3 * copy.c: set the right chattrs before copying files and others after
5 * Many manager configuration settings that are only applicable to user
6 manager or system manager can be always set. It would be better to reject
7 them when parsing config.
9 * Clarify what IPAddress* matches (source, destination, both?)
13 * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
15 * wiki: update journal format documentation for lz4 additions
19 * Rearrange tests so that the various test-xyz.c match a specific src/basic/xyz.c again
21 * rework mount.c and swap.c to follow proper state enumeration/deserialization
22 semantics, like we do for device.c now
26 * make MAINPID= message reception checks even stricter: if service uses User=,
27 then check sending UID and ignore message if it doesn't match the user or
30 * maybe trigger a uevent "change" on a device if "systemctl reload xyz.device"
33 * when importing an fs tree with machined, optionally apply userns-rec-chown
35 * when importing an fs tree with machined, complain if image is not an OS
37 * Maybe introduce a helper safe_exec() or so, which is to execve() which
38 safe_fork() is to fork(). And then make revert the RLIMIT_NOFILE soft limit
39 to 1K implicitly, unless explicitly opted-out.
41 * rework seccomp/nnp logic that that even if User= is used in combination with
42 a seccomp option we don't have to set NNP. For that, change uid first whil
43 keeping CAP_SYS_ADMIN, then apply seccomp, the drop cap.
45 * add a concept for automatically loading per-unit secrets off disk and
46 inserting them into the kernel keyring. Maybe SecretsDirectory= similar to
47 ConfigurationDirectory=.
49 * when no locale is configured, default to UEFI's PlatformLang variable
51 * When logind.conf contains HandleLidSwitch=suspend-then-hibernate and we can't
52 hibernate because the swap partition isn't large enough, still suspend
54 * bootctl: implement Type #2 boot loader entry discovery
56 * bootctl,sd-boot: actually honour the "architecture" key
58 * when a socket unit is spawned with an AF_UNIX path in /var/run, complain and
59 patch it to use /run instead
61 * set memory.oom.group in cgroup v2 for all leaf cgroups (kernel v4.19+)
63 * add a new syscall group "@esoteric" for more esoteric stuff such as bpf() and
64 usefaultd() and make systemd-analyze check for it.
66 * paranoia: whenever we process passwords, call mlock() on the memory
67 first. i.e. look for all places we use string_erase()/string_free_erase() and
68 augment them with mlock()
70 * whenever oom_kill memory.event event is triggered print a nice log message
72 * Move RestrictAddressFamily= to the new cgroup create socket
74 * support the bind/connect/sendmsg cgroup stuff for sandboxing, and possibly
77 * maybe implicitly attach monotonic+realtime timestamps to outgoing messages in
78 log.c and sd-journal-send
80 * chown() tty a service is attached to after the service goes down
82 * optionally: turn on cgroup delegation for per-session scope units
84 * introduce per-unit (i.e. per-slice, per-service) journal log size limits.
86 * optionally, if a per-partition GPT flag is set for the root/home/… partitions
87 format the partition on next boot and unset the flag, in order to implement
88 factory reset. also, add a second flag that simply indicates whether such a
89 scheme is supported. then, add a tool (or maybe beef up systemd-dissect) to
90 show state of these flags, and optionally trigger such a factory reset on
91 next boot by setting the flag.
93 * sd-boot: search drop-ins in $BOOT, too
95 * sd-boot: add "oneshot boot timeout" variable support
97 * sd-boot: automatically load EFI modules from some drop-in dir, so that people
98 can add in file system drivers and such
100 * esp generator: also mount $BOOT if found
102 * sd-boot: optionally, show boot menu when previous default boot item has
103 non-zero "tries done" count
105 * logind: add "boot into bootmenu" API, and possibly even "boot into windows"
106 and "boot into macos".
108 * bootspec.c: also enumerate EFI unified kernel images.
110 * maybe set a special xattr on cgroups that have delegate=yes set, to make it
111 easy to mark cut points
113 * introduce an option (or replacement) for "systemctl show" that outputs all
114 properties as JSON, similar to busctl's new JSON output. In contrast to that
115 it should skip the variant type string though.
117 * augment CODE_FILE=, CODE_LINE= with something like CODE_BASE= or so which
118 contains some identifier for the project, which allows us to include
119 clickable links to source files generating these log messages. The identifier
120 could be some abberviated URL prefix or so (taking inspiration from Go
121 imports). For example, for systemd we could use
122 CODE_BASE=github.com/systemd/systemd/blob/98b0b1123cc or so which is
123 sufficient to build a link by prefixing "http://" and suffixing the
126 * when outputting log data with journalctl and the log data includes references
127 to configuration files (CONFIG_FILE=), create a clickable link for it.
129 * Augment MESSAGE_ID with MESSAGE_BASE, in a similar fashion so that we can
130 make clickable links from log messages carrying a MESSAGE_ID, that lead to
131 some explanatory text online.
133 * maybe extend .path units to expose fanotify() per-mount change events
135 * Add a "systemctl list-units --by-slice" mode or so, which rearranges the
136 output of "systemctl list-units" slightly by showing the tree structure of
137 the slices, and the units attached to them.
139 * the a-posteriori stopping of units bound to units that disappeared logic
140 should be reworked: there should be a queue of units, and we should only
141 enqeue stop jobs from a defer event that processes queue instead of
142 right-away when we find a unit that is bound to one that doesn't exist
143 anymore. (similar to how the stop-unneeded queue has been reworked the same
146 * nspawn: make nspawn suitable for shell pipelines: instead of triggering a
147 hangup when input is finished, send ^D, which synthesizes an EOF. Then wait
148 for hangup or ^D before passing on the EOF.
150 * When reloading configuration PID 1 should reset all its properties to the
151 original defaults before calling parse_config()
153 * Add OnTimezoneChange= and OnTimeChange= stanzas to .timer units in order to
154 schedule events based on time and timezone changes.
156 * nspawn: greater control over selinux label?
158 * hibernate/s2h: make this robust and safe to enable in Fedora by default.
161 1. add resume_offset support to the resume code (i.e. support swap files
163 2. check if swap is on weird storage and refuse if so
164 3. add autodetection of hibernation images
166 * cgroups: use inotify to get notified when somebody else modifies cgroups
167 owned by us, then log a friendly warning.
169 * beef up log.c with support for stripping ANSI sequences from strings, so that
170 it is OK to include them in log strings. This would be particularly useful so
171 that our log messages could contain clickable links for example for unit
172 files and suchlike we operate on.
174 * add support for "portablectl attach http://foobar.com/waaa.raw (i.e. importd integration)
176 * add attach --enable and attach --now (for attach+enable+start)
178 * sync dynamic uids/gids between host+portable srvice (i.e. if DynamicUser=1 is set for a service, make sure that the
179 selected user is resolvable in the service even if it ships its own /etc/passwd)
181 * Fix DECIMAL_STR_MAX or DECIMAL_STR_WIDTH. One includes a trailing NUL, the
182 other doesn't. What a desaster. Probably to exclude it. Also
183 DECIMAL_STR_WIDTH should probably add an extra "-" into account for negative
186 * Check that users of inotify's IN_DELETE_SELF flag are using it properly, as
187 usually IN_ATTRIB is the right way to watch deleted files, as the former only
188 fires when a file is actually removed from disk, i.e. the link count drops to
189 zero and is not open anymore, while the latter happens when a file is
190 unlinked from any dir.
192 * port systemctl, busctl, … over to format-table.[ch]'s table formatters
194 * pid1: lock image configured with RootDirectory=/RootImage= using the usual nspawn semantics while the unit is up
196 * add --vacuum-xyz options to coredumpctl, matching those journalctl already has.
198 * SuccessExitStatus= and friends should probably also accept symbolic exit
199 codes names, i.e. error codes from the list maintained in exit-codes.[ch]
201 * introduce Ephemeral= unit file switch, that creates an ephemeral copy of all
202 files and directories that are left writable for a unit, and which are
203 removed after the unit goes down again. A bit like --ephemeral for
204 systemd-nspawn but for system services. If used together with RootImage= this
205 should reflink the image file itself.
207 Related: add Ephemeral=<path1> <path2> … which would allow marking
208 specific paths only like this.
210 * add CopyFile= or so as unit file setting that may be used to copy files or
211 directory trees from the host to te services RootImage= and RootDirectory=
212 environment. Which we can use for /etc/machine-id and in particular
213 /etc/resolv.conf. Should be smart and do something useful on read-only
214 images, for example fallback to read-only bind mounting the file instead.
216 * nspawn's console TTY should be allocated from within the container, not
217 mounted in from the outside
219 * show invocation ID in systemd-run output
221 * bypass SIGTERM state in unit files if KillSignal is SIGKILL
223 * tree-wide: ensure we always block the signals we hook into with
224 sd_event_add_signal() first
226 * add proper dbus APIs for the various sd_notify() commands, such as MAINPID=1
227 and so on, which would mean we could report errors and such.
229 * teach tmpfiles.d q/Q logic something sensible in the context of XFS/ext4
232 * introduce DefaultSlice= or so in system.conf that allows changing where we
233 place our units by default, i.e. change system.slice to something
234 else. Similar, ManagerSlice= should exist so that PID1's own scope unit could
235 be moved somewhere else too. Finally machined and logind should get similar
236 options so that it is possible to move user session scopes and machines to a
237 different slice too by default. Usecase: people who want to put resources on
238 the entire system, with the exception of one specific service. See:
239 https://lists.freedesktop.org/archives/systemd-devel/2018-February/040369.html
241 * maybe rework get_user_creds() to query the user database if $SHELL is used
242 for root, but only then.
244 * be stricter with fds we receive for the fdstore: close them asynchronously
246 * calenderspec: add support for week numbers and day numbers within a
247 year. This would allow us to define "bi-weekly" triggers safely.
249 * sd-bus: add vtable flag, that may be used to request client creds implicitly
250 and asynchronously before dispatching the operation
252 * make use of ethtool veth peer info in machined, for automatically finding out
253 host-side interface pointing to the container.
255 * add some special mode to LogsDirectory=/StateDirectory=… that allows
256 declaring these directories without necessarily pulling in deps for them, or
257 creating them when starting up. That way, we could declare that
258 systemd-journald writes to /var/log/journal, which could be useful when we
259 doing disk usage calculations and so on.
261 * taint systemd if there are fewer than 65536 users assigned (userns) to the system.
263 * deprecate RootDirectoryStartOnly= in favour of a new ExecStart= prefix char
265 * add a new RuntimeDirectoryPreserve= mode that defines a similar lifecycle for
266 the runtime dir as we maintain for the fdstore: i.e. keep it around as long
267 as the unit is running or has a job queued.
269 * support projid-based quota in machinectl for containers
271 * Add NetworkNamespacePath= to specify a path to a network namespace
273 * maybe use SOURCE_DATE_EPOCH (i.e. the env var the reproducible builds folks
274 introduced) as the RTC epoch, instead of the mtime of NEWS.
276 * add a way to lock down cgroup migration: a boolean, which when set for a unit
277 makes sure the processes in it can never migrate out of it
279 * blog about fd store and restartable services
281 * document Environment=SYSTEMD_LOG_LEVEL=debug drop-in in debugging document
283 * rework ExecOutput and ExecInput enums so that EXEC_OUTPUT_NULL loses its
284 magic meaning and is no longer upgraded to something else if set explicitly.
286 * in the long run: permit a system with /etc/machine-id linked to /dev/null, to
287 make it lose its identity, i.e. be anonymous. For this we'd have to patch
288 through the whole tree to make all code deal with the case where no machine
291 * optionally, collect cgroup resource data, and store it in per-unit RRD files,
292 suitable for processing with rrdtool. Add bus API to access this data, and
293 possibly implement a CPULoad property based on it.
295 * beef up pam_systemd to take unit file settings such as cgroups properties as
298 * maybe hook of xfs/ext4 quotactl() with services? i.e. automatically manage
299 the quota of a the user indicated in User= via unit file settings, like the
300 other resource management concepts. Would mix nicely with DynamicUser=1. Or
301 alternatively, do this with projids, so that we can also cover services
302 running as root. Quota should probably cover all the special dirs such as
303 StateDirectory=, LogsDirectory=, CacheDirectory=, as well as RootDirectory= if it
304 is set, plus the whole disk space any image configured with RootImage=.
306 * In DynamicUser= mode: before selecting a UID, use disk quota APIs on relevant
307 disks to see if the UID is already in use.
309 * add "systemctl wait" or so, which does what "systemd-run --wait" does, but
310 for all units. It should be both a way to pin units into memory as well as a
311 wait to retrieve their exit data.
313 * expose IO accounting data on the bus, show it in systemd-run --wait and log
314 about it in the resource log message
316 * add "systemctl purge" for flushing out configuration, state, logs, ... of a
317 unit when it is stopped
319 * show whether a service has out-of-date configuration in "systemctl status" by
320 using mtime data of ConfigurationDirectory=.
322 * replace all remaining uses of fgets() + LINE_MAX by read_line()
324 * Add AddUser= setting to unit files, similar to DynamicUser=1 which however
325 creates a static, persistent user rather than a dynamic, transient user. We
326 can leverage code from sysusers.d for this.
328 * add some optional flag to ReadWritePaths= and friends, that has the effect
329 that we create the dir in question when the service is started. Example:
331 ReadWritePaths=:/var/lib/foobar
333 * maybe add call sd_journal_set_block_timeout() or so to set SO_SNDTIMEO for
334 the sd-journal logging socket, and, if the timeout is set to 0, sets
335 O_NONBLOCK on it. That way people can control if and when to block for
338 * hostnamed: populate form factor data from a new hwdb database, so that old
339 yogas can be recognized as "convertible" too, even if they predate the DMI
340 "convertible" form factor
342 * Maybe add a small tool invoked early at boot, that adds in or resizes
343 partitions automatically, to be used when the media used is actually larger
344 than the image written onto it is.
346 * Maybe add PrivatePIDs= as new unit setting, and do minimal PID namespacing
347 after all. Be strict however, only support the equivalent of nspawn's
348 --as-pid2 switch, and sanely proxy sd_notify() messages dropping stuff such
351 * Add ExecMonitor= setting. May be used multiple times. Forks off a process in
352 the service cgroup, which is supposed to monitor the service, and when it
353 exits the service is considered failed by its monitor.
355 * track the per-service PAM process properly (i.e. as an additional control
356 process), so that it may be queried on the bus and everything.
358 * add a new "debug" job mode, that is propagated to unit_start() and for
359 services results in two things: we raise SIGSTOP right before invoking
360 execve() and turn off watchdog support. Then, use that to implement
361 "systemd-gdb" for attaching to the start-up of any system service in its
364 * maybe introduce gpt auto discovery for /var/tmp?
366 * maybe add gpt-partition-based user management: each user gets his own
367 LUKS-encrypted GPT partition with a new GPT type. A small nss module
368 enumerates users via udev partition enumeration. UIDs are assigned in a fixed
369 way: the partition index is added as offset to some fixed base uid. User name
370 is stored in GPT partition name. A PAM module authenticates the user via the
371 LUKS partition password. Benefits: strong per-user security, compatibility
372 with stateless/read-only/verity-enabled root. (other idea: do this based on
373 loopback files in /home, without GPT involvement)
375 * gpt-auto logic: introduce support for discovering /var matching an image. For
376 that, use a partition type UUID that is hashed from the OS name (as encoded
377 in /etc/os-release), the architecture, and 4 new bits from the gpt flags
378 field of the root partition. This way can easily support multiple OS
379 installations on the same GPT partition table, without problems with
380 unmatched /var partitions.
382 * gpt-auto logic: related to the above, maybe support a "secondary" root
383 partition, that is mounted to / and is writable, and where the actual root's
384 /usr is mounted into.
386 * 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
388 * drop nss-myhostname in favour of nss-resolve?
390 * add a percentage syntax for TimeoutStopSec=, e.g. TimeoutStopSec=150%, and
391 then use that for the setting used in user@.service. It should be understood
392 relative to the configured default value.
394 * in networkd, when matching device types, fix up DEVTYPE rubbish the kernel passes to us
396 * enable LockMLOCK to take a percentage value relative to physical memory
398 * Permit masking specific netlink APIs with RestrictAddressFamily=
400 * nspawn: support that /proc, /sys/, /dev are pre-mounted
402 * define gpt header bits to select volatility mode
404 * ProtectKernelLogs= (drops CAP_SYSLOG, add seccomp for syslog() syscall, and DeviceAllow to /dev/kmsg) in service files
406 * ProtectClock= (drops CAP_SYS_TIMES, adds seecomp filters for settimeofday, adjtimex), sets DeviceAllow o /dev/rtc
408 * ProtectTracing= (drops CAP_SYS_PTRACE, blocks ptrace syscall, makes /sys/kernel/tracing go away)
410 * ProtectMount= (drop mount/umount/pivot_root from seccomp, disallow fuse via DeviceAllow, imply Mountflags=slave)
412 * ProtectKeyRing= to take keyring calls away
414 * RemoveKeyRing= to remove all keyring entries of the specified user
416 * ProtectReboot= that masks reboot() and kexec_load() syscalls, prohibits kill
417 on PID 1 with the relevant signals, and makes relevant files in /sys and
418 /proc (such as the sysrq stuff) unavailable
420 * make sure the ratelimit object can deal with USEC_INFINITY as way to turn off things
422 * journalctl: make sure -f ends when the container indicated by -M terminates
424 * mount: automatically search for "main" partition of an image has multiple
427 * expose the "privileged" flag of ExecCommand on the bus, and open it up to
430 * in nss-systemd, if we run inside of RootDirectory= with PrivateUsers= set,
431 find a way to map the User=/Group= of the service to the right name. This way
432 a user/group for a service only has to exist on the host for the right
435 * add bus API for creating unit files in /etc, reusing the code for transient units
437 * add bus API to remove unit files from /etc
439 * add bus API to retrieve current unit file contents (i.e. implement "systemctl cat" on the bus only)
441 * rework fopen_temporary() to make use of open_tmpfile_linkable() (problem: the
442 kernel doesn't support linkat() that replaces existing files, currently)
444 * transient units: don't bother with actually setting unit properties, we
445 reload the unit file anyway
447 * journald: sigbus API via a signal-handler safe function that people may call
448 from the SIGBUS handler
450 * optionally, also require WATCHDOG=1 notifications during service start-up and shutdown
452 * resolved: when routing queries, make sure only look for the *longest* suffix...
454 * delay activation of logind until somebody logs in, or when /dev/tty0 pulls it
455 in or lingering is on (so that containers don't bother with it until PAM is used). also exit-on-idle
457 * cache sd_event_now() result from before the first iteration...
459 * add systemctl stop --job-mode=triggering that follows TRIGGERED_BY deps and adds them to the same transaction
461 * PID1: find a way how we can reload unit file configuration for
462 specific units only, without reloading the whole of systemd
464 * add an explicit parser for LimitRTPRIO= that verifies
465 the specified range and generates sane error messages for incorrect
468 * when we detect that there are waiting jobs but no running jobs, do something
470 * push CPUAffinity= also into the "cpuset" cgroup controller (only after the cpuset controller got ported to the unified hierarchy)
472 * PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn)
474 * there's probably something wrong with having user mounts below /sys,
475 as we have for debugfs. for exmaple, src/core/mount.c handles mounts
476 prefixed with /sys generally special.
477 http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html
479 * fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline
481 * docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date
483 * add a job mode that will fail if a transaction would mean stopping
484 running units. Use this in timedated to manage the NTP service
486 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html
488 * The udev blkid built-in should expose a property that reflects
489 whether media was sensed in USB CF/SD card readers. This should then
490 be used to control SYSTEMD_READY=1/0 so that USB card readers aren't
491 picked up by systemd unless they contain a medium. This would mirror
492 the behaviour we already have for CD drives.
494 * networkd/udev: implement SR_IOV configuration in .link files:
495 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html
497 * hostnamectl: show root image uuid
499 * Find a solution for SMACK capabilities stuff:
500 http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html
502 * "systemctl preset-all" should probably order the unit files it
503 operates on lexicographically before starting to work, in order to
504 ensure deterministic behaviour if two unit files conflict (like DMs
507 * synchronize console access with BSD locks:
508 http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html
510 * as soon as we have sender timestamps, revisit coalescing multiple parallel daemon reloads:
511 http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
513 * in systemctl list-unit-files: show the install value the presets would suggest for a service in a third column
515 * figure out when we can use the coarse timers
517 * add "systemctl start -v foobar.service" that shows logs of a service
518 while the start command runs. This is non-trivial to do without
519 races though, since we should flush out all journal messages before
520 returning from the "systemctl stop".
522 * 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
524 * maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment=
526 * consider showing the unit names during boot up in the status output, not just the unit descriptions
528 * maybe allow timer units with an empty Units= setting, so that they
529 can be used for resuming the system but nothing else.
531 * what to do about udev db binary stability for apps? (raw access is not an option)
533 * man: maybe use the word "inspect" rather than "introspect"?
535 * systemctl: if some operation fails, show log output?
537 * systemctl edit: use equvalent of cat() to insert existing config as a comment, prepended with #.
538 Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc.
540 * exponential backoff in timesyncd when we cannot reach a server
542 * timesyncd: add ugly bus calls to set NTP servers per-interface, for usage by NM
544 * merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share....
546 * systemd.show_status= should probably have a mode where only failed
549 * add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL
550 (throughout the codebase, not only PID1)
554 - service registration
555 - service/domain/types browsing
557 - DNS-SD service registration from socket units
558 - resolved should optionally register additional per-interface LLMNR
559 names, so that for the container case we can establish the same name
560 (maybe "host") for referencing the server, everywhere.
561 - allow clients to request DNSSEC for a single lookup even if DNSSEC is off (?)
562 - hook up resolved with machined-based address resolution
564 * refcounting in sd-resolve is borked
566 * Add a new verb "systemctl top"
568 * add new gpt type for btrfs volumes
570 * support empty /etc boots nicely:
571 - nspawn/gpt-generator: introduce new gpt partition type for /usr
573 * generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
575 * a way for container managers to turn off getty starting via $container_headless= or so...
577 * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
579 * For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
580 they run added to the initial transaction and thus confuse Type=idle.
582 * add bus api to query unit file's X fields.
584 * gpt-auto-generator:
585 - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap?
586 - Make /home automount rather than mount?
588 * add generator that pulls in systemd-network from containers when
589 CAP_NET_ADMIN is set, more than the loopback device is defined, even
590 when it is otherwise off
592 * MessageQueueMessageSize= (and suchlike) should use parse_iec_size().
594 * implement Distribute= in socket units to allow running multiple
595 service instances processing the listening socket, and open this up
598 * socket units: support creating sockets in different namespace,
599 opening it up for JoinsNamespaceOf=. This would require to fork off
600 a tiny process that joins the namespace and creates/binds the socket
601 and passes this back to PID1 via SCM_RIGHTS. This also could be used
602 to allow Chown/chgrp on sockets without requiring NSS in PID 1.
604 * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze
605 $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls
606 should SIGSTOP all unit processes in a loop until all processes of
607 it are fully stopped. This can later be used for app management by
608 desktop UIs such as gnome-shell to freeze apps that are not visible
609 on screen, not unlike how job control works on the shell
612 - implement per-slice CPUFairScheduling=1 switch
613 - introduce high-level settings for RT budget, swappiness
614 - how to reset dynamically changed unit cgroup attributes sanely?
615 - when reloading configuration, apply new cgroup configuration
616 - when recursively showing the cgroup hierarchy, optionally also show
617 the hierarchies of child processes
620 - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
622 * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk?
624 * when we detect low battery and no AC on boot, show pretty splash and refuse boot
626 * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
628 * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
630 * 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
632 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
634 * If we try to find a unit via a dangling symlink, generate a clean
635 error. Currently, we just ignore it and read the unit from the search
638 * refuse boot if /usr/lib/os-release is missing or /etc/machine-id cannot be set up
640 * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
642 * load .d/*.conf dropins for device units
644 * allow implementation of InaccessibleDirectories=/ plus
645 ReadOnlyDirectories=... for whitelisting files for a service.
649 - GetAllProperties() on a non-existing object does not result in a failure currently
650 - port to sd-resolve for connecting to TCP dbus servers
651 - 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
652 - see if we can drop more message validation on the sending side
653 - add API to clone sd_bus_message objects
654 - longer term: priority inheritance
656 - NameLost/NameAcquired obsolete
659 - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
662 - allow multiple signal handlers per signal?
663 - document chaining of signal handler for SIGCHLD and child handlers
664 - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ...
665 - generate a failure of a default event loop is executed out-of-thread
667 * investigate endianness issues of UUID vs. GUID
669 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
670 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
672 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
674 * add a pam module that on password changes updates any LUKS slot where the password matches
677 - add unit tests for config_parse_device_allow()
679 * seems that when we follow symlinks to units we prefer the symlink
680 destination path over /etc and /usr. We should not do that. Instead
681 /etc should always override /run+/usr and also any symlink
684 * when isolating, try to figure out a way how we implicitly can order
685 all units we stop before the isolating unit...
687 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
689 * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify
690 that the boot loader should be installed to the ESP. Define a way
691 how an installer can figure out whether a BLS compliant boot loader
694 * think about requeuing jobs when daemon-reload is issued? usecase:
695 the initrd issues a reload after fstab from the host is accessible
696 and we might want to requeue the mounts local-fs acquired through
699 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
701 * remove any syslog support from log.c — we probably cannot do this before split-off udev is gone for good
703 * shutdown logging: store to EFI var, and store to USB stick?
705 * merge unit_kill_common() and unit_kill_context()
707 * introduce ExecCondition= in services
710 - honor language efi variables for default language selection (if there are any?)
711 - honor timezone efi variables for default timezone selection (if there are any?)
712 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
714 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
716 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
718 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
721 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
722 - logind: wakelock/opportunistic suspend support
723 - Add pretty name for seats in logind
724 - logind: allow showing logout dialog from system?
725 - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
726 - if pam_systemd is invoked by su from a process that is outside of a
727 any session we should probably just become a NOP, since that's
728 usually not a real user session but just some system code that just
730 - logind: make the Suspend()/Hibernate() bus calls wait for the for
731 the job to be completed. before returning, so that clients can wait
732 for "systemctl suspend" to finish to know when the suspending is
734 - logind: when the power button is pressed short, just popup a
735 logout dialog. If it is pressed for 1s, do the usual
736 shutdown. Inspiration are Macs here.
737 - expose "Locked" property on logind sesison objects
738 - maybe allow configuration of the StopTimeout for session scopes
739 - rename session scope so that it includes the UID. THat way
740 the session scope can be arranged freely in slices and we don't have
741 make assumptions about their slice anymore.
742 - follow PropertiesChanged state more closely, to deal with quick logouts and
745 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
748 - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields
749 - import and delete pstore filesystem content at startup
750 - journald: also get thread ID from client, plus thread name
751 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
752 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
753 - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively?
754 - declare the local journal protocol stable in the wiki interface chart
755 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
756 - journald: when dropping msgs due to ratelimit make sure to write
757 "dropped %u messages" not only when we are about to print the next
758 message that works, but alraedy after a short tiemout
759 - check if we can make journalctl by default use --follow mode inside of less if called without args?
760 - maybe add API to send pairs of iovecs via sd_journal_send
761 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
762 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
763 and !FOOBAR for events without FOOBAR.
764 - journal: store timestamp of journal_file_set_offline() int he header,
765 so it is possible to display when the file was last synced.
766 - 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.
767 - journal: find a way to allow dropping history early, based on priority, other rules
768 - journal: When used on NFS, check payload hashes
769 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
770 - refuse taking lower-case variable names in sd_journal_send() and friends.
771 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
772 - journal: deal nicely with byte-by-byte copied files, especially regards header
773 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
774 - Replace utmp, wtmp, btmp, and lastlog completely with journal
775 - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
776 - when a kernel driver logs in a tight loop, we should ratelimit that too.
777 - journald: optionally, log debug messages to /run but everything else to /var
778 - journald: when we drop syslog messages because the syslog socket is
779 full, make sure to write how many messages are lost as first thing
780 to syslog when it works again.
781 - change systemd-journal-flush into a service that stays around during
782 boot, and causes the journal to be moved back to /run on shutdown,
783 so that we do not keep /var busy. This needs to happen synchronously,
784 hence doing this via signals is not going to work.
785 - optionally support running journald from the command line for testing purposes in external projects
786 - journald: allow per-priority and per-service retention times when rotating/vacuuming
787 - journald: make use of uid-range.h to managed uid ranges to split
789 - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
790 - improve journalctl performance by loading journal files
791 lazily. Encode just enough information in the file name, so that we
792 do not have to open it to know that it is not interesting for us, for
793 the most common operations.
794 - man: document that corrupted journal files is nothing to act on
795 - rework journald sigbus stuff to use mutex
796 - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our
797 services that run under their own user ids, and use User= (but only
798 in a world where userns is ubiquitous since otherwise we cannot
799 invoke those daemons on the host AND in a container anymore). Also,
800 if LimitNPROC= is used without User= we should warn and refuse
802 - journalctl --verify: don't show files that are currently being
803 written to as FAIL, but instead show that their are being written to.
804 - add journalctl -H that talks via ssh to a remote peer and passes through
806 - add a version of --merge which also merges /var/log/journal/remote
807 - journalctl: -m should access container journals directly by enumerating
808 them via machined, and also watch containers coming and going.
809 Benefit: nspawn --ephemeral would start working nicely with the journal.
810 - assign MESSAGE_ID to log messages about failed services
812 * add a test if all entries in the catalog are properly formatted.
813 (Adding dashes in a catalog entry currently results in the catalog entry
814 being silently skipped. journalctl --update-catalog must warn about this,
815 and we should also have a unit test to check that all our message are OK.)
818 - document that deps in [Unit] sections ignore Alias= fields in
819 [Install] units of other units, unless those units are disabled
820 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
821 - document that service reload may be implemented as service reexec
822 - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
823 - document systemd-journal-flush.service properly
824 - documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
825 - man: document the very specific env the shutdown drop-in tools live in
826 - man: add more examples to man pages
827 - man: maybe sort directives in man pages, and take sections from --help and apply them to man too
830 - add systemctl switch to dump transaction without executing it
831 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
832 - "systemctl disable" on a static unit prints no message and does
833 nothing. "systemctl enable" does nothing, and gives a bad message
834 about it. Should fix both to print nice actionable messages.
835 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
836 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
837 - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
838 - systemctl: "Journal has been rotated since unit was started." message is misleading
839 - systemctl status output should include list of triggering units and their status
842 - "systemctl mask" should find all names by which a unit is accessible
843 (i.e. by scanning for symlinks to it) and link them all to /dev/null
846 - timer units should get the ability to trigger when:
847 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
850 - Modulate timer frequency based on battery state
852 * add libsystemd-password or so to query passwords during boot using the password agent logic
854 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
856 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
858 * make repeated alt-ctrl-del presses printing a dump
860 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
862 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
865 - emulate /dev/kmsg using CUSE and turn off the syslog syscall
866 with seccomp. That should provide us with a useful log buffer that
867 systemd can log to during early boot, and disconnect container logs
868 from the kernel's logs.
869 - as soon as networkd has a bus interface, hook up --network-interface=,
870 --network-bridge= with networkd, to trigger netdev creation should an
872 - a nice way to boot up without machine id set, so that it is set at boot
873 automatically for supporting --ephemeral. Maybe hash the host machine id
874 together with the machine name to generate the machine id for the container
875 - fix logic always print a final newline on output.
876 https://github.com/systemd/systemd/pull/272#issuecomment-113153176
877 - should optionally support receiving WATCHDOG=1 messages from its payload
879 - optionally automatically add FORWARD rules to iptables whenever nspawn is
880 running, remove them when shut down.
883 - refuse mounting over a mount point
884 - automatically discover .roothash files in dissect, similarly to nspawn
887 - add an API so that libvirt-lxc can inform us about network interfaces being
888 removed or added to an existing machine
889 - "machinectl migrate" or similar to copy a container from or to a
890 difference host, via ssh
891 - introduce systemd-nspawn-ephemeral@.service, and hook it into
892 "machinectl start" with a new --ephemeral switch
893 - "machinectl status" should also show internal logs of the container in
895 - "machinectl history"
897 - "machinectl commit" that takes a writable snapshot of a tree, invokes a
898 shell in it, and marks it read-only after use
901 - cryptsetup-generator: allow specification of passwords in crypttab itself
902 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
904 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
905 https://bugs.freedesktop.org/show_bug.cgi?id=54712
907 * add a dependency on standard-conf.xml and other included files to man pages
909 * MountFlags=shared acts as MountFlags=slave right now.
911 * properly handle loop back mounts via fstab, especially regards to fsck/passno
913 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
918 - add trigger --subsystem-match=usb/usb_device device
919 - reimport udev db after MOVE events for devices without dev_t
921 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
923 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
926 - save coredump in Windows/Mozilla minidump format
927 - when truncating coredumps, also log the full size that the process had, and make a metadata field so we can report truncated coredumps
929 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
931 * default to actual 32-bit PIDs, via /proc/sys/kernel/pid_max
933 * be able to specify a forced restart of service A where service B depends on, in case B
934 needs to be auto-respawned?
937 - apply "x" on "D" too (see patch from William Douglas)
939 - instead of ignoring unknown fields, reject them.
940 - creating new directories/subvolumes/fifos/device nodes
941 should not follow symlinks. None of the other adjustment or creation
942 calls follow symlinks.
944 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
946 * verify that the AF_UNIX sockets of a service in the fs still exist
947 when we start a service in order to avoid confusion when a user
948 assumes starting a service is enough to make it accessible
950 * Make it possible to set the keymap independently from the font on
951 the kernel cmdline. Right now setting one resets also the other.
953 * and a dbus call to generate target from current state
955 * write blog stories about:
956 - hwdb: what belongs into it, lsusb
957 - enabling dbus services
958 - how to make changes to sysctl and sysfs attributes
960 - how to pass throw-away units to systemd, or dynamically change properties of existing units
961 - testing with Harald's awesome test kit
963 - how to develop against journal browsing APIs
964 - the journal HTTP iface
965 - non-cgroup resource management
966 - dynamic resource management with cgroups
967 - refreshed, longer missions statement
968 - calendar time events
969 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
970 - how to create your own target
971 - instantiated apache, dovecot and so on
972 - hooking a script into various stages of shutdown/rearly booot
974 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
976 * dot output for --test showing the 'initial transaction'
979 - When logging about multiple units (stopping BoundTo units, conflicts, etc.),
980 log both units as UNIT=, so that journalctl -u triggers on both.
981 - generate better errors when people try to set transient properties
982 that are not supported...
983 http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html
984 - maybe introduce WantsMountsFor=? Usecase:
985 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html
986 - recreate systemd's D-Bus private socket file on SIGUSR2
987 - move PAM code into its own binary
988 - when we automatically restart a service, ensure we restart its rdeps, too.
989 - hide PAM options in fragment parser when compile time disabled
990 - Support --test based on current system state
991 - 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().
992 - after deserializing sockets in socket.c we should reapply sockopts and things
993 - drop PID 1 reloading, only do reexecing (difficult: Reload()
994 currently is properly synchronous, Reexec() is weird, because we
995 cannot delay the response properly until we are back, so instead of
996 being properly synchronous we just keep open the fd and close it
997 when done. That means clients do not get a successful method reply,
998 but much rather a disconnect on success.
999 - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
1000 - when a bus name of a service disappears from the bus make sure to queue further activation requests
1003 - allow port=0 in .socket units
1004 - maybe introduce ExecRestartPre=
1005 - add ReloadSignal= for configuring a reload signal to use
1006 - implement Register= switch in .socket units to enable registration
1007 in Avahi, RPC and other socket registration services.
1008 - allow Type=simple with PIDFile=
1009 https://bugzilla.redhat.com/show_bug.cgi?id=723942
1010 - allow writing multiple conditions in unit files on one line
1011 - load-fragment: when loading a unit file via a chain of symlinks
1012 verify that it is not masked via any of the names traversed.
1013 - introduce Type=pid-file
1014 - introduce mix of BindTo and Requisite
1015 - add a concept of RemainAfterExit= to scope units
1016 - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
1019 - Make sure ID_PATH is always exported and complete for
1020 network devices where possible, so we can safely rely
1022 - check MTUBytes parsing (expecting size_t but we are using unsigned)
1025 - add support for more attribute types
1026 - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places
1029 - add more keys to [Route] and [Address] sections
1030 - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
1031 - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
1032 - add reduced [Link] support to .network files
1033 - add Scope= parsing option for [Network]
1034 - properly handle routerless dhcp leases
1035 - work with non-Ethernet devices
1036 - add support for more bond options
1037 - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from?
1038 - the DHCP lease data (such as NTP/DNS) is still made available when
1039 a carrier is lost on a link. It should be removed instantly.
1040 - expose in the API the following bits:
1041 - option 15, domain name and/or option 119, search list
1042 - option 12, host name and/or option 81, fqdn
1043 - option 123, 144, geolocation
1044 - option 252, configure http proxy (PAC/wpad)
1045 - provide a way to define a per-network interface default metric value
1046 for all routes to it. possibly a second default for DHCP routes.
1047 - allow Name= to be specified repeatedly in the [Match] section. Maybe also
1048 support Name=foo*|bar*|baz ?
1049 - duplicate address check for static IPs (like ARPCHECK in network-scripts)
1050 - allow DUID/IAID to be customized, see issue #394.
1051 - whenever uplink info changes, make DHCP server send out FORCERENEW
1053 * networkd-wait-online:
1054 - make operstates to wait for configurable?
1057 - figure out how much we can increase Maximum Message Size
1060 - add functions to set previously stored IPv6 addresses on startup and get
1061 them at shutdown; store them in client->ia_na
1062 - write more test cases
1063 - implement reconfigure support, see 5.3., 15.11. and 22.20.
1064 - implement support for temporary adressess (IA_TA)
1065 - implement dhcpv6 authentication
1066 - investigate the usefulness of Confirm messages; i.e. are there any
1067 situations where the link changes without any loss in carrier detection
1069 - some servers don't do rapid commit without a filled in IA_NA, verify
1076 - natively watch for dbus-*.service symlinks (PENDING)
1077 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
1079 * fix alsa mixer restore to not print error when no config is stored
1081 * make cryptsetup lower --iter-time
1083 * patch kernel for xattr support in /dev, /proc/, /sys?
1085 * kernel: add device_type = "fb", "fbcon" to class "graphics"
1087 * /usr/bin/service should actually show the new command line
1089 * 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=
1091 * fedora: F20: go timer units all the way, leave cron.daily for cron
1093 * neither pkexec nor sudo initialize environ[] from the PAM environment?
1095 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
1097 * register catalog database signature as file magic
1099 * zsh shell completion:
1100 - <command> <verb> -<TAB> should complete options, but currently does not
1101 - systemctl add-wants,add-requires
1106 * look for close() vs. close_nointr() vs. close_nointr_nofail()
1108 * check for strerror(r) instead of strerror(-r)
1112 * set_put(), hashmap_put() return values check. i.e. == 0 does not free()!
1114 * use secure_getenv() instead of getenv() where appropriate
1116 * link up selected blog stories from man pages and unit files Documentation= fields