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.
11 * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
13 * wiki: update journal format documentation for lz4 additions
17 * Rearrange tests so that the various test-xyz.c match a specific src/basic/xyz.c again
19 * rework mount.c and swap.c to follow proper state enumeration/deserialization
20 semantics, like we do for device.c now
24 * bootctl: implement Type #2 boot loader entry discovery
26 * bootctl,sd-boot: actually honour the "architecture" key
28 * consider splitting out all temporary file creation APIs (we have so many in
29 fileio.h and elsewhere!) into a new util file of its own.
31 * set memory.oom.group in cgroupsv2 for all leaf cgroups
33 * drop umask() calls and suchlike from our generators, pid1 should set things up correctly anyway
35 * paranoia: whenever we process passwords, call mlock() on the memory
36 first. i.e. look for all places we use string_erase()/string_free_erase() and
37 augment them with mlock()
39 * whenever oom_kill memory.event event is triggered print a nice log message
41 * Move RestrictAddressFamily= to the new cgroup create socket
43 * support the bind/connect/sendmsg cgroup stuff for sandboxing, and possibly
46 * chown() tty a service is attached to after the service goes down
48 * optionally: turn on cgroup delegation for per-session scope units
50 * introduce per-unit (i.e. per-slice, per-service) journal log size limits.
52 * optionally, if a per-partition GPT flag is set for the root/home/… partitions
53 format the partition on next boot and unset the flag, in order to implement
54 factory reset. also, add a second flag that simply indicates whether such a
55 scheme is supported. then, add a tool (or maybe beef up systemd-dissect) to
56 show state of these flags, and optionally trigger such a factory reset on
57 next boot by setting the flag.
59 * sd-boot: search drop-ins in $BOOT, too
61 * sd-boot: add "oneshot boot timeout" variable support
63 * sd-boot: automatically load EFI modules from some drop-in dir, so that people
64 can add in file system drivers and such
66 * esp generator: also mount $BOOT if found
68 * sd-boot: optionally, show boot menu when previous default boot item has
69 non-zero "tries done" count
71 * logind: add "boot into bootmenu" API, and possibly even "boot into windows"
72 and "boot into macos".
74 * bootspec.c: also enumerate EFI unified kernel images.
76 * maybe extend .path units to expose fanotify() per-mount change events
78 * Add a "systemctl list-units --by-slice" mode or so, which rearranges the
79 output of "systemctl list-units" slightly by showing the tree structure of
80 the slices, and the units attached to them.
82 * the a-posteriori stopping of units bound to units that disappeared logic
83 should be reworked: there should be a queue of units, and we should only
84 enqeue stop jobs from a defer event that processes queue instead of
85 right-away when we find a unit that is bound to one that doesn't exist
86 anymore. (similar to how the stop-unneeded queue has been reworked the same
89 * nspawn: make nspawn suitable for shell pipelines: instead of triggering a
90 hangup when input is finished, send ^D, which synthesizes an EOF. Then wait
91 for hangup or ^D before passing on the EOF.
93 * When reloading configuration PID 1 should reset all its properties to the
94 original defaults before calling parse_config()
96 * Add OnTimezoneChange= and OnTimeChange= stanzas to .timer units in order to
97 schedule events based on time and timezone changes.
99 * nspawn: greater control over selinux label?
101 * cgroups: figure out if we can somehow communicate in a cleaner way whether a
102 systemd instance not running in the cgroup root shall or shall not manage the
103 attributes of its top-level cgroup. Currently it assumes it manages all, but
104 then might get EPERM due to permission porblems/userns, which is OK, but this
105 should be revisited to make clearer and also work if the payload systemd runs
106 with full privs and without userns.
108 * hibernate/s2h: make this robust and safe to enable in Fedora by default.
111 1. add resume_offset support to the resume code (i.e. support swap files
113 2. check if swap is on weird storage and refuse if so
114 3. add autodetection of hibernation images
116 * cgroups: use inotify to get notified when somebody else modifies cgroups
117 owned by us, then log a friendly warning.
119 * beef up log.c with support for stripping ANSI sequences from strings, so that
120 it is OK to include them in log strings. This would be particularly useful so
121 that our log messages could contain clickable links for example for unit
122 files and suchlike we operate on.
124 * add support for "portablectl attach http://foobar.com/waaa.raw (i.e. importd integration)
126 * add attach --enable and attach --now (for attach+enable+start)
128 * sync dynamic uids/gids between host+portable srvice (i.e. if DynamicUser=1 is set for a service, make sure that the
129 selected user is resolvable in the service even if it ships its own /etc/passwd)
131 * Fix DECIMAL_STR_MAX or DECIMAL_STR_WIDTH. One includes a trailing NUL, the
132 other doesn't. What a desaster. Probably to exclude it. Also
133 DECIMAL_STR_WIDTH should probably add an extra "-" into account for negative
136 * Check that users of inotify's IN_DELETE_SELF flag are using it properly, as
137 usually IN_ATTRIB is the right way to watch deleted files, as the former only
138 fires when a file is actually removed from disk, i.e. the link count drops to
139 zero and is not open anymore, while the latter happens when a file is
140 unlinked from any dir.
142 * port systemctl, busctl, … over to format-table.[ch]'s table formatters
144 * pid1: lock image configured with RootDirectory=/RootImage= using the usual nspawn semantics while the unit is up
146 * add --vacuum-xyz options to coredumpctl, matching those journalctl already has.
148 * SuccessExitStatus= and friends should probably also accept symbolic exit
149 codes names, i.e. error codes from the list maintained in exit-codes.[ch]
151 * introduce Ephemeral= unit file switch, that creates an ephemeral copy of all
152 files and directories that are left writable for a unit, and which are
153 removed after the unit goes down again. A bit like --ephemeral for
154 systemd-nspawn but for system services. If used together with RootImage= this
155 should reflink the image file itself.
157 Related: add Ephemeral=<path1> <path2> … which would allow marking
158 specific paths only like this.
160 * add CopyFile= or so as unit file setting that may be used to copy files or
161 directory trees from the host to te services RootImage= and RootDirectory=
162 environment. Which we can use for /etc/machine-id and in particular
163 /etc/resolv.conf. Should be smart and do something useful on read-only
164 images, for example fallback to read-only bind mounting the file instead.
166 * nspawn's console TTY should be allocated from within the container, not
167 mounted in from the outside
169 * show invocation ID in systemd-run output
171 * bypass SIGTERM state in unit files if KillSignal is SIGKILL
173 * tree-wide: ensure we always block the signals we hook into with
174 sd_event_add_signal() first
176 * add proper dbus APIs for the various sd_notify() commands, such as MAINPID=1
177 and so on, which would mean we could report errors and such.
179 * teach tmpfiles.d q/Q logic something sensible in the context of XFS/ext4
182 * introduce DefaultSlice= or so in system.conf that allows changing where we
183 place our units by default, i.e. change system.slice to something
184 else. Similar, ManagerSlice= should exist so that PID1's own scope unit could
185 be moved somewhere else too. Finally machined and logind should get similar
186 options so that it is possible to move user session scopes and machines to a
187 different slice too by default. Usecase: people who want to put resources on
188 the entire system, with the exception of one specific service. See:
189 https://lists.freedesktop.org/archives/systemd-devel/2018-February/040369.html
191 * maybe rework get_user_creds() to query the user database if $SHELL is used
192 for root, but only then.
194 * be stricter with fds we receive for the fdstore: close them asynchronously
196 * calenderspec: add support for week numbers and day numbers within a
197 year. This would allow us to define "bi-weekly" triggers safely.
199 * sd-bus: add vtable flag, that may be used to request client creds implicitly
200 and asynchronously before dispatching the operation
202 * make use of ethtool veth peer info in machined, for automatically finding out
203 host-side interface pointing to the container.
205 * add some special mode to LogsDirectory=/StateDirectory=… that allows
206 declaring these directories without necessarily pulling in deps for them, or
207 creating them when starting up. That way, we could declare that
208 systemd-journald writes to /var/log/journal, which could be useful when we
209 doing disk usage calculations and so on.
211 * taint systemd if there are fewer than 65536 users assigned (userns) to the system.
213 * deprecate PermissionsStartOnly= and RootDirectoryStartOnly= in favour of the ExecStart= prefix chars
215 * add a new RuntimeDirectoryPreserve= mode that defines a similar lifecycle for
216 the runtime dir as we maintain for the fdstore: i.e. keep it around as long
217 as the unit is running or has a job queued.
219 * support projid-based quota in machinectl for containers, and then drop
220 implicit btrfs loopback magic in machined
222 * Add NetworkNamespacePath= to specify a path to a network namespace
224 * maybe use SOURCE_DATE_EPOCH (i.e. the env var the reproducible builds folks
225 introduced) as the RTC epoch, instead of the mtime of NEWS.
227 * add a way to lock down cgroup migration: a boolean, which when set for a unit
228 makes sure the processes in it can never migrate out of it
230 * blog about fd store and restartable services
232 * document Environment=SYSTEMD_LOG_LEVEL=debug drop-in in debugging document
234 * rework ExecOutput and ExecInput enums so that EXEC_OUTPUT_NULL loses its
235 magic meaning and is no longer upgraded to something else if set explicitly.
237 * in the long run: permit a system with /etc/machine-id linked to /dev/null, to
238 make it lose its identity, i.e. be anonymous. For this we'd have to patch
239 through the whole tree to make all code deal with the case where no machine
242 * optionally, collect cgroup resource data, and store it in per-unit RRD files,
243 suitable for processing with rrdtool. Add bus API to access this data, and
244 possibly implement a CPULoad property based on it.
246 * beef up pam_systemd to take unit file settings such as cgroups properties as
249 * a new "systemd-analyze security" tool outputting a checklist of security
250 features a service does and does not implement
252 * maybe hook of xfs/ext4 quotactl() with services? i.e. automatically manage
253 the quota of a the user indicated in User= via unit file settings, like the
254 other resource management concepts. Would mix nicely with DynamicUser=1. Or
255 alternatively, do this with projids, so that we can also cover services
256 running as root. Quota should probably cover all the special dirs such as
257 StateDirectory=, LogsDirectory=, CacheDirectory=, as well as RootDirectory= if it
258 is set, plus the whole disk space any image configured with RootImage=.
260 * Introduce "exit" as an EmergencyAction value, and allow to configure a
261 per-unit success/failure exit code to configure. This would be useful for
262 running commands inside of services inside of containers, which could then
263 propagate their failure state all the way up.
265 * In DynamicUser= mode: before selecting a UID, use disk quota APIs on relevant
266 disks to see if the UID is already in use.
268 * add "systemctl wait" or so, which does what "systemd-run --wait" does, but
269 for all units. It should be both a way to pin units into memory as well as a
270 wait to retrieve their exit data.
272 * expose IO accounting data on the bus, show it in systemd-run --wait and log
273 about it in the resource log message
275 * add "systemctl purge" for flushing out configuration, state, logs, ... of a
276 unit when it is stopped
278 * show whether a service has out-of-date configuration in "systemctl status" by
279 using mtime data of ConfigurationDirectory=.
281 * replace all uses of fgets() + LINE_MAX by read_line()
283 * Add AddUser= setting to unit files, similar to DynamicUser=1 which however
284 creates a static, persistent user rather than a dynamic, transient user. We
285 can leverage code from sysusers.d for this.
287 * add some optional flag to ReadWritePaths= and friends, that has the effect
288 that we create the dir in question when the service is started. Example:
290 ReadWritePaths=:/var/lib/foobar
292 * maybe add call sd_journal_set_block_timeout() or so to set SO_SNDTIMEO for
293 the sd-journal logging socket, and, if the timeout is set to 0, sets
294 O_NONBLOCK on it. That way people can control if and when to block for
297 * hostnamed: populate form factor data from a new hwdb database, so that old
298 yogas can be recognized as "convertible" too, even if they predate the DMI
299 "convertible" form factor
301 * Maybe add a small tool invoked early at boot, that adds in or resizes
302 partitions automatically, to be used when the media used is actually larger
303 than the image written onto it is.
305 * Maybe add PrivatePIDs= as new unit setting, and do minimal PID namespacing
306 after all. Be strict however, only support the equivalent of nspawn's
307 --as-pid2 switch, and sanely proxy sd_notify() messages dropping stuff such
310 * Add ExecMonitor= setting. May be used multiple times. Forks off a process in
311 the service cgroup, which is supposed to monitor the service, and when it
312 exits the service is considered failed by its monitor.
314 * track the per-service PAM process properly (i.e. as an additional control
315 process), so that it may be queried on the bus and everything.
317 * add a new "debug" job mode, that is propagated to unit_start() and for
318 services results in two things: we raise SIGSTOP right before invoking
319 execve() and turn off watchdog support. Then, use that to implement
320 "systemd-gdb" for attaching to the start-up of any system service in its
323 * maybe introduce gpt auto discovery for /var/tmp?
325 * maybe add gpt-partition-based user management: each user gets his own
326 LUKS-encrypted GPT partition with a new GPT type. A small nss module
327 enumerates users via udev partition enumeration. UIDs are assigned in a fixed
328 way: the partition index is added as offset to some fixed base uid. User name
329 is stored in GPT partition name. A PAM module authenticates the user via the
330 LUKS partition password. Benefits: strong per-user security, compatibility
331 with stateless/read-only/verity-enabled root. (other idea: do this based on
332 loopback files in /home, without GPT involvement)
334 * gpt-auto logic: introduce support for discovering /var matching an image. For
335 that, use a partition type UUID that is hashed from the OS name (as encoded
336 in /etc/os-release), the architecture, and 4 new bits from the gpt flags
337 field of the root partition. This way can easily support multiple OS
338 installations on the same GPT partition table, without problems with
339 unmatched /var partitions.
341 * gpt-auto logic: related to the above, maybe support a "secondary" root
342 partition, that is mounted to / and is writable, and where the actual root's
343 /usr is mounted into.
345 * 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
347 * drop nss-myhostname in favour of nss-resolve?
349 * add a percentage syntax for TimeoutStopSec=, e.g. TimeoutStopSec=150%, and
350 then use that for the setting used in user@.service. It should be understood
351 relative to the configured default value.
353 * in networkd, when matching device types, fix up DEVTYPE rubbish the kernel passes to us
355 * enable LockMLOCK to take a percentage value relative to physical memory
357 * Permit masking specific netlink APIs with RestrictAddressFamily=
359 * nspawn: support that /proc, /sys/, /dev are pre-mounted
361 * define gpt header bits to select volatility mode
363 * ProtectKernelLogs= (drops CAP_SYSLOG, add seccomp for syslog() syscall, and DeviceAllow to /dev/kmsg) in service files
365 * ProtectClock= (drops CAP_SYS_TIMES, adds seecomp filters for settimeofday, adjtimex), sets DeviceAllow o /dev/rtc
367 * ProtectTracing= (drops CAP_SYS_PTRACE, blocks ptrace syscall, makes /sys/kernel/tracing go away)
369 * ProtectMount= (drop mount/umount/pivot_root from seccomp, disallow fuse via DeviceAllow, imply Mountflags=slave)
371 * ProtectKeyRing= to take keyring calls away
373 * RemoveKeyRing= to remove all keyring entries of the specified user
375 * ProtectReboot= that masks reboot() and kexec_load() syscalls, prohibits kill
376 on PID 1 with the relevant signals, and makes relevant files in /sys and
377 /proc (such as the sysrq stuff) unavailable
379 * DeviceAllow= should also generate seccomp filters for mknod()
381 * make sure the ratelimit object can deal with USEC_INFINITY as way to turn off things
383 * journalctl: make sure -f ends when the container indicated by -M terminates
385 * mount: automatically search for "main" partition of an image has multiple
388 * expose the "privileged" flag of ExecCommand on the bus, and open it up to
391 * in nss-systemd, if we run inside of RootDirectory= with PrivateUsers= set,
392 find a way to map the User=/Group= of the service to the right name. This way
393 a user/group for a service only has to exist on the host for the right
396 * add bus API for creating unit files in /etc, reusing the code for transient units
398 * add bus API to remove unit files from /etc
400 * add bus API to retrieve current unit file contents (i.e. implement "systemctl cat" on the bus only)
402 * rework fopen_temporary() to make use of open_tmpfile_linkable() (problem: the
403 kernel doesn't support linkat() that replaces existing files, currently)
405 * transient units: don't bother with actually setting unit properties, we
406 reload the unit file anyway
408 * journald: sigbus API via a signal-handler safe function that people may call
409 from the SIGBUS handler
411 * optionally, also require WATCHDOG=1 notifications during service start-up and shutdown
413 * resolved: when routing queries, make sure only look for the *longest* suffix...
415 * delay activation of logind until somebody logs in, or when /dev/tty0 pulls it
416 in or lingering is on (so that containers don't bother with it until PAM is used). also exit-on-idle
418 * cache sd_event_now() result from before the first iteration...
420 * add systemctl stop --job-mode=triggering that follows TRIGGERED_BY deps and adds them to the same transaction
422 * PID1: find a way how we can reload unit file configuration for
423 specific units only, without reloading the whole of systemd
425 * add an explicit parser for LimitRTPRIO= that verifies
426 the specified range and generates sane error messages for incorrect
429 * when we detect that there are waiting jobs but no running jobs, do something
431 * push CPUAffinity= also into the "cpuset" cgroup controller (only after the cpuset controller got ported to the unified hierarchy)
433 * PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn)
435 * there's probably something wrong with having user mounts below /sys,
436 as we have for debugfs. for exmaple, src/core/mount.c handles mounts
437 prefixed with /sys generally special.
438 http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html
440 * fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline
442 * docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date
444 * add a job mode that will fail if a transaction would mean stopping
445 running units. Use this in timedated to manage the NTP service
447 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html
449 * Maybe add support for the equivalent of "ethtool advertise" to .link files?
450 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030112.html
452 * The udev blkid built-in should expose a property that reflects
453 whether media was sensed in USB CF/SD card readers. This should then
454 be used to control SYSTEMD_READY=1/0 so that USB card readers aren't
455 picked up by systemd unless they contain a medium. This would mirror
456 the behaviour we already have for CD drives.
458 * networkd/udev: implement SR_IOV configuration in .link files:
459 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html
461 * hostnamectl: show root image uuid
463 * Find a solution for SMACK capabilities stuff:
464 http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html
466 * "systemctl preset-all" should probably order the unit files it
467 operates on lexicographically before starting to work, in order to
468 ensure deterministic behaviour if two unit files conflict (like DMs
471 * synchronize console access with BSD locks:
472 http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html
474 * as soon as we have sender timestamps, revisit coalescing multiple parallel daemon reloads:
475 http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
477 * in systemctl list-unit-files: show the install value the presets would suggest for a service in a third column
479 * figure out when we can use the coarse timers
481 * add "systemctl start -v foobar.service" that shows logs of a service
482 while the start command runs. This is non-trivial to do without
483 races though, since we should flush out all journal messages before
484 returning from the "systemctl stop".
486 * 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
488 * maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment=
490 * consider showing the unit names during boot up in the status output, not just the unit descriptions
492 * maybe allow timer units with an empty Units= setting, so that they
493 can be used for resuming the system but nothing else.
495 * what to do about udev db binary stability for apps? (raw access is not an option)
497 * man: maybe use the word "inspect" rather than "introspect"?
499 * systemctl: if some operation fails, show log output?
501 * systemctl edit: use equvalent of cat() to insert existing config as a comment, prepended with #.
502 Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc.
504 * exponential backoff in timesyncd when we cannot reach a server
506 * timesyncd: add ugly bus calls to set NTP servers per-interface, for usage by NM
508 * merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share....
510 * systemd.show_status= should probably have a mode where only failed
513 * add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL
514 (throughout the codebase, not only PID1)
518 - service registration
519 - service/domain/types browsing
521 - DNS-SD service registration from socket units
522 - resolved should optionally register additional per-interface LLMNR
523 names, so that for the container case we can establish the same name
524 (maybe "host") for referencing the server, everywhere.
525 - allow clients to request DNSSEC for a single lookup even if DNSSEC is off (?)
526 - hook up resolved with machined-based address resolution
528 * refcounting in sd-resolve is borked
530 * Add a new verb "systemctl top"
532 * add new gpt type for btrfs volumes
534 * support empty /etc boots nicely:
535 - nspawn/gpt-generator: introduce new gpt partition type for /usr
537 * generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
539 * a way for container managers to turn off getty starting via $container_headless= or so...
541 * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
543 * For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
544 they run added to the initial transaction and thus confuse Type=idle.
546 * add bus api to query unit file's X fields.
548 * gpt-auto-generator:
549 - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap?
550 - Make /home automount rather than mount?
552 * add generator that pulls in systemd-network from containers when
553 CAP_NET_ADMIN is set, more than the loopback device is defined, even
554 when it is otherwise off
556 * MessageQueueMessageSize= (and suchlike) should use parse_iec_size().
558 * implement Distribute= in socket units to allow running multiple
559 service instances processing the listening socket, and open this up
562 * socket units: support creating sockets in different namespace,
563 opening it up for JoinsNamespaceOf=. This would require to fork off
564 a tiny process that joins the namespace and creates/binds the socket
565 and passes this back to PID1 via SCM_RIGHTS. This also could be used
566 to allow Chown/chgrp on sockets without requiring NSS in PID 1.
568 * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze
569 $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls
570 should SIGSTOP all unit processes in a loop until all processes of
571 it are fully stopped. This can later be used for app management by
572 desktop UIs such as gnome-shell to freeze apps that are not visible
573 on screen, not unlike how job control works on the shell
576 - implement per-slice CPUFairScheduling=1 switch
577 - handle jointly mounted controllers correctly
578 - introduce high-level settings for RT budget, swappiness
579 - how to reset dynamically changed unit cgroup attributes sanely?
580 - when reloading configuration, apply new cgroup configuration
581 - when recursively showing the cgroup hierarchy, optionally also show
582 the hierarchies of child processes
585 - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
587 * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk?
589 * when we detect low battery and no AC on boot, show pretty splash and refuse boot
591 * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
593 * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
595 * 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
597 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
599 * If we try to find a unit via a dangling symlink, generate a clean
600 error. Currently, we just ignore it and read the unit from the search
603 * refuse boot if /usr/lib/os-release is missing or /etc/machine-id cannot be set up
605 * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
607 * load .d/*.conf dropins for device units
609 * allow implementation of InaccessibleDirectories=/ plus
610 ReadOnlyDirectories=... for whitelisting files for a service.
614 - GetAllProperties() on a non-existing object does not result in a failure currently
615 - port to sd-resolve for connecting to TCP dbus servers
616 - 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
617 - see if we can drop more message validation on the sending side
618 - add API to clone sd_bus_message objects
619 - longer term: priority inheritance
621 - NameLost/NameAcquired obsolete
624 - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
627 - allow multiple signal handlers per signal?
628 - document chaining of signal handler for SIGCHLD and child handlers
629 - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ...
630 - generate a failure of a default event loop is executed out-of-thread
632 * investigate endianness issues of UUID vs. GUID
634 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
635 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
637 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
639 * add a pam module that on password changes updates any LUKS slot where the password matches
641 * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases...
644 - add unit tests for config_parse_device_allow()
646 * seems that when we follow symlinks to units we prefer the symlink
647 destination path over /etc and /usr. We should not do that. Instead
648 /etc should always override /run+/usr and also any symlink
651 * when isolating, try to figure out a way how we implicitly can order
652 all units we stop before the isolating unit...
654 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
656 * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify
657 that the boot loader should be installed to the ESP. Define a way
658 how an installer can figure out whether a BLS compliant boot loader
661 * think about requeuing jobs when daemon-reload is issued? usecase:
662 the initrd issues a reload after fstab from the host is accessible
663 and we might want to requeue the mounts local-fs acquired through
666 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
668 * remove any syslog support from log.c — we probably cannot do this before split-off udev is gone for good
670 * shutdown logging: store to EFI var, and store to USB stick?
672 * merge unit_kill_common() and unit_kill_context()
674 * introduce ExecCondition= in services
677 - honor language efi variables for default language selection (if there are any?)
678 - honor timezone efi variables for default timezone selection (if there are any?)
679 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
681 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
683 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
685 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
688 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
689 - logind: wakelock/opportunistic suspend support
690 - Add pretty name for seats in logind
691 - logind: allow showing logout dialog from system?
692 - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
693 - if pam_systemd is invoked by su from a process that is outside of a
694 any session we should probably just become a NOP, since that's
695 usually not a real user session but just some system code that just
697 - logind: make the Suspend()/Hibernate() bus calls wait for the for
698 the job to be completed. before returning, so that clients can wait
699 for "systemctl suspend" to finish to know when the suspending is
701 - logind: when the power button is pressed short, just popup a
702 logout dialog. If it is pressed for 1s, do the usual
703 shutdown. Inspiration are Macs here.
704 - expose "Locked" property on logind sesison objects
705 - maybe allow configuration of the StopTimeout for session scopes
706 - rename session scope so that it includes the UID. THat way
707 the session scope can be arranged freely in slices and we don't have
708 make assumptions about their slice anymore.
709 - follow PropertiesChanged state more closely, to deal with quick logouts and
712 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
715 - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields
716 - import and delete pstore filesystem content at startup
717 - journald: also get thread ID from client, plus thread name
718 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
719 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
720 - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively?
721 - declare the local journal protocol stable in the wiki interface chart
722 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
723 - journald: when dropping msgs due to ratelimit make sure to write
724 "dropped %u messages" not only when we are about to print the next
725 message that works, but alraedy after a short tiemout
726 - check if we can make journalctl by default use --follow mode inside of less if called without args?
727 - maybe add API to send pairs of iovecs via sd_journal_send
728 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
729 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
730 and !FOOBAR for events without FOOBAR.
731 - journal: store timestamp of journal_file_set_offline() int he header,
732 so it is possible to display when the file was last synced.
733 - 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.
734 - journal: find a way to allow dropping history early, based on priority, other rules
735 - journal: When used on NFS, check payload hashes
736 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
737 - refuse taking lower-case variable names in sd_journal_send() and friends.
738 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
739 - journal: deal nicely with byte-by-byte copied files, especially regards header
740 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
741 - Replace utmp, wtmp, btmp, and lastlog completely with journal
742 - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
743 - when a kernel driver logs in a tight loop, we should ratelimit that too.
744 - journald: optionally, log debug messages to /run but everything else to /var
745 - journald: when we drop syslog messages because the syslog socket is
746 full, make sure to write how many messages are lost as first thing
747 to syslog when it works again.
748 - change systemd-journal-flush into a service that stays around during
749 boot, and causes the journal to be moved back to /run on shutdown,
750 so that we do not keep /var busy. This needs to happen synchronously,
751 hence doing this via signals is not going to work.
752 - optionally support running journald from the command line for testing purposes in external projects
753 - journald: allow per-priority and per-service retention times when rotating/vacuuming
754 - journald: make use of uid-range.h to managed uid ranges to split
756 - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
757 - improve journalctl performance by loading journal files
758 lazily. Encode just enough information in the file name, so that we
759 do not have to open it to know that it is not interesting for us, for
760 the most common operations.
761 - man: document that corrupted journal files is nothing to act on
762 - rework journald sigbus stuff to use mutex
763 - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our
764 services that run under their own user ids, and use User= (but only
765 in a world where userns is ubiquitous since otherwise we cannot
766 invoke those daemons on the host AND in a container anymore). Also,
767 if LimitNPROC= is used without User= we should warn and refuse
769 - journalctl --verify: don't show files that are currently being
770 written to as FAIL, but instead show that their are being written to.
771 - add journalctl -H that talks via ssh to a remote peer and passes through
773 - add a version of --merge which also merges /var/log/journal/remote
774 - journalctl: -m should access container journals directly by enumerating
775 them via machined, and also watch containers coming and going.
776 Benefit: nspawn --ephemeral would start working nicely with the journal.
777 - assign MESSAGE_ID to log messages about failed services
779 * add a test if all entries in the catalog are properly formatted.
780 (Adding dashes in a catalog entry currently results in the catalog entry
781 being silently skipped. journalctl --update-catalog must warn about this,
782 and we should also have a unit test to check that all our message are OK.)
785 - document that deps in [Unit] sections ignore Alias= fields in
786 [Install] units of other units, unless those units are disabled
787 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
788 - document that service reload may be implemented as service reexec
789 - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
790 - document systemd-journal-flush.service properly
791 - documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
792 - man: document the very specific env the shutdown drop-in tools live in
793 - man: add more examples to man pages
794 - man: maybe sort directives in man pages, and take sections from --help and apply them to man too
797 - add systemctl switch to dump transaction without executing it
798 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
799 - "systemctl disable" on a static unit prints no message and does
800 nothing. "systemctl enable" does nothing, and gives a bad message
801 about it. Should fix both to print nice actionable messages.
802 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
803 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
804 - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
805 - systemctl: "Journal has been rotated since unit was started." message is misleading
806 - systemctl status output should include list of triggering units and their status
809 - "systemctl mask" should find all names by which a unit is accessible
810 (i.e. by scanning for symlinks to it) and link them all to /dev/null
813 - timer units should get the ability to trigger when:
814 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
817 - Modulate timer frequency based on battery state
819 * add libsystemd-password or so to query passwords during boot using the password agent logic
821 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
823 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
825 * make repeated alt-ctrl-del presses printing a dump
827 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
829 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
832 - emulate /dev/kmsg using CUSE and turn off the syslog syscall
833 with seccomp. That should provide us with a useful log buffer that
834 systemd can log to during early boot, and disconnect container logs
835 from the kernel's logs.
836 - as soon as networkd has a bus interface, hook up --network-interface=,
837 --network-bridge= with networkd, to trigger netdev creation should an
839 - a nice way to boot up without machine id set, so that it is set at boot
840 automatically for supporting --ephemeral. Maybe hash the host machine id
841 together with the machine name to generate the machine id for the container
842 - fix logic always print a final newline on output.
843 https://github.com/systemd/systemd/pull/272#issuecomment-113153176
844 - should optionally support receiving WATCHDOG=1 messages from its payload
846 - optionally automatically add FORWARD rules to iptables whenever nspawn is
847 running, remove them when shut down.
848 - maybe make copying of /etc/resolv.conf optional, and skip it if --read-only
852 - refuse mounting over a mount point
853 - automatically discover .roothash files in dissect, similarly to nspawn
856 - add an API so that libvirt-lxc can inform us about network interfaces being
857 removed or added to an existing machine
858 - "machinectl migrate" or similar to copy a container from or to a
859 difference host, via ssh
860 - introduce systemd-nspawn-ephemeral@.service, and hook it into
861 "machinectl start" with a new --ephemeral switch
862 - "machinectl status" should also show internal logs of the container in
864 - "machinectl history"
866 - "machinectl commit" that takes a writable snapshot of a tree, invokes a
867 shell in it, and marks it read-only after use
870 - generate a nice warning if mkfs.btrfs is missing
873 - cryptsetup-generator: allow specification of passwords in crypttab itself
874 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
876 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
877 https://bugs.freedesktop.org/show_bug.cgi?id=54712
879 * add a dependency on standard-conf.xml and other included files to man pages
881 * MountFlags=shared acts as MountFlags=slave right now.
883 * properly handle loop back mounts via fstab, especially regards to fsck/passno
885 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
890 - add trigger --subsystem-match=usb/usb_device device
891 - reimport udev db after MOVE events for devices without dev_t
893 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
895 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
898 - save coredump in Windows/Mozilla minidump format
899 - when truncating coredumps, also log the full size that the process had, and make a metadata field so we can report truncated coredumps
901 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
903 * default to actual 32-bit PIDs, via /proc/sys/kernel/pid_max
905 * be able to specify a forced restart of service A where service B depends on, in case B
906 needs to be auto-respawned?
909 - apply "x" on "D" too (see patch from William Douglas)
911 - instead of ignoring unknown fields, reject them.
912 - creating new directories/subvolumes/fifos/device nodes
913 should not follow symlinks. None of the other adjustment or creation
914 calls follow symlinks.
916 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
918 * verify that the AF_UNIX sockets of a service in the fs still exist
919 when we start a service in order to avoid confusion when a user
920 assumes starting a service is enough to make it accessible
922 * Make it possible to set the keymap independently from the font on
923 the kernel cmdline. Right now setting one resets also the other.
925 * and a dbus call to generate target from current state
927 * write blog stories about:
928 - hwdb: what belongs into it, lsusb
929 - enabling dbus services
930 - how to make changes to sysctl and sysfs attributes
932 - how to pass throw-away units to systemd, or dynamically change properties of existing units
933 - testing with Harald's awesome test kit
935 - how to develop against journal browsing APIs
936 - the journal HTTP iface
937 - non-cgroup resource management
938 - dynamic resource management with cgroups
939 - refreshed, longer missions statement
940 - calendar time events
941 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
942 - how to create your own target
943 - instantiated apache, dovecot and so on
944 - hooking a script into various stages of shutdown/rearly booot
946 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
948 * dot output for --test showing the 'initial transaction'
951 - When logging about multiple units (stopping BoundTo units, conflicts, etc.),
952 log both units as UNIT=, so that journalctl -u triggers on both.
953 - generate better errors when people try to set transient properties
954 that are not supported...
955 http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html
956 - maybe introduce WantsMountsFor=? Usecase:
957 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html
958 - recreate systemd's D-Bus private socket file on SIGUSR2
959 - move PAM code into its own binary
960 - when we automatically restart a service, ensure we restart its rdeps, too.
961 - hide PAM options in fragment parser when compile time disabled
962 - Support --test based on current system state
963 - 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().
964 - after deserializing sockets in socket.c we should reapply sockopts and things
965 - drop PID 1 reloading, only do reexecing (difficult: Reload()
966 currently is properly synchronous, Reexec() is weird, because we
967 cannot delay the response properly until we are back, so instead of
968 being properly synchronous we just keep open the fd and close it
969 when done. That means clients do not get a successful method reply,
970 but much rather a disconnect on success.
971 - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
972 - when a bus name of a service disappears from the bus make sure to queue further activation requests
975 - allow port=0 in .socket units
976 - maybe introduce ExecRestartPre=
977 - add ReloadSignal= for configuring a reload signal to use
978 - implement Register= switch in .socket units to enable registration
979 in Avahi, RPC and other socket registration services.
980 - allow Type=simple with PIDFile=
981 https://bugzilla.redhat.com/show_bug.cgi?id=723942
982 - allow writing multiple conditions in unit files on one line
983 - load-fragment: when loading a unit file via a chain of symlinks
984 verify that it is not masked via any of the names traversed.
985 - introduce Type=pid-file
986 - introduce mix of BindTo and Requisite
987 - add a concept of RemainAfterExit= to scope units
988 - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
991 - Make sure ID_PATH is always exported and complete for
992 network devices where possible, so we can safely rely
994 - check MTUBytes parsing (expecting size_t but we are using unsigned)
997 - add support for more attribute types
998 - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places
1001 - add more keys to [Route] and [Address] sections
1002 - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
1003 - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
1004 - add reduced [Link] support to .network files
1005 - add Scope= parsing option for [Network]
1006 - properly handle routerless dhcp leases
1007 - work with non-Ethernet devices
1008 - add support for more bond options
1009 - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from?
1010 - the DHCP lease data (such as NTP/DNS) is still made available when
1011 a carrier is lost on a link. It should be removed instantly.
1012 - expose in the API the following bits:
1013 - option 15, domain name and/or option 119, search list
1014 - option 12, host name and/or option 81, fqdn
1015 - option 123, 144, geolocation
1016 - option 252, configure http proxy (PAC/wpad)
1017 - provide a way to define a per-network interface default metric value
1018 for all routes to it. possibly a second default for DHCP routes.
1019 - allow Name= to be specified repeatedly in the [Match] section. Maybe also
1020 support Name=foo*|bar*|baz ?
1021 - duplicate address check for static IPs (like ARPCHECK in network-scripts)
1022 - allow DUID/IAID to be customized, see issue #394.
1023 - whenever uplink info changes, make DHCP server send out FORCERENEW
1025 * networkd-wait-online:
1026 - make operstates to wait for configurable?
1029 - figure out how much we can increase Maximum Message Size
1032 - add functions to set previously stored IPv6 addresses on startup and get
1033 them at shutdown; store them in client->ia_na
1034 - write more test cases
1035 - implement reconfigure support, see 5.3., 15.11. and 22.20.
1036 - implement support for temporary adressess (IA_TA)
1037 - implement dhcpv6 authentication
1038 - investigate the usefulness of Confirm messages; i.e. are there any
1039 situations where the link changes without any loss in carrier detection
1041 - some servers don't do rapid commit without a filled in IA_NA, verify
1048 - natively watch for dbus-*.service symlinks (PENDING)
1049 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
1051 * fix alsa mixer restore to not print error when no config is stored
1053 * make cryptsetup lower --iter-time
1055 * patch kernel for xattr support in /dev, /proc/, /sys?
1057 * kernel: add device_type = "fb", "fbcon" to class "graphics"
1059 * /usr/bin/service should actually show the new command line
1061 * 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=
1063 * fedora: F20: go timer units all the way, leave cron.daily for cron
1065 * neither pkexec nor sudo initialize environ[] from the PAM environment?
1067 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
1069 * register catalog database signature as file magic
1071 * zsh shell completion:
1072 - <command> <verb> -<TAB> should complete options, but currently does not
1073 - systemctl add-wants,add-requires
1078 * look for close() vs. close_nointr() vs. close_nointr_nofail()
1080 * check for strerror(r) instead of strerror(-r)
1084 * set_put(), hashmap_put() return values check. i.e. == 0 does not free()!
1086 * use secure_getenv() instead of getenv() where appropriate
1088 * link up selected blog stories from man pages and unit files Documentation= fields