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 * maybe add a seccomp-based high-level filter that blocks creation of suid/sgid
29 * make MAINPID= message reception checks even stricter: if service uses User=,
30 then check sending UID and ignore message if it doesn't match the user or
33 * maybe trigger a uevent "change" on a device if "systemctl reload xyz.device"
36 * when importing an fs tree with machined, optionally apply userns-rec-chown
38 * when importing an fs tree with machined, complain if image is not an OS
40 * Maybe introduce a helper safe_exec() or so, which is to execve() which
41 safe_fork() is to fork(). And then make revert the RLIMIT_NOFILE soft limit
42 to 1K implicitly, unless explicitly opted-out.
44 * rework seccomp/nnp logic that that even if User= is used in combination with
45 a seccomp option we don't have to set NNP. For that, change uid first whil
46 keeping CAP_SYS_ADMIN, then apply seccomp, the drop cap.
48 * add a concept for automatically loading per-unit secrets off disk and
49 inserting them into the kernel keyring. Maybe SecretsDirectory= similar to
50 ConfigurationDirectory=.
52 * when no locale is configured, default to UEFI's PlatformLang variable
54 * When logind.conf contains HandleLidSwitch=suspend-then-hibernate and we can't
55 hibernate because the swap partition isn't large enough, still suspend
57 * bootctl,sd-boot: actually honour the "architecture" key
59 * when a socket unit is spawned with an AF_UNIX path in /var/run, complain and
60 patch it to use /run instead
62 * set memory.oom.group in cgroup v2 for all leaf cgroups (kernel v4.19+)
64 * add a new syscall group "@esoteric" for more esoteric stuff such as bpf() and
65 usefaultd() and make systemd-analyze check for it.
67 * paranoia: whenever we process passwords, call mlock() on the memory
68 first. i.e. look for all places we use string_erase()/string_free_erase() and
69 augment them with mlock(). Also use MADV_DONTDUMP
71 * whenever oom_kill memory.event event is triggered print a nice log message
73 * Move RestrictAddressFamily= to the new cgroup create socket
75 * support the bind/connect/sendmsg cgroup stuff for sandboxing, and possibly
78 * maybe implicitly attach monotonic+realtime timestamps to outgoing messages in
79 log.c and sd-journal-send
81 * chown() tty a service is attached to after the service goes down
83 * optionally: turn on cgroup delegation for per-session scope units
85 * introduce per-unit (i.e. per-slice, per-service) journal log size limits.
87 * optionally, if a per-partition GPT flag is set for the root/home/… partitions
88 format the partition on next boot and unset the flag, in order to implement
89 factory reset. also, add a second flag that simply indicates whether such a
90 scheme is supported. then, add a tool (or maybe beef up systemd-dissect) to
91 show state of these flags, and optionally trigger such a factory reset on
92 next boot by setting the flag.
94 * sd-boot: automatically load EFI modules from some drop-in dir, so that people
95 can add in file system drivers and such
97 * sd-boot: optionally, show boot menu when previous default boot item has
98 non-zero "tries done" count
100 * maybe set a special xattr on cgroups that have delegate=yes set, to make it
101 easy to mark cut points
103 * introduce an option (or replacement) for "systemctl show" that outputs all
104 properties as JSON, similar to busctl's new JSON output. In contrast to that
105 it should skip the variant type string though.
107 * augment CODE_FILE=, CODE_LINE= with something like CODE_BASE= or so which
108 contains some identifier for the project, which allows us to include
109 clickable links to source files generating these log messages. The identifier
110 could be some abberviated URL prefix or so (taking inspiration from Go
111 imports). For example, for systemd we could use
112 CODE_BASE=github.com/systemd/systemd/blob/98b0b1123cc or so which is
113 sufficient to build a link by prefixing "http://" and suffixing the
116 * when outputting log data with journalctl and the log data includes references
117 to configuration files (CONFIG_FILE=), create a clickable link for it.
119 * Augment MESSAGE_ID with MESSAGE_BASE, in a similar fashion so that we can
120 make clickable links from log messages carrying a MESSAGE_ID, that lead to
121 some explanatory text online.
123 * maybe extend .path units to expose fanotify() per-mount change events
125 * Add a "systemctl list-units --by-slice" mode or so, which rearranges the
126 output of "systemctl list-units" slightly by showing the tree structure of
127 the slices, and the units attached to them.
129 * the a-posteriori stopping of units bound to units that disappeared logic
130 should be reworked: there should be a queue of units, and we should only
131 enqeue stop jobs from a defer event that processes queue instead of
132 right-away when we find a unit that is bound to one that doesn't exist
133 anymore. (similar to how the stop-unneeded queue has been reworked the same
136 * nspawn: make nspawn suitable for shell pipelines: instead of triggering a
137 hangup when input is finished, send ^D, which synthesizes an EOF. Then wait
138 for hangup or ^D before passing on the EOF.
140 * When reloading configuration PID 1 should reset all its properties to the
141 original defaults before calling parse_config()
143 * Add OnTimezoneChange= and OnTimeChange= stanzas to .timer units in order to
144 schedule events based on time and timezone changes.
146 * nspawn: greater control over selinux label?
148 * hibernate/s2h: make this robust and safe to enable in Fedora by default.
151 1. add resume_offset support to the resume code (i.e. support swap files
153 2. check if swap is on weird storage and refuse if so
154 3. add autodetection of hibernation images
156 * cgroups: use inotify to get notified when somebody else modifies cgroups
157 owned by us, then log a friendly warning.
159 * beef up log.c with support for stripping ANSI sequences from strings, so that
160 it is OK to include them in log strings. This would be particularly useful so
161 that our log messages could contain clickable links for example for unit
162 files and suchlike we operate on.
164 * add support for "portablectl attach http://foobar.com/waaa.raw (i.e. importd integration)
166 * add attach --enable and attach --now (for attach+enable+start)
168 * sync dynamic uids/gids between host+portable srvice (i.e. if DynamicUser=1 is set for a service, make sure that the
169 selected user is resolvable in the service even if it ships its own /etc/passwd)
171 * Fix DECIMAL_STR_MAX or DECIMAL_STR_WIDTH. One includes a trailing NUL, the
172 other doesn't. What a desaster. Probably to exclude it. Also
173 DECIMAL_STR_WIDTH should probably add an extra "-" into account for negative
176 * Check that users of inotify's IN_DELETE_SELF flag are using it properly, as
177 usually IN_ATTRIB is the right way to watch deleted files, as the former only
178 fires when a file is actually removed from disk, i.e. the link count drops to
179 zero and is not open anymore, while the latter happens when a file is
180 unlinked from any dir.
182 * port systemctl, busctl, … over to format-table.[ch]'s table formatters
184 * pid1: lock image configured with RootDirectory=/RootImage= using the usual nspawn semantics while the unit is up
186 * add --vacuum-xyz options to coredumpctl, matching those journalctl already has.
188 * SuccessExitStatus= and friends should probably also accept symbolic exit
189 codes names, i.e. error codes from the list maintained in exit-codes.[ch]
191 * introduce Ephemeral= unit file switch, that creates an ephemeral copy of all
192 files and directories that are left writable for a unit, and which are
193 removed after the unit goes down again. A bit like --ephemeral for
194 systemd-nspawn but for system services. If used together with RootImage= this
195 should reflink the image file itself.
197 Related: add Ephemeral=<path1> <path2> … which would allow marking
198 specific paths only like this.
200 * add CopyFile= or so as unit file setting that may be used to copy files or
201 directory trees from the host to te services RootImage= and RootDirectory=
202 environment. Which we can use for /etc/machine-id and in particular
203 /etc/resolv.conf. Should be smart and do something useful on read-only
204 images, for example fallback to read-only bind mounting the file instead.
206 * nspawn's console TTY should be allocated from within the container, not
207 mounted in from the outside
209 * show invocation ID in systemd-run output
211 * bypass SIGTERM state in unit files if KillSignal is SIGKILL
213 * tree-wide: ensure we always block the signals we hook into with
214 sd_event_add_signal() first
216 * add proper dbus APIs for the various sd_notify() commands, such as MAINPID=1
217 and so on, which would mean we could report errors and such.
219 * teach tmpfiles.d q/Q logic something sensible in the context of XFS/ext4
222 * introduce DefaultSlice= or so in system.conf that allows changing where we
223 place our units by default, i.e. change system.slice to something
224 else. Similar, ManagerSlice= should exist so that PID1's own scope unit could
225 be moved somewhere else too. Finally machined and logind should get similar
226 options so that it is possible to move user session scopes and machines to a
227 different slice too by default. Usecase: people who want to put resources on
228 the entire system, with the exception of one specific service. See:
229 https://lists.freedesktop.org/archives/systemd-devel/2018-February/040369.html
231 * maybe rework get_user_creds() to query the user database if $SHELL is used
232 for root, but only then.
234 * be stricter with fds we receive for the fdstore: close them asynchronously
236 * calenderspec: add support for week numbers and day numbers within a
237 year. This would allow us to define "bi-weekly" triggers safely.
239 * sd-bus: add vtable flag, that may be used to request client creds implicitly
240 and asynchronously before dispatching the operation
242 * make use of ethtool veth peer info in machined, for automatically finding out
243 host-side interface pointing to the container.
245 * add some special mode to LogsDirectory=/StateDirectory=… that allows
246 declaring these directories without necessarily pulling in deps for them, or
247 creating them when starting up. That way, we could declare that
248 systemd-journald writes to /var/log/journal, which could be useful when we
249 doing disk usage calculations and so on.
251 * taint systemd if there are fewer than 65536 users assigned (userns) to the system.
253 * deprecate RootDirectoryStartOnly= in favour of a new ExecStart= prefix char
255 * add a new RuntimeDirectoryPreserve= mode that defines a similar lifecycle for
256 the runtime dir as we maintain for the fdstore: i.e. keep it around as long
257 as the unit is running or has a job queued.
259 * support projid-based quota in machinectl for containers
261 * Add NetworkNamespacePath= to specify a path to a network namespace
263 * maybe use SOURCE_DATE_EPOCH (i.e. the env var the reproducible builds folks
264 introduced) as the RTC epoch, instead of the mtime of NEWS.
266 * add a way to lock down cgroup migration: a boolean, which when set for a unit
267 makes sure the processes in it can never migrate out of it
269 * blog about fd store and restartable services
271 * document Environment=SYSTEMD_LOG_LEVEL=debug drop-in in debugging document
273 * rework ExecOutput and ExecInput enums so that EXEC_OUTPUT_NULL loses its
274 magic meaning and is no longer upgraded to something else if set explicitly.
276 * in the long run: permit a system with /etc/machine-id linked to /dev/null, to
277 make it lose its identity, i.e. be anonymous. For this we'd have to patch
278 through the whole tree to make all code deal with the case where no machine
281 * optionally, collect cgroup resource data, and store it in per-unit RRD files,
282 suitable for processing with rrdtool. Add bus API to access this data, and
283 possibly implement a CPULoad property based on it.
285 * beef up pam_systemd to take unit file settings such as cgroups properties as
288 * maybe hook of xfs/ext4 quotactl() with services? i.e. automatically manage
289 the quota of a the user indicated in User= via unit file settings, like the
290 other resource management concepts. Would mix nicely with DynamicUser=1. Or
291 alternatively, do this with projids, so that we can also cover services
292 running as root. Quota should probably cover all the special dirs such as
293 StateDirectory=, LogsDirectory=, CacheDirectory=, as well as RootDirectory= if it
294 is set, plus the whole disk space any image configured with RootImage=.
296 * In DynamicUser= mode: before selecting a UID, use disk quota APIs on relevant
297 disks to see if the UID is already in use.
299 * add "systemctl wait" or so, which does what "systemd-run --wait" does, but
300 for all units. It should be both a way to pin units into memory as well as a
301 wait to retrieve their exit data.
303 * expose IO accounting data on the bus, show it in systemd-run --wait and log
304 about it in the resource log message
306 * add "systemctl purge" for flushing out configuration, state, logs, ... of a
307 unit when it is stopped
309 * show whether a service has out-of-date configuration in "systemctl status" by
310 using mtime data of ConfigurationDirectory=.
312 * replace all remaining uses of fgets() + LINE_MAX by read_line()
314 * Add AddUser= setting to unit files, similar to DynamicUser=1 which however
315 creates a static, persistent user rather than a dynamic, transient user. We
316 can leverage code from sysusers.d for this.
318 * add some optional flag to ReadWritePaths= and friends, that has the effect
319 that we create the dir in question when the service is started. Example:
321 ReadWritePaths=:/var/lib/foobar
323 * maybe add call sd_journal_set_block_timeout() or so to set SO_SNDTIMEO for
324 the sd-journal logging socket, and, if the timeout is set to 0, sets
325 O_NONBLOCK on it. That way people can control if and when to block for
328 * hostnamed: populate form factor data from a new hwdb database, so that old
329 yogas can be recognized as "convertible" too, even if they predate the DMI
330 "convertible" form factor
332 * Maybe add a small tool invoked early at boot, that adds in or resizes
333 partitions automatically, to be used when the media used is actually larger
334 than the image written onto it is.
336 * Maybe add PrivatePIDs= as new unit setting, and do minimal PID namespacing
337 after all. Be strict however, only support the equivalent of nspawn's
338 --as-pid2 switch, and sanely proxy sd_notify() messages dropping stuff such
341 * Add ExecMonitor= setting. May be used multiple times. Forks off a process in
342 the service cgroup, which is supposed to monitor the service, and when it
343 exits the service is considered failed by its monitor.
345 * track the per-service PAM process properly (i.e. as an additional control
346 process), so that it may be queried on the bus and everything.
348 * add a new "debug" job mode, that is propagated to unit_start() and for
349 services results in two things: we raise SIGSTOP right before invoking
350 execve() and turn off watchdog support. Then, use that to implement
351 "systemd-gdb" for attaching to the start-up of any system service in its
354 * maybe introduce gpt auto discovery for /var/tmp?
356 * maybe add gpt-partition-based user management: each user gets his own
357 LUKS-encrypted GPT partition with a new GPT type. A small nss module
358 enumerates users via udev partition enumeration. UIDs are assigned in a fixed
359 way: the partition index is added as offset to some fixed base uid. User name
360 is stored in GPT partition name. A PAM module authenticates the user via the
361 LUKS partition password. Benefits: strong per-user security, compatibility
362 with stateless/read-only/verity-enabled root. (other idea: do this based on
363 loopback files in /home, without GPT involvement)
365 * gpt-auto logic: introduce support for discovering /var matching an image. For
366 that, use a partition type UUID that is hashed from the OS name (as encoded
367 in /etc/os-release), the architecture, and 4 new bits from the gpt flags
368 field of the root partition. This way can easily support multiple OS
369 installations on the same GPT partition table, without problems with
370 unmatched /var partitions.
372 * gpt-auto logic: related to the above, maybe support a "secondary" root
373 partition, that is mounted to / and is writable, and where the actual root's
374 /usr is mounted into.
376 * 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
378 * drop nss-myhostname in favour of nss-resolve?
380 * add a percentage syntax for TimeoutStopSec=, e.g. TimeoutStopSec=150%, and
381 then use that for the setting used in user@.service. It should be understood
382 relative to the configured default value.
384 * in networkd, when matching device types, fix up DEVTYPE rubbish the kernel passes to us
386 * enable LockMLOCK to take a percentage value relative to physical memory
388 * Permit masking specific netlink APIs with RestrictAddressFamily=
390 * nspawn: support that /proc, /sys/, /dev are pre-mounted
392 * define gpt header bits to select volatility mode
394 * ProtectKernelLogs= (drops CAP_SYSLOG, add seccomp for syslog() syscall, and DeviceAllow to /dev/kmsg) in service files
396 * ProtectClock= (drops CAP_SYS_TIMES, adds seecomp filters for settimeofday, adjtimex), sets DeviceAllow o /dev/rtc
398 * ProtectTracing= (drops CAP_SYS_PTRACE, blocks ptrace syscall, makes /sys/kernel/tracing go away)
400 * ProtectMount= (drop mount/umount/pivot_root from seccomp, disallow fuse via DeviceAllow, imply Mountflags=slave)
402 * ProtectKeyRing= to take keyring calls away
404 * RemoveKeyRing= to remove all keyring entries of the specified user
406 * ProtectReboot= that masks reboot() and kexec_load() syscalls, prohibits kill
407 on PID 1 with the relevant signals, and makes relevant files in /sys and
408 /proc (such as the sysrq stuff) unavailable
410 * make sure the ratelimit object can deal with USEC_INFINITY as way to turn off things
412 * journalctl: make sure -f ends when the container indicated by -M terminates
414 * mount: automatically search for "main" partition of an image has multiple
417 * expose the "privileged" flag of ExecCommand on the bus, and open it up to
420 * in nss-systemd, if we run inside of RootDirectory= with PrivateUsers= set,
421 find a way to map the User=/Group= of the service to the right name. This way
422 a user/group for a service only has to exist on the host for the right
425 * add bus API for creating unit files in /etc, reusing the code for transient units
427 * add bus API to remove unit files from /etc
429 * add bus API to retrieve current unit file contents (i.e. implement "systemctl cat" on the bus only)
431 * rework fopen_temporary() to make use of open_tmpfile_linkable() (problem: the
432 kernel doesn't support linkat() that replaces existing files, currently)
434 * transient units: don't bother with actually setting unit properties, we
435 reload the unit file anyway
437 * journald: sigbus API via a signal-handler safe function that people may call
438 from the SIGBUS handler
440 * optionally, also require WATCHDOG=1 notifications during service start-up and shutdown
442 * resolved: when routing queries, make sure only look for the *longest* suffix...
444 * delay activation of logind until somebody logs in, or when /dev/tty0 pulls it
445 in or lingering is on (so that containers don't bother with it until PAM is used). also exit-on-idle
447 * cache sd_event_now() result from before the first iteration...
449 * add systemctl stop --job-mode=triggering that follows TRIGGERED_BY deps and adds them to the same transaction
451 * PID1: find a way how we can reload unit file configuration for
452 specific units only, without reloading the whole of systemd
454 * add an explicit parser for LimitRTPRIO= that verifies
455 the specified range and generates sane error messages for incorrect
458 * when we detect that there are waiting jobs but no running jobs, do something
460 * push CPUAffinity= also into the "cpuset" cgroup controller (only after the cpuset controller got ported to the unified hierarchy)
462 * PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn)
464 * there's probably something wrong with having user mounts below /sys,
465 as we have for debugfs. for exmaple, src/core/mount.c handles mounts
466 prefixed with /sys generally special.
467 http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html
469 * fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline
471 * docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date
473 * add a job mode that will fail if a transaction would mean stopping
474 running units. Use this in timedated to manage the NTP service
476 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html
478 * The udev blkid built-in should expose a property that reflects
479 whether media was sensed in USB CF/SD card readers. This should then
480 be used to control SYSTEMD_READY=1/0 so that USB card readers aren't
481 picked up by systemd unless they contain a medium. This would mirror
482 the behaviour we already have for CD drives.
484 * networkd/udev: implement SR_IOV configuration in .link files:
485 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html
487 * hostnamectl: show root image uuid
489 * Find a solution for SMACK capabilities stuff:
490 http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html
492 * "systemctl preset-all" should probably order the unit files it
493 operates on lexicographically before starting to work, in order to
494 ensure deterministic behaviour if two unit files conflict (like DMs
497 * synchronize console access with BSD locks:
498 http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html
500 * as soon as we have sender timestamps, revisit coalescing multiple parallel daemon reloads:
501 http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
503 * in systemctl list-unit-files: show the install value the presets would suggest for a service in a third column
505 * figure out when we can use the coarse timers
507 * add "systemctl start -v foobar.service" that shows logs of a service
508 while the start command runs. This is non-trivial to do without
509 races though, since we should flush out all journal messages before
510 returning from the "systemctl stop".
512 * 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
514 * maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment=
516 * consider showing the unit names during boot up in the status output, not just the unit descriptions
518 * maybe allow timer units with an empty Units= setting, so that they
519 can be used for resuming the system but nothing else.
521 * what to do about udev db binary stability for apps? (raw access is not an option)
523 * man: maybe use the word "inspect" rather than "introspect"?
525 * systemctl: if some operation fails, show log output?
527 * systemctl edit: use equvalent of cat() to insert existing config as a comment, prepended with #.
528 Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc.
530 * exponential backoff in timesyncd when we cannot reach a server
532 * timesyncd: add ugly bus calls to set NTP servers per-interface, for usage by NM
534 * merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share....
536 * systemd.show_status= should probably have a mode where only failed
539 * add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL
540 (throughout the codebase, not only PID1)
544 - service registration
545 - service/domain/types browsing
547 - DNS-SD service registration from socket units
548 - resolved should optionally register additional per-interface LLMNR
549 names, so that for the container case we can establish the same name
550 (maybe "host") for referencing the server, everywhere.
551 - allow clients to request DNSSEC for a single lookup even if DNSSEC is off (?)
552 - hook up resolved with machined-based address resolution
554 * refcounting in sd-resolve is borked
556 * Add a new verb "systemctl top"
558 * add new gpt type for btrfs volumes
560 * support empty /etc boots nicely:
561 - nspawn/gpt-generator: introduce new gpt partition type for /usr
563 * generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
565 * a way for container managers to turn off getty starting via $container_headless= or so...
567 * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
569 * For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
570 they run added to the initial transaction and thus confuse Type=idle.
572 * add bus api to query unit file's X fields.
574 * gpt-auto-generator:
575 - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap?
576 - Make /home automount rather than mount?
578 * add generator that pulls in systemd-network from containers when
579 CAP_NET_ADMIN is set, more than the loopback device is defined, even
580 when it is otherwise off
582 * MessageQueueMessageSize= (and suchlike) should use parse_iec_size().
584 * implement Distribute= in socket units to allow running multiple
585 service instances processing the listening socket, and open this up
588 * socket units: support creating sockets in different namespace,
589 opening it up for JoinsNamespaceOf=. This would require to fork off
590 a tiny process that joins the namespace and creates/binds the socket
591 and passes this back to PID1 via SCM_RIGHTS. This also could be used
592 to allow Chown/chgrp on sockets without requiring NSS in PID 1.
594 * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze
595 $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls
596 should SIGSTOP all unit processes in a loop until all processes of
597 it are fully stopped. This can later be used for app management by
598 desktop UIs such as gnome-shell to freeze apps that are not visible
599 on screen, not unlike how job control works on the shell
602 - implement per-slice CPUFairScheduling=1 switch
603 - introduce high-level settings for RT budget, swappiness
604 - how to reset dynamically changed unit cgroup attributes sanely?
605 - when reloading configuration, apply new cgroup configuration
606 - when recursively showing the cgroup hierarchy, optionally also show
607 the hierarchies of child processes
610 - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
612 * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk?
614 * when we detect low battery and no AC on boot, show pretty splash and refuse boot
616 * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
618 * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
620 * 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
622 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
624 * If we try to find a unit via a dangling symlink, generate a clean
625 error. Currently, we just ignore it and read the unit from the search
628 * refuse boot if /usr/lib/os-release is missing or /etc/machine-id cannot be set up
630 * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
632 * load .d/*.conf dropins for device units
634 * allow implementation of InaccessibleDirectories=/ plus
635 ReadOnlyDirectories=... for whitelisting files for a service.
639 - GetAllProperties() on a non-existing object does not result in a failure currently
640 - port to sd-resolve for connecting to TCP dbus servers
641 - 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
642 - see if we can drop more message validation on the sending side
643 - add API to clone sd_bus_message objects
644 - longer term: priority inheritance
646 - NameLost/NameAcquired obsolete
649 - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
652 - allow multiple signal handlers per signal?
653 - document chaining of signal handler for SIGCHLD and child handlers
654 - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ...
655 - generate a failure of a default event loop is executed out-of-thread
657 * investigate endianness issues of UUID vs. GUID
659 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
660 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
662 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
664 * add a pam module that on password changes updates any LUKS slot where the password matches
667 - add unit tests for config_parse_device_allow()
669 * seems that when we follow symlinks to units we prefer the symlink
670 destination path over /etc and /usr. We should not do that. Instead
671 /etc should always override /run+/usr and also any symlink
674 * when isolating, try to figure out a way how we implicitly can order
675 all units we stop before the isolating unit...
677 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
679 * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify
680 that the boot loader should be installed to the ESP. Define a way
681 how an installer can figure out whether a BLS compliant boot loader
684 * think about requeuing jobs when daemon-reload is issued? usecase:
685 the initrd issues a reload after fstab from the host is accessible
686 and we might want to requeue the mounts local-fs acquired through
689 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
691 * remove any syslog support from log.c — we probably cannot do this before split-off udev is gone for good
693 * shutdown logging: store to EFI var, and store to USB stick?
695 * merge unit_kill_common() and unit_kill_context()
697 * introduce ExecCondition= in services
700 - honor language efi variables for default language selection (if there are any?)
701 - honor timezone efi variables for default timezone selection (if there are any?)
702 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
704 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
706 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
708 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
711 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
712 - logind: wakelock/opportunistic suspend support
713 - Add pretty name for seats in logind
714 - logind: allow showing logout dialog from system?
715 - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
716 - if pam_systemd is invoked by su from a process that is outside of a
717 any session we should probably just become a NOP, since that's
718 usually not a real user session but just some system code that just
720 - logind: make the Suspend()/Hibernate() bus calls wait for the for
721 the job to be completed. before returning, so that clients can wait
722 for "systemctl suspend" to finish to know when the suspending is
724 - logind: when the power button is pressed short, just popup a
725 logout dialog. If it is pressed for 1s, do the usual
726 shutdown. Inspiration are Macs here.
727 - expose "Locked" property on logind sesison objects
728 - maybe allow configuration of the StopTimeout for session scopes
729 - rename session scope so that it includes the UID. THat way
730 the session scope can be arranged freely in slices and we don't have
731 make assumptions about their slice anymore.
732 - follow PropertiesChanged state more closely, to deal with quick logouts and
735 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
738 - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields
739 - import and delete pstore filesystem content at startup
740 - journald: also get thread ID from client, plus thread name
741 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
742 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
743 - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively?
744 - declare the local journal protocol stable in the wiki interface chart
745 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
746 - journald: when dropping msgs due to ratelimit make sure to write
747 "dropped %u messages" not only when we are about to print the next
748 message that works, but alraedy after a short tiemout
749 - check if we can make journalctl by default use --follow mode inside of less if called without args?
750 - maybe add API to send pairs of iovecs via sd_journal_send
751 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
752 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
753 and !FOOBAR for events without FOOBAR.
754 - journal: store timestamp of journal_file_set_offline() int he header,
755 so it is possible to display when the file was last synced.
756 - 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.
757 - journal: find a way to allow dropping history early, based on priority, other rules
758 - journal: When used on NFS, check payload hashes
759 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
760 - refuse taking lower-case variable names in sd_journal_send() and friends.
761 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
762 - journal: deal nicely with byte-by-byte copied files, especially regards header
763 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
764 - Replace utmp, wtmp, btmp, and lastlog completely with journal
765 - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
766 - when a kernel driver logs in a tight loop, we should ratelimit that too.
767 - journald: optionally, log debug messages to /run but everything else to /var
768 - journald: when we drop syslog messages because the syslog socket is
769 full, make sure to write how many messages are lost as first thing
770 to syslog when it works again.
771 - change systemd-journal-flush into a service that stays around during
772 boot, and causes the journal to be moved back to /run on shutdown,
773 so that we do not keep /var busy. This needs to happen synchronously,
774 hence doing this via signals is not going to work.
775 - optionally support running journald from the command line for testing purposes in external projects
776 - journald: allow per-priority and per-service retention times when rotating/vacuuming
777 - journald: make use of uid-range.h to managed uid ranges to split
779 - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
780 - improve journalctl performance by loading journal files
781 lazily. Encode just enough information in the file name, so that we
782 do not have to open it to know that it is not interesting for us, for
783 the most common operations.
784 - man: document that corrupted journal files is nothing to act on
785 - rework journald sigbus stuff to use mutex
786 - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our
787 services that run under their own user ids, and use User= (but only
788 in a world where userns is ubiquitous since otherwise we cannot
789 invoke those daemons on the host AND in a container anymore). Also,
790 if LimitNPROC= is used without User= we should warn and refuse
792 - journalctl --verify: don't show files that are currently being
793 written to as FAIL, but instead show that their are being written to.
794 - add journalctl -H that talks via ssh to a remote peer and passes through
796 - add a version of --merge which also merges /var/log/journal/remote
797 - journalctl: -m should access container journals directly by enumerating
798 them via machined, and also watch containers coming and going.
799 Benefit: nspawn --ephemeral would start working nicely with the journal.
800 - assign MESSAGE_ID to log messages about failed services
802 * add a test if all entries in the catalog are properly formatted.
803 (Adding dashes in a catalog entry currently results in the catalog entry
804 being silently skipped. journalctl --update-catalog must warn about this,
805 and we should also have a unit test to check that all our message are OK.)
808 - document that deps in [Unit] sections ignore Alias= fields in
809 [Install] units of other units, unless those units are disabled
810 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
811 - document that service reload may be implemented as service reexec
812 - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
813 - document systemd-journal-flush.service properly
814 - documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
815 - man: document the very specific env the shutdown drop-in tools live in
816 - man: add more examples to man pages
817 - man: maybe sort directives in man pages, and take sections from --help and apply them to man too
818 - document root=gpt-auto properly
821 - add systemctl switch to dump transaction without executing it
822 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
823 - "systemctl disable" on a static unit prints no message and does
824 nothing. "systemctl enable" does nothing, and gives a bad message
825 about it. Should fix both to print nice actionable messages.
826 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
827 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
828 - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
829 - systemctl: "Journal has been rotated since unit was started." message is misleading
830 - systemctl status output should include list of triggering units and their status
833 - "systemctl mask" should find all names by which a unit is accessible
834 (i.e. by scanning for symlinks to it) and link them all to /dev/null
837 - timer units should get the ability to trigger when:
838 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
841 - Modulate timer frequency based on battery state
843 * add libsystemd-password or so to query passwords during boot using the password agent logic
845 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
847 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
849 * make repeated alt-ctrl-del presses printing a dump
851 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
853 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
856 - emulate /dev/kmsg using CUSE and turn off the syslog syscall
857 with seccomp. That should provide us with a useful log buffer that
858 systemd can log to during early boot, and disconnect container logs
859 from the kernel's logs.
860 - as soon as networkd has a bus interface, hook up --network-interface=,
861 --network-bridge= with networkd, to trigger netdev creation should an
863 - a nice way to boot up without machine id set, so that it is set at boot
864 automatically for supporting --ephemeral. Maybe hash the host machine id
865 together with the machine name to generate the machine id for the container
866 - fix logic always print a final newline on output.
867 https://github.com/systemd/systemd/pull/272#issuecomment-113153176
868 - should optionally support receiving WATCHDOG=1 messages from its payload
870 - optionally automatically add FORWARD rules to iptables whenever nspawn is
871 running, remove them when shut down.
874 - refuse mounting over a mount point
875 - automatically discover .roothash files in dissect, similarly to nspawn
878 - add an API so that libvirt-lxc can inform us about network interfaces being
879 removed or added to an existing machine
880 - "machinectl migrate" or similar to copy a container from or to a
881 difference host, via ssh
882 - introduce systemd-nspawn-ephemeral@.service, and hook it into
883 "machinectl start" with a new --ephemeral switch
884 - "machinectl status" should also show internal logs of the container in
886 - "machinectl history"
888 - "machinectl commit" that takes a writable snapshot of a tree, invokes a
889 shell in it, and marks it read-only after use
892 - cryptsetup-generator: allow specification of passwords in crypttab itself
893 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
895 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
896 https://bugs.freedesktop.org/show_bug.cgi?id=54712
898 * add a dependency on standard-conf.xml and other included files to man pages
900 * MountFlags=shared acts as MountFlags=slave right now.
902 * properly handle loop back mounts via fstab, especially regards to fsck/passno
904 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
909 - add trigger --subsystem-match=usb/usb_device device
910 - reimport udev db after MOVE events for devices without dev_t
912 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
914 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
917 - save coredump in Windows/Mozilla minidump format
918 - when truncating coredumps, also log the full size that the process had, and make a metadata field so we can report truncated coredumps
920 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
922 * default to actual 32-bit PIDs, via /proc/sys/kernel/pid_max
924 * be able to specify a forced restart of service A where service B depends on, in case B
925 needs to be auto-respawned?
928 - apply "x" on "D" too (see patch from William Douglas)
930 - instead of ignoring unknown fields, reject them.
931 - creating new directories/subvolumes/fifos/device nodes
932 should not follow symlinks. None of the other adjustment or creation
933 calls follow symlinks.
935 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
937 * verify that the AF_UNIX sockets of a service in the fs still exist
938 when we start a service in order to avoid confusion when a user
939 assumes starting a service is enough to make it accessible
941 * Make it possible to set the keymap independently from the font on
942 the kernel cmdline. Right now setting one resets also the other.
944 * and a dbus call to generate target from current state
946 * write blog stories about:
947 - hwdb: what belongs into it, lsusb
948 - enabling dbus services
949 - how to make changes to sysctl and sysfs attributes
951 - how to pass throw-away units to systemd, or dynamically change properties of existing units
952 - testing with Harald's awesome test kit
954 - how to develop against journal browsing APIs
955 - the journal HTTP iface
956 - non-cgroup resource management
957 - dynamic resource management with cgroups
958 - refreshed, longer missions statement
959 - calendar time events
960 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
961 - how to create your own target
962 - instantiated apache, dovecot and so on
963 - hooking a script into various stages of shutdown/rearly booot
965 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
967 * dot output for --test showing the 'initial transaction'
970 - When logging about multiple units (stopping BoundTo units, conflicts, etc.),
971 log both units as UNIT=, so that journalctl -u triggers on both.
972 - generate better errors when people try to set transient properties
973 that are not supported...
974 http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html
975 - maybe introduce WantsMountsFor=? Usecase:
976 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html
977 - recreate systemd's D-Bus private socket file on SIGUSR2
978 - move PAM code into its own binary
979 - when we automatically restart a service, ensure we restart its rdeps, too.
980 - hide PAM options in fragment parser when compile time disabled
981 - Support --test based on current system state
982 - 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().
983 - after deserializing sockets in socket.c we should reapply sockopts and things
984 - drop PID 1 reloading, only do reexecing (difficult: Reload()
985 currently is properly synchronous, Reexec() is weird, because we
986 cannot delay the response properly until we are back, so instead of
987 being properly synchronous we just keep open the fd and close it
988 when done. That means clients do not get a successful method reply,
989 but much rather a disconnect on success.
990 - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
991 - when a bus name of a service disappears from the bus make sure to queue further activation requests
994 - allow port=0 in .socket units
995 - maybe introduce ExecRestartPre=
996 - add ReloadSignal= for configuring a reload signal to use
997 - implement Register= switch in .socket units to enable registration
998 in Avahi, RPC and other socket registration services.
999 - allow Type=simple with PIDFile=
1000 https://bugzilla.redhat.com/show_bug.cgi?id=723942
1001 - allow writing multiple conditions in unit files on one line
1002 - load-fragment: when loading a unit file via a chain of symlinks
1003 verify that it is not masked via any of the names traversed.
1004 - introduce Type=pid-file
1005 - introduce mix of BindTo and Requisite
1006 - add a concept of RemainAfterExit= to scope units
1007 - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
1010 - Make sure ID_PATH is always exported and complete for
1011 network devices where possible, so we can safely rely
1013 - check MTUBytes parsing (expecting size_t but we are using unsigned)
1016 - add support for more attribute types
1017 - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places
1020 - add more keys to [Route] and [Address] sections
1021 - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
1022 - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
1023 - add reduced [Link] support to .network files
1024 - add Scope= parsing option for [Network]
1025 - properly handle routerless dhcp leases
1026 - work with non-Ethernet devices
1027 - add support for more bond options
1028 - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from?
1029 - the DHCP lease data (such as NTP/DNS) is still made available when
1030 a carrier is lost on a link. It should be removed instantly.
1031 - expose in the API the following bits:
1032 - option 15, domain name and/or option 119, search list
1033 - option 12, host name and/or option 81, fqdn
1034 - option 123, 144, geolocation
1035 - option 252, configure http proxy (PAC/wpad)
1036 - provide a way to define a per-network interface default metric value
1037 for all routes to it. possibly a second default for DHCP routes.
1038 - allow Name= to be specified repeatedly in the [Match] section. Maybe also
1039 support Name=foo*|bar*|baz ?
1040 - duplicate address check for static IPs (like ARPCHECK in network-scripts)
1041 - allow DUID/IAID to be customized, see issue #394.
1042 - whenever uplink info changes, make DHCP server send out FORCERENEW
1044 * networkd-wait-online:
1045 - make operstates to wait for configurable?
1048 - figure out how much we can increase Maximum Message Size
1051 - add functions to set previously stored IPv6 addresses on startup and get
1052 them at shutdown; store them in client->ia_na
1053 - write more test cases
1054 - implement reconfigure support, see 5.3., 15.11. and 22.20.
1055 - implement support for temporary adressess (IA_TA)
1056 - implement dhcpv6 authentication
1057 - investigate the usefulness of Confirm messages; i.e. are there any
1058 situations where the link changes without any loss in carrier detection
1060 - some servers don't do rapid commit without a filled in IA_NA, verify
1067 - natively watch for dbus-*.service symlinks (PENDING)
1068 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
1070 * fix alsa mixer restore to not print error when no config is stored
1072 * make cryptsetup lower --iter-time
1074 * patch kernel for xattr support in /dev, /proc/, /sys?
1076 * kernel: add device_type = "fb", "fbcon" to class "graphics"
1078 * /usr/bin/service should actually show the new command line
1080 * 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=
1082 * fedora: F20: go timer units all the way, leave cron.daily for cron
1084 * neither pkexec nor sudo initialize environ[] from the PAM environment?
1086 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
1088 * register catalog database signature as file magic
1090 * zsh shell completion:
1091 - <command> <verb> -<TAB> should complete options, but currently does not
1092 - systemctl add-wants,add-requires
1097 * look for close() vs. close_nointr() vs. close_nointr_nofail()
1099 * check for strerror(r) instead of strerror(-r)
1103 * set_put(), hashmap_put() return values check. i.e. == 0 does not free()!
1105 * use secure_getenv() instead of getenv() where appropriate
1107 * link up selected blog stories from man pages and unit files Documentation= fields
1108 String is not UTF-8 clean, ignoring assignment
1109 timedatex.service: Consumed 26ms CPU time.