
363 * a way for container managers to turn off getty starting via $container_headless= or so...
365 * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
367 * mount_cgroup_controllers(): symlinks need to get the label applied
369 * For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
370 they run added to the initial transaction and thus confuse Type=idle.
372 * add bus api to query unit file's X fields.
374 * gpt-auto-generator:
375 - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap?
376 - Make /home automount rather than mount?
378 * add generator that pulls in systemd-network from containers when
379 CAP_NET_ADMIN is set, more than the loopback device is defined, even
380 when it is otherwise off
382 * MessageQueueMessageSize= (and suchlike) should use parse_iec_size().
384 * implement Distribute= in socket units to allow running multiple
385 service instances processing the listening socket, and open this up
388 * socket units: support creating sockets in different namespace,
389 opening it up for JoinsNamespaceOf=. This would require to fork off
390 a tiny process that joins the namespace and creates/binds the socket
391 and passes this back to PID1 via SCM_RIGHTS. This also could be used
392 to allow Chown/chgrp on sockets without requiring NSS in PID 1.
394 * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze
395 $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls
396 should SIGSTOP all unit processes in a loop until all processes of
397 it are fully stopped. This can later be used for app management by
398 desktop UIs such as gnome-shell to freeze apps that are not visible
399 on screen, not unlike how job control works on the shell
402 - implement per-slice CPUFairScheduling=1 switch
403 - handle jointly mounted controllers correctly
404 - introduce high-level settings for RT budget, swappiness
405 - how to reset dynamically changed unit cgroup attributes sanely?
406 - when reloading configuration, apply new cgroup configuration
407 - when recursively showing the cgroup hierarchy, optionally also show
408 the hierarchies of child processes
411 - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
412 - ensure scope units may be started only a single time
414 * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk?
416 * when we detect low battery and no AC on boot, show pretty splash and refuse boot
418 * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
420 * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
422 * 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
424 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
426 * If we try to find a unit via a dangling symlink, generate a clean
427 error. Currently, we just ignore it and read the unit from the search
430 * refuse boot if /usr/lib/os-release is missing or /etc/machine-id cannot be set up
432 * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
434 * load .d/*.conf dropins for device units
436 * allow implementation of InaccessibleDirectories=/ plus
437 ReadOnlyDirectories=... for whitelisting files for a service.
441 - GetAllProperties() on a non-existing object does not result in a failure currently
442 - port to sd-resolve for connecting to TCP dbus servers
443 - see if we can introduce a new sd_bus_get_owner_machine_id() call to retrieve the machine ID of the machine of the bus itself
444 - see if we can drop more message validation on the sending side
445 - add API to clone sd_bus_message objects
446 - make AddMatch calls on dbus1 transports async?
447 - longer term: priority inheritance
449 - NameLost/NameAcquired obsolete
452 - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
453 - test bloom filter generation indexes
456 - allow multiple signal handlers per signal?
457 - document chaining of signal handler for SIGCHLD and child handlers
458 - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ...
459 - generate a failure of a default event loop is executed out-of-thread
460 - maybe add support for inotify events
462 * investigate endianness issues of UUID vs. GUID
464 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
465 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
467 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
469 * add a pam module that on password changes updates any LUKS slot where the password matches
471 * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases...
473 * cgtop: make cgtop useful in a container
476 - add 'set -e' to scripts in test/
477 - make stuff in test/ work with separate output dir
479 * seems that when we follow symlinks to units we prefer the symlink
480 destination path over /etc and /usr. We should not do that. Instead
481 /etc should always override /run+/usr and also any symlink
484 * when isolating, try to figure out a way how we implicitly can order
485 all units we stop before the isolating unit...
487 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
489 * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify
490 that the boot loader should be installed to the ESP. Define a way
491 how an installer can figure out whether a BLS compliant boot loader
494 * think about requeuing jobs when daemon-reload is issued? usecase:
495 the initrd issues a reload after fstab from the host is accessible
496 and we might want to requeue the mounts local-fs acquired through
499 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
501 * remove any syslog support from log.c — we probably cannot do this before split-off udev is gone for good
503 * shutdown logging: store to EFI var, and store to USB stick?
505 * think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service
507 * merge unit_kill_common() and unit_kill_context()
509 * introduce ExecCondition= in services
512 - honor language efi variables for default language selection (if there are any?)
513 - honor timezone efi variables for default timezone selection (if there are any?)
514 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
516 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
518 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
520 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
523 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
524 - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
525 - logind: wakelock/opportunistic suspend support
526 - Add pretty name for seats in logind
527 - logind: allow showing logout dialog from system?
528 - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case
529 - session scopes/user unit: add RequiresMountsFor for the home directory of the user
530 - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
531 - if pam_systemd is invoked by su from a process that is outside of a
532 any session we should probably just become a NOP, since that's
533 usually not a real user session but just some system code that just
535 - logind: make the Suspend()/Hibernate() bus calls wait for the for
536 the job to be completed. before returning, so that clients can wait
537 for "systemctl suspend" to finish to know when the suspending is
539 - logind: when the power button is pressed short, just popup a
540 logout dialog. If it is pressed for 1s, do the usual
541 shutdown. Inspiration are Macs here.
542 - expose "Locked" property on logind sesison objects
543 - given that logind now lets PID 1 do all nasty work, we can
544 probably reduce the capability set it retains substantially.
545 (we need CAP_SYS_ADMIN for drmSetMaster(), so maybe not worth it)
546 - expose orientation sensors and tablet mode through logind
547 - maybe allow configuration of the StopTimeout for session scopes
548 - rename session scope so that it includes the UID. THat way
549 the session scope can be arranged freely in slices and we don't have
550 make assumptions about their slice anymore.
551 - follow PropertiesChanged state more closely, to deal with quick logouts and
554 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
556 * service: watchdog logic: for testing purposes allow ping, but do not require pong
559 - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields
560 - import and delete pstore filesystem content at startup
561 - journald: also get thread ID from client, plus thread name
562 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
563 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
564 - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively?
565 - declare the local journal protocol stable in the wiki interface chart
566 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
567 - journald: when dropping msgs due to ratelimit make sure to write
568 "dropped %u messages" not only when we are about to print the next
569 message that works, but alraedy after a short tiemout
570 - check if we can make journalctl by default use --follow mode inside of less if called without args?
571 - maybe add API to send pairs of iovecs via sd_journal_send
572 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
573 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
574 and !FOOBAR for events without FOOBAR.
575 - journal: store timestamp of journal_file_set_offline() int he header,
576 so it is possible to display when the file was last synced.
577 - 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.
578 - journal: find a way to allow dropping history early, based on priority, other rules
579 - journal: When used on NFS, check payload hashes
580 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
581 - refuse taking lower-case variable names in sd_journal_send() and friends.
582 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
583 - journal: deal nicely with byte-by-byte copied files, especially regards header
584 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
585 - Replace utmp, wtmp, btmp, and lastlog completely with journal
586 - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
587 - when a kernel driver logs in a tight loop, we should ratelimit that too.
588 - journald: optionally, log debug messages to /run but everything else to /var
589 - journald: when we drop syslog messages because the syslog socket is
590 full, make sure to write how many messages are lost as first thing
591 to syslog when it works again.
592 - journald: make sure ratelimit is actually really per-service with the new cgroup changes
593 - change systemd-journal-flush into a service that stays around during
594 boot, and causes the journal to be moved back to /run on shutdown,
595 so that we do not keep /var busy. This needs to happen synchronously,
596 hence doing this via signals is not going to work.
597 - optionally support running journald from the command line for testing purposes in external projects
598 - journald: allow per-priority and per-service retention times when rotating/vacuuming
599 - journald: make use of uid-range.h to managed uid ranges to split
601 - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
602 - improve journalctl performance by loading journal files
603 lazily. Encode just enough information in the file name, so that we
604 do not have to open it to know that it is not interesting for us, for
605 the most common operations.
606 - journal-or-kmsg is currently broken? See reverted
607 commit 4a01181e460686d8b4a543b1dfa7f77c9e3c5ab8.
608 - man: document that corrupted journal files is nothing to act on
609 - rework journald sigbus stuff to use mutex
610 - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our
611 services that run under their own user ids, and use User= (but only
612 in a world where userns is ubiquitous since otherwise we cannot
613 invoke those daemons on the host AND in a container anymore). Also,
614 if LimitNPROC= is used without User= we should warn and refuse
616 - journalctl --verify: don't show files that are currently being
617 written to as FAIL, but instead show that their are being written to.
618 - add journalctl -H that talks via ssh to a remote peer and passes through
620 - add a version of --merge which also merges /var/log/journal/remote
621 - log accumulated resource usage after each service invocation
622 - journalctl: -m should access container journals directly by enumerating
623 them via machined, and also watch containers coming and going.
624 Benefit: nspawn --ephemeral would start working nicely with the journal.
625 - assign MESSAGE_ID to log messages about failed services
628 - document that deps in [Unit] sections ignore Alias= fields in
629 [Install] units of other units, unless those units are disabled
630 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
631 - document the exit codes when services fail before they are exec()ed
632 - document that service reload may be implemented as service reexec
633 - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications
634 - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
635 - document systemd-journal-flush.service properly
636 - documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
637 - man: document the very specific env the shutdown drop-in tools live in
638 - man: add more examples to man pages
639 - man: maybe sort directives in man pages, and take sections from --help and apply them to man too
642 - add systemctl switch to dump transaction without executing it
643 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
644 - "systemctl disable" on a static unit prints no message and does
645 nothing. "systemctl enable" does nothing, and gives a bad message
646 about it. Should fix both to print nice actionable messages.
647 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
648 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
649 - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
650 - systemctl: "Journal has been rotated since unit was started." message is misleading
651 - better error message if you run systemctl without systemd running
652 - systemctl status output should include list of triggering units and their status
655 - "systemctl mask" should find all names by which a unit is accessible
656 (i.e. by scanning for symlinks to it) and link them all to /dev/null
659 - timer units should get the ability to trigger when:
660 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
662 - Modulate timer frequency based on battery state
664 * add libsystemd-password or so to query passwords during boot using the password agent logic
666 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
668 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
670 * make repeated alt-ctrl-del presses printing a dump
672 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
674 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
677 - emulate /dev/kmsg using CUSE and turn off the syslog syscall
678 with seccomp. That should provide us with a useful log buffer that
679 systemd can log to during early boot, and disconnect container logs
680 from the kernel's logs.
681 - as soon as networkd has a bus interface, hook up --network-interface=,
682 --network-bridge= with networkd, to trigger netdev creation should an
684 - a nice way to boot up without machine id set, so that it is set at boot
685 automatically for supporting --ephemeral. Maybe hash the host machine id
686 together with the machine name to generate the machine id for the container
687 - fix logic always print a final newline on output.
688 https://github.com/systemd/systemd/pull/272#issuecomment-113153176
689 - should optionally support receiving WATCHDOG=1 messages from its payload
691 - should send out sd_notify("WATCHDOG=1") messages
692 - optionally automatically add FORWARD rules to iptables whenever nspawn is
693 running, remove them when shut down.
694 - Improve error message when --bind= is used on a non-existing source
696 - maybe make copying of /etc/resolv.conf optional, and skip it if --read-only
700 - refuse mounting over a mount point
701 - automatically discover .roothash files in dissect, similarly to nspawn
704 - add an API so that libvirt-lxc can inform us about network interfaces being
705 removed or added to an existing machine
706 - "machinectl migrate" or similar to copy a container from or to a
707 difference host, via ssh
708 - introduce systemd-nspawn-ephemeral@.service, and hook it into
709 "machinectl start" with a new --ephemeral switch
710 - "machinectl status" should also show internal logs of the container in
712 - "machinectl list-images" should show os-release data, as well as
713 machine-info data (including deployment level)
714 - "machinectl history"
716 - "machinectl commit" that takes a writable snapshot of a tree, invokes a
717 shell in it, and marks it read-only after use
720 - generate a nice warning if mkfs.btrfs is missing
723 - cryptsetup-generator: allow specification of passwords in crypttab itself
724 - move cryptsetup key caching into kernel keyctl?
725 https://bugs.freedesktop.org/show_bug.cgi?id=54982
726 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
728 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
729 https://bugs.freedesktop.org/show_bug.cgi?id=54712
731 * create /sbin/init symlinks from the build system
733 * MountFlags=shared acts as MountFlags=slave right now.
735 * properly handle loop back mounts via fstab, especially regards to fsck/passno
737 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
742 - add trigger --subsystem-match=usb/usb_device device
743 - reimport udev db after MOVE events for devices without dev_t
745 * 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
747 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
749 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
752 - save coredump in Windows/Mozilla minidump format
753 - when truncating coredumps, also log the full size that the process had, and make a metadata field so we can report truncated coredumps
755 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
757 * default to actual 32-bit PIDs, via /proc/sys/kernel/pid_max
759 * be able to specify a forced restart of service A where service B depends on, in case B
760 needs to be auto-respawned?
763 - apply "x" on "D" too (see patch from William Douglas)
765 - instead of ignoring unknown fields, reject them.
766 - creating new directories/subvolumes/fifos/device nodes
767 should not follow symlinks. None of the other adjustment or creation
768 calls follow symlinks.
770 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
772 * verify that the AF_UNIX sockets of a service in the fs still exist
773 when we start a service in order to avoid confusion when a user
774 assumes starting a service is enough to make it accessible
776 * Make it possible to set the keymap independently from the font on
777 the kernel cmdline. Right now setting one resets also the other.
779 * and a dbus call to generate target from current state
781 * write blog stories about:
782 - hwdb: what belongs into it, lsusb
783 - enabling dbus services
785 - how to make changes to sysctl and sysfs attributes
787 - how to pass throw-away units to systemd, or dynamically change properties of existing units
788 - testing with Harald's awesome test kit
790 - how to develop against journal browsing APIs
791 - the journal HTTP iface
792 - non-cgroup resource management
793 - dynamic resource management with cgroups
794 - refreshed, longer missions statement
795 - calendar time events
796 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
797 - how to create your own target
798 - instantiated apache, dovecot and so on
799 - hooking a script into various stages of shutdown/rearly booot
801 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
803 * dot output for --test showing the 'initial transaction'
805 * fingerprint.target, wireless.target, gps.target, netdevice.target
808 - .timer units should optionally support CLOCK_BOOTTIME in addition to CLOCK_MONOTONIC
809 - When logging about multiple units (stopping BoundTo units, conflicts, etc.),
810 log both units as UNIT=, so that journalctl -u triggers on both.
811 - generate better errors when people try to set transient properties
812 that are not supported...
813 http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html
814 - maybe introduce WantsMountsFor=? Usecase:
815 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html
816 - recreate systemd's D-Bus private socket file on SIGUSR2
817 - move PAM code into its own binary
818 - when we automatically restart a service, ensure we restart its rdeps, too.
819 - hide PAM options in fragment parser when compile time disabled
820 - Support --test based on current system state
821 - 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().
822 - after deserializing sockets in socket.c we should reapply sockopts and things
823 - drop PID 1 reloading, only do reexecing (difficult: Reload()
824 currently is properly synchronous, Reexec() is weird, because we
825 cannot delay the response properly until we are back, so instead of
826 being properly synchronous we just keep open the fd and close it
827 when done. That means clients do not get a successful method reply,
828 but much rather a disconnect on success.
829 - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
830 - when a bus name of a service disappears from the bus make sure to queue further activation requests
833 - allow port=0 in .socket units
834 - maybe introduce ExecRestartPre=
835 - add ReloadSignal= for configuring a reload signal to use
836 - implement Register= switch in .socket units to enable registration
837 in Avahi, RPC and other socket registration services.
838 - allow Type=simple with PIDFile=
839 https://bugzilla.redhat.com/show_bug.cgi?id=723942
840 - allow writing multiple conditions in unit files on one line
841 - load-fragment: when loading a unit file via a chain of symlinks
842 verify that it is not masked via any of the names traversed.
843 - introduce Type=pid-file
844 - ExecOnFailure=/usr/bin/foo
845 - introduce mix of BindTo and Requisite
846 - add a concept of RemainAfterExit= to scope units
847 - Set NoNewPrivileges= on all of our own services, where that makes sense
848 - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
849 - consider adding RuntimeDirectoryUser= + RuntimeDirectoryGroup=
852 - Make sure ID_PATH is always exported and complete for
853 network devices where possible, so we can safely rely
855 - check MTUBytes parsing (expecting size_t but we are using unsigned)
858 - add support for more attribute types
859 - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places
862 - add more keys to [Route] and [Address] sections
863 - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
864 - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
865 - add reduced [Link] support to .network files
866 - add Scope= parsing option for [Network]
867 - properly handle routerless dhcp leases
868 - work with non-Ethernet devices
869 - add support for more bond options
870 - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from?
871 - the DHCP lease data (such as NTP/DNS) is still made available when
872 a carrier is lost on a link. It should be removed instantly.
873 - expose in the API the following bits:
874 - option 15, domain name and/or option 119, search list
875 - option 12, host name and/or option 81, fqdn
876 - option 123, 144, geolocation
877 - option 252, configure http proxy (PAC/wpad)
878 - provide a way to define a per-network interface default metric value
879 for all routes to it. possibly a second default for DHCP routes.
880 - allow Name= to be specified repeatedly in the [Match] section. Maybe also
881 support Name=foo*|bar*|baz ?
882 - duplicate address check for static IPs (like ARPCHECK in network-scripts)
883 - allow DUID/IAID to be customized, see issue #394.
884 - whenever uplink info changes, make DHCP server send out FORCERENEW
886 * networkd-wait-online:
887 - make operstates to wait for configurable?
890 - figure out how much we can increase Maximum Message Size
891 - support RFC4702 (pass FQDN)
894 - add functions to set previously stored IPv6 addresses on startup and get
895 them at shutdown; store them in client->ia_na
896 - write more test cases
897 - implement reconfigure support, see 5.3., 15.11. and 22.20.
898 - implement support for temporary adressess (IA_TA)
899 - implement dhcpv6 authentication
900 - investigate the usefulness of Confirm messages; i.e. are there any
901 situations where the link changes without any loss in carrier detection
903 - some servers don't do rapid commit without a filled in IA_NA, verify
910 - natively watch for dbus-*.service symlinks (PENDING)
911 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
913 * fix alsa mixer restore to not print error when no config is stored
915 * make cryptsetup lower --iter-time
917 * patch kernel for xattr support in /dev, /proc/, /sys?
919 * kernel: add device_type = "fb", "fbcon" to class "graphics"
921 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
923 * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
925 * /usr/bin/service should actually show the new command line
927 * 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=
929 * fedora: F20: go timer units all the way, leave cron.daily for cron
931 * neither pkexec nor sudo initialize environ[] from the PAM environment?
933 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
935 * register catalog database signature as file magic
937 * zsh shell completion:
938 - <command> <verb> -<TAB> should complete options, but currently does not
939 - systemctl add-wants,add-requires
944 * look for close() vs. close_nointr() vs. close_nointr_nofail()
946 * check for strerror(r) instead of strerror(-r)
948 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
952 * set_put(), hashmap_put() return values check. i.e. == 0 does not free()!
954 * use secure_getenv() instead of getenv() where appropriate
956 * link up selected blog stories from man pages and unit files Documentation= fields