3 * Many manager configuration settings that are only applicable to user
4 manager or system manager can be always set. It would be better to reject
5 them when parsing config.
9 * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
11 * wiki: update journal format documentation for lz4 additions
15 * Rearrange tests so that the various test-xyz.c match a specific src/basic/xyz.c again
17 * rework mount.c and swap.c to follow proper state enumeration/deserialization
18 semantics, like we do for device.c now
22 * cryptsetup/homed: also support FIDO2 HMAC password logic for unlocking
23 devices. (see: https://github.com/mjec/fido2-hmac-secret)
25 * systemd-gpt-auto should probably set x-systemd.growfs on the mounts it
28 * homed/userdb: distuingish passwords and recovery keys in the records, since
29 we probably want to use different PBKDF algorithms/settings for them:
30 passwords have low entropy but recovery keys should have good entropy key
31 hence we can make them quicker to work.
34 - teach it to prepare an ESP wholesale, i.e. with mkfs.vfat invocation
35 - teach it to copy in unified kernel images and maybe type #1 boot loader spec entries from host
36 - make it operate on loopback files, dissecting enough to find ESP to operate on
38 * by default, in systemd --user service bump the OOMAdjust to 100, as privs
39 allow so that systemd survives
41 * honour specifiers in unit files that resolve to some very basic
42 /etc/os-release data, such as ID, VERSION_ID, BUILD_ID, VARIANT_ID.
44 * cryptsetup: allow encoding key directly in /etc/crypttab, maybe with a
45 "base64:" prefix. Useful in particular for pkcs11 mode.
47 * cryptsetup: reimplement the mkswap/mke2fs in cryptsetup-generator to use
48 systemd-makefs.service instead.
50 * socket units: allow creating a udev monitor socket with ListenDevices= or so,
51 with matches, then actviate app thorugh that passing socket oveer
54 - port sd_id128_get_machine_app_specific() over from khash
55 - port resolved over from libgcrypt (DNSSEC code)
56 - port journald + fsprg over from libgcrypt
57 - port importd over from libgcrypt
58 - when that's done: kill khash.c
59 - when that's done: kill gnutls support in resolved
61 * kill zenata, all hail weblate?
63 * when we resize disks (homed?) always round up to 4K sectors, not 512K
65 * add growvol and makevol options for /etc/crypttab, similar to
66 x-systemd.growfs and x-systemd-makefs.
68 * hook up the TPM to /etc/crypttab, with a new option that is similar to the
69 new PKCS#11 option in crypttab, and allows unlocking a LUKS volume via a key
70 unsealed from the TPM. Optionally, if TPM is not available fall back to
71 TPM-less mode, and set up linear DM mapping instead (inspired by kpartx), so
72 that the device paths stay the same, regardless if crypto is used or not.
74 * systemd-repart: by default generate minimized partition tables (i.e. tables
75 that only covere the space actually used, excluding any free space at the
76 end), in order to maximize dd'ability. Requires libfdisk work, see
77 https://github.com/karelzak/util-linux/issues/907
79 * systemd-repart: optionally, allow specifiying a path to initialize new
80 partitions from, i.e. an fs image file or a source device node. This would
81 then turn systemd-repart into a simple installer: with a few .repart files
82 you could replicate the host system on another device. a full installer would
83 then be: "systemd-repart /dev/sda && bootctl install /dev/sda &&
84 systemd-firstboot --image= …"
86 * systemd-repart: MBR partition table support. Care needs to be taken regarding
87 Type=, so that partition definitions can sanely apply to both the GPT and the
88 MBR case. Idea: accept syntax "Type=gpt:home mbr:0x83" for setting the types
89 for the two partition types explicitly. And provide an internal mapping so
90 that "Type=linux-generic" maps to the right types for both partition tables
93 * systemd-repart: allow sizing partitions as factor of available RAM, so that
94 we can reasonably size swap partitions for hibernation.
96 * systemd-repart: allow running mkfs before making partitions pop up +
97 encryption via LUKS to allow booting into an empty root with only /usr mounted in
99 * systemd-repart: allow managing the gpt read-only partition flag + auto-mount flag
101 * systemd-repart: allow disabling growing of specific partitions, or making
102 them (think ESP: we don't ever want to grow it, since we cannot resize vfat)
104 * systemd-repart: add specifier expansion, add especifier that refers to root
105 device node of current system, /usr device node, and matching verity, so that
106 an installer can be made a "copy" installer of the booted OS
108 * systemd-repart: make it a static checker during early boot for existence and
109 absence of other partitions for trusted boot environments
111 * systemd-repart: when no configuration is found, exit early do not check
112 partition table, so that it is safe to run in the initrd on any system
114 * systemd-repart: allow config of partition uuid
116 * userdb: allow username prefix searches in varlink API
118 * userdb: allow existence checks
120 * pid: activation by journal search expression
122 * when switching root from initrd to host, set the machine_id env var so that
123 if the host has no machine ID set yet we continue to use the random one the
126 * sd-event: add native support for P_ALL waitid() watching, then move PID 1 to
127 it fo reaping assigned but unknown children. This needs to some special care
128 to operate somewhat sensibly in light of priorities: P_ALL will return
129 arbitrary processes, regardless of the priority we want to watch them with,
130 hence on each event loop iteration check all processes which we shall watch
131 with higher prio explicitly, and then watch the entire rest with P_ALL.
133 * tweak sd-event's child watching: keep a prioq of children to watch and use
134 waitid() only on the children with the highest priority until one is waitable
135 and ignore all lower-prio ones from that point on
137 * maybe introduce xattrs that can be set on the root dir of the root fs
138 partition that declare the volatility mode to use the image in. Previously I
139 thought marking this via GPT partition flags but that's not ideal since
140 that's outside of the LUKS encryption/verity verification, and we probably
141 shouldn't operate in a volatile mode unless we got told so from a trusted
144 * figure out automatic partition discovery when combining writable root dir
147 * coredump: maybe when coredumping read a new xattr from /proc/$PID/exe that
148 may be used to mark a whole binary as non-coredumpable. Would fix:
149 https://bugs.freedesktop.org/show_bug.cgi?id=69447
151 * teach parse_timestamp() timezones like the calendar spec already knows it
153 * beef up hibernation to optionally do swapon/swapoff immediately before/after
156 * beef up s2h to implement a battery watch loop: instead of entering
157 hibernation unconditionally after coming back from resume make a decision
158 based on the battery load level: if battery level is above a specific
159 threshold, go to suspend again, only hibernate if below it. This means we'd
160 stick to suspend usually, but fall back to hibernation only when battery runs
161 empty (well, subject to our sampling interval). Related to this, check if we
162 can make ACPI _BTP (i.e. /sys/class/power_supply/*/alarm) work for us too,
163 i.e. see if it can wake up machines from suspend, so that we could resume
164 automatically when the system is low on power and move automatically to
165 hibernation mode. (see
166 https://uefi.org/sites/default/files/resources/ACPI%206_2_A_Sept29.pdf
168 https://docs.microsoft.com/en-us/windows-hardware/design/device-experiences/modern-standby-wake-sources
171 * add an explicit "vertical" mode to format-table, so that "systemctl
172 status"-like outputs (i.e. with a series of field names left and values
173 right) become genuine first class citizens, and we gain automatic, sane JSON
176 * systemd-firstboot: teach it dissector magic, so that you can point it to some
177 disk image and it will just set everything in it all behind the scenes.
179 * systemd-firstboot: add --force mode that replaces existing configuration.
181 * We should probably replace /var/log/README, /etc/rc.d/README with symlinks
182 that are linked to these places instead of copied. After all they are
183 constant vendor data.
185 * maybe add kernel cmdline params: 1) to force first-boot mode + 2) to force
186 random seed crediting
188 * nspawn: on cgroupsv1 issue cgroup empty handler process based on host events,
189 so that we make cgroup agent logic safe
191 * nspawn/machined: add API to invoke binary in container, then use that as
192 fallback in "machinectl shell"
194 * logind: rework pam_logind to also do a bus call in case of invocation from
195 user@.service, which returns the XDG_RUNTIME_DIR value, and make this
196 behaviour selectable via pam module option.
199 - when user tries to log into record signed by unrecognized key, automatically add key to our chain after polkit auth
200 - hook up machined/nspawn users with a varlink user query interface
201 - rollback when resize fails mid-operation
202 - GNOME's side for forget key on suspend (requires rework so that lock screen runs outside of uid)
205 - shrink fs on logout?
206 - update LUKS password on login if we find there's a password that unlocks the JSON record but not the LUKS device.
207 - create on activate?
208 - properties: icon url?, preferred session type?, administrator bool (which translates to 'wheel' membership)?, address?, telephone?, vcard?, samba stuff?, parental controls?
209 - communicate clearly when usb stick is safe to remove. probably involves
210 beefing up logind to make pam session close hook synchronous and wait until
211 systemd --user is shut down.
212 - logind: maybe keep a "busy fd" as long as there's a non-released session around or the user@.service
213 - maybe make automatic, read-only, time-based reflink-copies of LUKS disk images (think: time machine)
214 - distuingish destroy / remove (i.e. currently we can unregister a user, unregister+remove their home directory, but not just remove their home directory)
215 - in systemd's PAMName= logic: query passwords with ssh-askpassword, so that we can make "loginctl set-linger" mode work
216 - fingerprint authentication, pattern authentication, …
217 - make sure "classic" user records can also be managed by homed
218 - description field for groups
219 - make size of $XDG_RUNTIME_DIR configurable in user record
220 - reuse pwquality magic in firstboot
221 - query password from kernel keyring first
222 - update even if record is "absent"
223 - add a "access mode" + "fstype" field to the "status" section of json identity records reflecting the actually used access mode and fstype, even on non-luks backends
224 - move acct mgmt stuff from pam_systemd_home to pam_systemd?
225 - when "homectl --pkcs11-token-uri=" is used, synthesize ssh-authorized-keys records for all keys we have private keys on the stick for
226 - make slice for users configurable (requires logind rework)
227 - logind: populate auto-login list bus property from PKCS#11 token
228 - when determining state of a LUKS home directory, check DM suspended sysfs file
230 * introduce a new per-process uuid, similar to the boot id, the machine id, the
231 invocation id, that is derived from process creds, specifically a hashed
232 combination of AT_RANDOM + getpid() + the starttime from
233 /proc/self/status. Then add these ids implicitly when logging. Deriving this
234 uuid from these three things has the benefit that it can be derived easily
235 from /proc/$PID/ in a stable, and unique way that changes on both fork() and
238 * let's not GC a unit while its ratelimits are still pending
240 * when killing due to service watchdog timeout maybe detect whether target
241 process is under ptracing and then log loudly and continue instead.
243 * introduce a new group to own TPM devices
245 * make rfkill uaccess controllable by default, i.e. steal rule from
246 gnome-bluetooth and friends
248 * tweak journald context caching. In addition to caching per-process attributes
249 keyed by PID, cache per-cgroup attributes (i.e. the various xattrs we read)
250 keyed by cgroup path, and guarded by ctime changes. This should provide us
251 with a nice speed-up on services that have many processes running in the same
254 * make MAINPID= message reception checks even stricter: if service uses User=,
255 then check sending UID and ignore message if it doesn't match the user or
258 * maybe trigger a uevent "change" on a device if "systemctl reload xyz.device"
261 * when importing an fs tree with machined, optionally apply userns-rec-chown
263 * when importing an fs tree with machined, complain if image is not an OS
265 * Maybe introduce a helper safe_exec() or so, which is to execve() which
266 safe_fork() is to fork(). And then make revert the RLIMIT_NOFILE soft limit
267 to 1K implicitly, unless explicitly opted-out.
269 * rework seccomp/nnp logic that even if User= is used in combination with
270 a seccomp option we don't have to set NNP. For that, change uid first whil
271 keeping CAP_SYS_ADMIN, then apply seccomp, the drop cap.
273 * add a concept for automatically loading per-unit secrets off disk and
274 inserting them into the kernel keyring. Maybe SecretsDirectory= similar to
275 ConfigurationDirectory=.
277 * when no locale is configured, default to UEFI's PlatformLang variable
279 * bootctl,sd-boot: actually honour the "architecture" key
281 * add a new syscall group "@esoteric" for more esoteric stuff such as bpf() and
282 usefaultd() and make systemd-analyze check for it.
284 * paranoia: whenever we process passwords, call mlock() on the memory
285 first. i.e. look for all places we use free_and_erasep() and
286 augment them with mlock(). Also use MADV_DONTDUMP.
288 * Move RestrictAddressFamily= to the new cgroup create socket
290 * support the bind/connect/sendmsg cgroup stuff for sandboxing, and possibly
293 * maybe implicitly attach monotonic+realtime timestamps to outgoing messages in
294 log.c and sd-journal-send
296 * optionally: turn on cgroup delegation for per-session scope units
298 * introduce per-unit (i.e. per-slice, per-service) journal log size limits.
300 * sd-boot: automatically load EFI modules from some drop-in dir, so that people
301 can add in file system drivers and such
303 * sd-boot: optionally, show boot menu when previous default boot item has
304 non-zero "tries done" count
306 * introduce an option (or replacement) for "systemctl show" that outputs all
307 properties as JSON, similar to busctl's new JSON output. In contrast to that
308 it should skip the variant type string though.
310 * augment CODE_FILE=, CODE_LINE= with something like CODE_BASE= or so which
311 contains some identifier for the project, which allows us to include
312 clickable links to source files generating these log messages. The identifier
313 could be some abberviated URL prefix or so (taking inspiration from Go
314 imports). For example, for systemd we could use
315 CODE_BASE=github.com/systemd/systemd/blob/98b0b1123cc or so which is
316 sufficient to build a link by prefixing "http://" and suffixing the
319 * Augment MESSAGE_ID with MESSAGE_BASE, in a similar fashion so that we can
320 make clickable links from log messages carrying a MESSAGE_ID, that lead to
321 some explanatory text online.
323 * maybe extend .path units to expose fanotify() per-mount change events
325 * Add a "systemctl list-units --by-slice" mode or so, which rearranges the
326 output of "systemctl list-units" slightly by showing the tree structure of
327 the slices, and the units attached to them.
329 * the a-posteriori stopping of units bound to units that disappeared logic
330 should be reworked: there should be a queue of units, and we should only
331 enqueue stop jobs from a defer event that processes queue instead of
332 right-away when we find a unit that is bound to one that doesn't exist
333 anymore. (similar to how the stop-unneeded queue has been reworked the same
336 * nspawn: make nspawn suitable for shell pipelines: instead of triggering a
337 hangup when input is finished, send ^D, which synthesizes an EOF. Then wait
338 for hangup or ^D before passing on the EOF.
340 * When reloading configuration PID 1 should reset all its properties to the
341 original defaults before calling parse_config()
343 * nspawn: greater control over selinux label?
345 * hibernate/s2h: make this robust and safe to enable in Fedora by default.
348 1. add resume_offset support to the resume code (i.e. support swap files
350 2. check if swap is on weird storage and refuse if so
351 3. add auto-detection of hibernation images
353 * cgroups: use inotify to get notified when somebody else modifies cgroups
354 owned by us, then log a friendly warning.
356 * beef up log.c with support for stripping ANSI sequences from strings, so that
357 it is OK to include them in log strings. This would be particularly useful so
358 that our log messages could contain clickable links for example for unit
359 files and suchlike we operate on.
361 * add support for "portablectl attach http://foobar.com/waaa.raw (i.e. importd integration)
363 * add attach --enable and attach --now (for attach+enable+start)
365 * sync dynamic uids/gids between host+portable srvice (i.e. if DynamicUser=1 is set for a service, make sure that the
366 selected user is resolvable in the service even if it ships its own /etc/passwd)
368 * Fix DECIMAL_STR_MAX or DECIMAL_STR_WIDTH. One includes a trailing NUL, the
369 other doesn't. What a disaster. Probably to exclude it. Also
370 DECIMAL_STR_WIDTH should probably add an extra "-" into account for negative
373 * Check that users of inotify's IN_DELETE_SELF flag are using it properly, as
374 usually IN_ATTRIB is the right way to watch deleted files, as the former only
375 fires when a file is actually removed from disk, i.e. the link count drops to
376 zero and is not open anymore, while the latter happens when a file is
377 unlinked from any dir.
379 * port systemctl, busctl, … over to format-table.[ch]'s table formatters
381 * pid1: lock image configured with RootDirectory=/RootImage= using the usual nspawn semantics while the unit is up
383 * add --vacuum-xyz options to coredumpctl, matching those journalctl already has.
385 * introduce Ephemeral= unit file switch, that creates an ephemeral copy of all
386 files and directories that are left writable for a unit, and which are
387 removed after the unit goes down again. A bit like --ephemeral for
388 systemd-nspawn but for system services. If used together with RootImage= this
389 should reflink the image file itself.
391 Related: add Ephemeral=<path1> <path2> … which would allow marking
392 specific paths only like this.
394 * add CopyFile= or so as unit file setting that may be used to copy files or
395 directory trees from the host to the services RootImage= and RootDirectory=
396 environment. Which we can use for /etc/machine-id and in particular
397 /etc/resolv.conf. Should be smart and do something useful on read-only
398 images, for example fallback to read-only bind mounting the file instead.
400 * show invocation ID in systemd-run output
402 * bypass SIGTERM state in unit files if KillSignal is SIGKILL
404 * add proper dbus APIs for the various sd_notify() commands, such as MAINPID=1
405 and so on, which would mean we could report errors and such.
407 * teach tmpfiles.d q/Q logic something sensible in the context of XFS/ext4
410 * introduce DefaultSlice= or so in system.conf that allows changing where we
411 place our units by default, i.e. change system.slice to something
412 else. Similar, ManagerSlice= should exist so that PID1's own scope unit could
413 be moved somewhere else too. Finally machined and logind should get similar
414 options so that it is possible to move user session scopes and machines to a
415 different slice too by default. Usecase: people who want to put resources on
416 the entire system, with the exception of one specific service. See:
417 https://lists.freedesktop.org/archives/systemd-devel/2018-February/040369.html
419 * maybe rework get_user_creds() to query the user database if $SHELL is used
420 for root, but only then.
422 * be stricter with fds we receive for the fdstore: close them asynchronously
424 * calenderspec: add support for week numbers and day numbers within a
425 year. This would allow us to define "bi-weekly" triggers safely.
427 * sd-bus: add vtable flag, that may be used to request client creds implicitly
428 and asynchronously before dispatching the operation
430 * make use of ethtool veth peer info in machined, for automatically finding out
431 host-side interface pointing to the container.
433 * add some special mode to LogsDirectory=/StateDirectory=… that allows
434 declaring these directories without necessarily pulling in deps for them, or
435 creating them when starting up. That way, we could declare that
436 systemd-journald writes to /var/log/journal, which could be useful when we
437 doing disk usage calculations and so on.
439 * taint systemd if there are fewer than 65536 users assigned (userns) to the system.
441 * deprecate RootDirectoryStartOnly= in favour of a new ExecStart= prefix char
443 * add a new RuntimeDirectoryPreserve= mode that defines a similar lifecycle for
444 the runtime dir as we maintain for the fdstore: i.e. keep it around as long
445 as the unit is running or has a job queued.
447 * support projid-based quota in machinectl for containers
449 * maybe use SOURCE_DATE_EPOCH (i.e. the env var the reproducible builds folks
450 introduced) as the RTC epoch, instead of the mtime of NEWS.
452 * add a way to lock down cgroup migration: a boolean, which when set for a unit
453 makes sure the processes in it can never migrate out of it
455 * blog about fd store and restartable services
457 * document Environment=SYSTEMD_LOG_LEVEL=debug drop-in in debugging document
459 * rework ExecOutput and ExecInput enums so that EXEC_OUTPUT_NULL loses its
460 magic meaning and is no longer upgraded to something else if set explicitly.
462 * in the long run: permit a system with /etc/machine-id linked to /dev/null, to
463 make it lose its identity, i.e. be anonymous. For this we'd have to patch
464 through the whole tree to make all code deal with the case where no machine
467 * optionally, collect cgroup resource data, and store it in per-unit RRD files,
468 suitable for processing with rrdtool. Add bus API to access this data, and
469 possibly implement a CPULoad property based on it.
471 * beef up pam_systemd to take unit file settings such as cgroups properties as
474 * maybe hook of xfs/ext4 quotactl() with services? i.e. automatically manage
475 the quota of a the user indicated in User= via unit file settings, like the
476 other resource management concepts. Would mix nicely with DynamicUser=1. Or
477 alternatively, do this with projids, so that we can also cover services
478 running as root. Quota should probably cover all the special dirs such as
479 StateDirectory=, LogsDirectory=, CacheDirectory=, as well as RootDirectory= if it
480 is set, plus the whole disk space any image configured with RootImage=.
482 * In DynamicUser= mode: before selecting a UID, use disk quota APIs on relevant
483 disks to see if the UID is already in use.
485 * add "systemctl wait" or so, which does what "systemd-run --wait" does, but
486 for all units. It should be both a way to pin units into memory as well as a
487 wait to retrieve their exit data.
489 * expose IO accounting data on the bus, show it in systemd-run --wait and log
490 about it in the resource log message
492 * show whether a service has out-of-date configuration in "systemctl status" by
493 using mtime data of ConfigurationDirectory=.
495 * Add AddUser= setting to unit files, similar to DynamicUser=1 which however
496 creates a static, persistent user rather than a dynamic, transient user. We
497 can leverage code from sysusers.d for this.
499 * add some optional flag to ReadWritePaths= and friends, that has the effect
500 that we create the dir in question when the service is started. Example:
502 ReadWritePaths=:/var/lib/foobar
504 * maybe add call sd_journal_set_block_timeout() or so to set SO_SNDTIMEO for
505 the sd-journal logging socket, and, if the timeout is set to 0, sets
506 O_NONBLOCK on it. That way people can control if and when to block for
509 * hostnamed: populate form factor data from a new hwdb database, so that old
510 yogas can be recognized as "convertible" too, even if they predate the DMI
511 "convertible" form factor
513 * Maybe add PrivatePIDs= as new unit setting, and do minimal PID namespacing
514 after all. Be strict however, only support the equivalent of nspawn's
515 --as-pid2 switch, and sanely proxy sd_notify() messages dropping stuff such
518 * Add ExecMonitor= setting. May be used multiple times. Forks off a process in
519 the service cgroup, which is supposed to monitor the service, and when it
520 exits the service is considered failed by its monitor.
522 * track the per-service PAM process properly (i.e. as an additional control
523 process), so that it may be queried on the bus and everything.
525 * add a new "debug" job mode, that is propagated to unit_start() and for
526 services results in two things: we raise SIGSTOP right before invoking
527 execve() and turn off watchdog support. Then, use that to implement
528 "systemd-gdb" for attaching to the start-up of any system service in its
531 * gpt-auto logic: related to the above, maybe support a "secondary" root
532 partition, that is mounted to / and is writable, and where the actual root's
533 /usr is mounted into.
535 * 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
537 * drop nss-myhostname in favour of nss-resolve?
539 * add a percentage syntax for TimeoutStopSec=, e.g. TimeoutStopSec=150%, and
540 then use that for the setting used in user@.service. It should be understood
541 relative to the configured default value.
543 * in networkd, when matching device types, fix up DEVTYPE rubbish the kernel passes to us
545 * enable LockMLOCK to take a percentage value relative to physical memory
547 * Permit masking specific netlink APIs with RestrictAddressFamily=
549 * nspawn: support that /proc, /sys/, /dev are pre-mounted
551 * define gpt header bits to select volatility mode
553 * ProtectClock= (drops CAP_SYS_TIMES, adds seecomp filters for settimeofday, adjtimex), sets DeviceAllow o /dev/rtc
555 * ProtectTracing= (drops CAP_SYS_PTRACE, blocks ptrace syscall, makes /sys/kernel/tracing go away)
557 * ProtectMount= (drop mount/umount/pivot_root from seccomp, disallow fuse via DeviceAllow, imply Mountflags=slave)
559 * ProtectKeyRing= to take keyring calls away
561 * RemoveKeyRing= to remove all keyring entries of the specified user
563 * ProtectReboot= that masks reboot() and kexec_load() syscalls, prohibits kill
564 on PID 1 with the relevant signals, and makes relevant files in /sys and
565 /proc (such as the sysrq stuff) unavailable
567 * make sure the ratelimit object can deal with USEC_INFINITY as way to turn off things
569 * journalctl: make sure -f ends when the container indicated by -M terminates
571 * mount: automatically search for "main" partition of an image has multiple
574 * in nss-systemd, if we run inside of RootDirectory= with PrivateUsers= set,
575 find a way to map the User=/Group= of the service to the right name. This way
576 a user/group for a service only has to exist on the host for the right
579 * add bus API for creating unit files in /etc, reusing the code for transient units
581 * add bus API to remove unit files from /etc
583 * add bus API to retrieve current unit file contents (i.e. implement "systemctl cat" on the bus only)
585 * rework fopen_temporary() to make use of open_tmpfile_linkable() (problem: the
586 kernel doesn't support linkat() that replaces existing files, currently)
588 * transient units: don't bother with actually setting unit properties, we
589 reload the unit file anyway
591 * journald: sigbus API via a signal-handler safe function that people may call
592 from the SIGBUS handler
594 * optionally, also require WATCHDOG=1 notifications during service start-up and shutdown
596 * delay activation of logind until somebody logs in, or when /dev/tty0 pulls it
597 in or lingering is on (so that containers don't bother with it until PAM is used). also exit-on-idle
599 * cache sd_event_now() result from before the first iteration...
601 * PID1: find a way how we can reload unit file configuration for
602 specific units only, without reloading the whole of systemd
604 * add an explicit parser for LimitRTPRIO= that verifies
605 the specified range and generates sane error messages for incorrect
608 * when we detect that there are waiting jobs but no running jobs, do something
610 * push CPUAffinity= also into the "cpuset" cgroup controller
612 * PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn)
614 * there's probably something wrong with having user mounts below /sys,
615 as we have for debugfs. for example, src/core/mount.c handles mounts
616 prefixed with /sys generally special.
617 http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html
619 * fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline
621 * docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date
623 * add a job mode that will fail if a transaction would mean stopping
624 running units. Use this in timedated to manage the NTP service
626 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html
628 * The udev blkid built-in should expose a property that reflects
629 whether media was sensed in USB CF/SD card readers. This should then
630 be used to control SYSTEMD_READY=1/0 so that USB card readers aren't
631 picked up by systemd unless they contain a medium. This would mirror
632 the behaviour we already have for CD drives.
634 * networkd/udev: implement SR_IOV configuration in .link files:
635 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html
637 * hostnamectl: show root image uuid
639 * Find a solution for SMACK capabilities stuff:
640 http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html
642 * "systemctl preset-all" should probably order the unit files it
643 operates on lexicographically before starting to work, in order to
644 ensure deterministic behaviour if two unit files conflict (like DMs
647 * synchronize console access with BSD locks:
648 http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html
650 * as soon as we have sender timestamps, revisit coalescing multiple parallel daemon reloads:
651 http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
653 * figure out when we can use the coarse timers
655 * add "systemctl start -v foobar.service" that shows logs of a service
656 while the start command runs. This is non-trivial to do without
657 races though, since we should flush out all journal messages before
658 returning from the "systemctl stop".
660 * 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
662 * maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment=
664 * maybe allow timer units with an empty Units= setting, so that they
665 can be used for resuming the system but nothing else.
667 * what to do about udev db binary stability for apps? (raw access is not an option)
669 * systemctl: if some operation fails, show log output?
671 * systemctl edit: use equivalent of cat() to insert existing config as a comment, prepended with #.
672 Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc.
674 * exponential backoff in timesyncd when we cannot reach a server
676 * timesyncd: add ugly bus calls to set NTP servers per-interface, for usage by NM
678 * merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share....
680 * add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL
681 (throughout the codebase, not only PID1)
685 - service registration
686 - service/domain/types browsing
688 - DNS-SD service registration from socket units
689 - resolved should optionally register additional per-interface LLMNR
690 names, so that for the container case we can establish the same name
691 (maybe "host") for referencing the server, everywhere.
692 - allow clients to request DNSSEC for a single lookup even if DNSSEC is off (?)
693 - hook up resolved with machined-based address resolution
695 * refcounting in sd-resolve is borked
697 * Add a new verb "systemctl top"
699 * add new gpt type for btrfs volumes
701 * support empty /etc boots nicely:
702 - nspawn/gpt-generator: introduce new gpt partition type for /usr
704 * generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
706 * a way for container managers to turn off getty starting via $container_headless= or so...
708 * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
710 * For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
711 they run added to the initial transaction and thus confuse Type=idle.
713 * add bus api to query unit file's X fields.
715 * gpt-auto-generator:
716 - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap?
717 - Make /home automount rather than mount?
719 * add generator that pulls in systemd-network from containers when
720 CAP_NET_ADMIN is set, more than the loopback device is defined, even
721 when it is otherwise off
723 * MessageQueueMessageSize= (and suchlike) should use parse_iec_size().
725 * implement Distribute= in socket units to allow running multiple
726 service instances processing the listening socket, and open this up
729 * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze
730 $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls
731 should SIGSTOP all unit processes in a loop until all processes of
732 it are fully stopped. This can later be used for app management by
733 desktop UIs such as gnome-shell to freeze apps that are not visible
734 on screen, not unlike how job control works on the shell
737 - implement per-slice CPUFairScheduling=1 switch
738 - introduce high-level settings for RT budget, swappiness
739 - how to reset dynamically changed unit cgroup attributes sanely?
740 - when reloading configuration, apply new cgroup configuration
741 - when recursively showing the cgroup hierarchy, optionally also show
742 the hierarchies of child processes
745 - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
747 * when we detect low battery and no AC on boot, show pretty splash and refuse boot
749 * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
751 * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
753 * 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
755 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
757 * If we try to find a unit via a dangling symlink, generate a clean
758 error. Currently, we just ignore it and read the unit from the search
761 * refuse boot if /usr/lib/os-release is missing or /etc/machine-id cannot be set up
763 * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
765 * load .d/*.conf dropins for device units
769 - GetAllProperties() on a non-existing object does not result in a failure currently
770 - port to sd-resolve for connecting to TCP dbus servers
771 - 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
772 - see if we can drop more message validation on the sending side
773 - add API to clone sd_bus_message objects
774 - longer term: priority inheritance
776 - NameLost/NameAcquired obsolete
779 - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
782 - allow multiple signal handlers per signal?
783 - document chaining of signal handler for SIGCHLD and child handlers
784 - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ...
786 * investigate endianness issues of UUID vs. GUID
788 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
789 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
791 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
793 * add a pam module that on password changes updates any LUKS slot where the password matches
796 - add unit tests for config_parse_device_allow()
798 * seems that when we follow symlinks to units we prefer the symlink
799 destination path over /etc and /usr. We should not do that. Instead
800 /etc should always override /run+/usr and also any symlink
803 * when isolating, try to figure out a way how we implicitly can order
804 all units we stop before the isolating unit...
806 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
808 * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify
809 that the boot loader should be installed to the ESP. Define a way
810 how an installer can figure out whether a BLS compliant boot loader
813 * think about requeuing jobs when daemon-reload is issued? usecase:
814 the initrd issues a reload after fstab from the host is accessible
815 and we might want to requeue the mounts local-fs acquired through
818 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
820 * remove any syslog support from log.c — we probably cannot do this before split-off udev is gone for good
822 * shutdown logging: store to EFI var, and store to USB stick?
824 * merge unit_kill_common() and unit_kill_context()
827 - honor language efi variables for default language selection (if there are any?)
828 - honor timezone efi variables for default timezone selection (if there are any?)
829 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
831 - recognize the case when not booted on EFI
833 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
835 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
837 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
840 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
841 - logind: wakelock/opportunistic suspend support
842 - Add pretty name for seats in logind
843 - logind: allow showing logout dialog from system?
844 - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
845 - if pam_systemd is invoked by su from a process that is outside of a
846 any session we should probably just become a NOP, since that's
847 usually not a real user session but just some system code that just
849 - logind: make the Suspend()/Hibernate() bus calls wait for the for
850 the job to be completed. before returning, so that clients can wait
851 for "systemctl suspend" to finish to know when the suspending is
853 - logind: when the power button is pressed short, just popup a
854 logout dialog. If it is pressed for 1s, do the usual
855 shutdown. Inspiration are Macs here.
856 - expose "Locked" property on logind sesison objects
857 - maybe allow configuration of the StopTimeout for session scopes
858 - rename session scope so that it includes the UID. THat way
859 the session scope can be arranged freely in slices and we don't have
860 make assumptions about their slice anymore.
861 - follow PropertiesChanged state more closely, to deal with quick logouts and
865 - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields
866 - journald: also get thread ID from client, plus thread name
867 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
868 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
869 - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively?
870 - declare the local journal protocol stable in the wiki interface chart
871 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
872 - journald: when dropping msgs due to ratelimit make sure to write
873 "dropped %u messages" not only when we are about to print the next
874 message that works, but already after a short timeout
875 - check if we can make journalctl by default use --follow mode inside of less if called without args?
876 - maybe add API to send pairs of iovecs via sd_journal_send
877 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
878 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
879 and !FOOBAR for events without FOOBAR.
880 - journal: store timestamp of journal_file_set_offline() int he header,
881 so it is possible to display when the file was last synced.
882 - 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.
883 - journal: find a way to allow dropping history early, based on priority, other rules
884 - journal: When used on NFS, check payload hashes
885 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
886 - refuse taking lower-case variable names in sd_journal_send() and friends.
887 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
888 - journal: deal nicely with byte-by-byte copied files, especially regards header
889 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
890 - Replace utmp, wtmp, btmp, and lastlog completely with journal
891 - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
892 - when a kernel driver logs in a tight loop, we should ratelimit that too.
893 - journald: optionally, log debug messages to /run but everything else to /var
894 - journald: when we drop syslog messages because the syslog socket is
895 full, make sure to write how many messages are lost as first thing
896 to syslog when it works again.
897 - journald: allow per-priority and per-service retention times when rotating/vacuuming
898 - journald: make use of uid-range.h to managed uid ranges to split
900 - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
901 - improve journalctl performance by loading journal files
902 lazily. Encode just enough information in the file name, so that we
903 do not have to open it to know that it is not interesting for us, for
904 the most common operations.
905 - man: document that corrupted journal files is nothing to act on
906 - rework journald sigbus stuff to use mutex
907 - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our
908 services that run under their own user ids, and use User= (but only
909 in a world where userns is ubiquitous since otherwise we cannot
910 invoke those daemons on the host AND in a container anymore). Also,
911 if LimitNPROC= is used without User= we should warn and refuse
913 - journalctl --verify: don't show files that are currently being
914 written to as FAIL, but instead show that their are being written to.
915 - add journalctl -H that talks via ssh to a remote peer and passes through
917 - add a version of --merge which also merges /var/log/journal/remote
918 - journalctl: -m should access container journals directly by enumerating
919 them via machined, and also watch containers coming and going.
920 Benefit: nspawn --ephemeral would start working nicely with the journal.
921 - assign MESSAGE_ID to log messages about failed services
923 * add a test if all entries in the catalog are properly formatted.
924 (Adding dashes in a catalog entry currently results in the catalog entry
925 being silently skipped. journalctl --update-catalog must warn about this,
926 and we should also have a unit test to check that all our message are OK.)
929 - document that deps in [Unit] sections ignore Alias= fields in
930 [Install] units of other units, unless those units are disabled
931 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
932 - document that service reload may be implemented as service reexec
933 - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
934 - document systemd-journal-flush.service properly
935 - documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
936 - man: document the very specific env the shutdown drop-in tools live in
937 - man: add more examples to man pages
938 - man: maybe sort directives in man pages, and take sections from --help and apply them to man too
939 - document root=gpt-auto properly
942 - add systemctl switch to dump transaction without executing it
943 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
944 - "systemctl disable" on a static unit prints no message and does
945 nothing. "systemctl enable" does nothing, and gives a bad message
946 about it. Should fix both to print nice actionable messages.
947 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
948 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
949 - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
950 - systemctl: "Journal has been rotated since unit was started." message is misleading
951 - systemctl status output should include list of triggering units and their status
954 - "systemctl mask" should find all names by which a unit is accessible
955 (i.e. by scanning for symlinks to it) and link them all to /dev/null
958 - timer units should get the ability to trigger when:
960 - Modulate timer frequency based on battery state
962 * add libsystemd-password or so to query passwords during boot using the password agent logic
964 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
966 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
968 * make repeated alt-ctrl-del presses printing a dump
970 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
972 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
975 - emulate /dev/kmsg using CUSE and turn off the syslog syscall
976 with seccomp. That should provide us with a useful log buffer that
977 systemd can log to during early boot, and disconnect container logs
978 from the kernel's logs.
979 - as soon as networkd has a bus interface, hook up --network-interface=,
980 --network-bridge= with networkd, to trigger netdev creation should an
982 - a nice way to boot up without machine id set, so that it is set at boot
983 automatically for supporting --ephemeral. Maybe hash the host machine id
984 together with the machine name to generate the machine id for the container
985 - fix logic always print a final newline on output.
986 https://github.com/systemd/systemd/pull/272#issuecomment-113153176
987 - should optionally support receiving WATCHDOG=1 messages from its payload
989 - optionally automatically add FORWARD rules to iptables whenever nspawn is
990 running, remove them when shut down.
993 - refuse mounting over a mount point
994 - automatically discover .roothash files in dissect, similarly to nspawn
997 - add an API so that libvirt-lxc can inform us about network interfaces being
998 removed or added to an existing machine
999 - "machinectl migrate" or similar to copy a container from or to a
1000 difference host, via ssh
1001 - introduce systemd-nspawn-ephemeral@.service, and hook it into
1002 "machinectl start" with a new --ephemeral switch
1003 - "machinectl status" should also show internal logs of the container in
1005 - "machinectl history"
1007 - "machinectl commit" that takes a writable snapshot of a tree, invokes a
1008 shell in it, and marks it read-only after use
1011 - cryptsetup-generator: allow specification of passwords in crypttab itself
1012 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
1014 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
1015 https://bugs.freedesktop.org/show_bug.cgi?id=54712
1017 * add a dependency on standard-conf.xml and other included files to man pages
1019 * MountFlags=shared acts as MountFlags=slave right now.
1021 * properly handle loop back mounts via fstab, especially regards to fsck/passno
1023 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
1028 - add trigger --subsystem-match=usb/usb_device device
1029 - reimport udev db after MOVE events for devices without dev_t
1031 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
1033 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
1036 - save coredump in Windows/Mozilla minidump format
1037 - when truncating coredumps, also log the full size that the process had, and make a metadata field so we can report truncated coredumps
1039 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
1041 * be able to specify a forced restart of service A where service B depends on, in case B
1042 needs to be auto-respawned?
1045 - apply "x" on "D" too (see patch from William Douglas)
1046 - instead of ignoring unknown fields, reject them.
1047 - creating new directories/subvolumes/fifos/device nodes
1048 should not follow symlinks. None of the other adjustment or creation
1049 calls follow symlinks.
1051 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
1053 * verify that the AF_UNIX sockets of a service in the fs still exist
1054 when we start a service in order to avoid confusion when a user
1055 assumes starting a service is enough to make it accessible
1057 * Make it possible to set the keymap independently from the font on
1058 the kernel cmdline. Right now setting one resets also the other.
1060 * and a dbus call to generate target from current state
1062 * write blog stories about:
1063 - hwdb: what belongs into it, lsusb
1064 - enabling dbus services
1065 - how to make changes to sysctl and sysfs attributes
1067 - how to pass throw-away units to systemd, or dynamically change properties of existing units
1068 - testing with Harald's awesome test kit
1070 - how to develop against journal browsing APIs
1071 - the journal HTTP iface
1072 - non-cgroup resource management
1073 - dynamic resource management with cgroups
1074 - refreshed, longer missions statement
1075 - calendar time events
1076 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
1077 - how to create your own target
1078 - instantiated apache, dovecot and so on
1079 - hooking a script into various stages of shutdown/rearly booot
1081 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
1083 * dot output for --test showing the 'initial transaction'
1086 - When logging about multiple units (stopping BoundTo units, conflicts, etc.),
1087 log both units as UNIT=, so that journalctl -u triggers on both.
1088 - generate better errors when people try to set transient properties
1089 that are not supported...
1090 http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html
1091 - maybe introduce WantsMountsFor=? Usecase:
1092 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html
1093 - recreate systemd's D-Bus private socket file on SIGUSR2
1094 - move PAM code into its own binary
1095 - when we automatically restart a service, ensure we restart its rdeps, too.
1096 - hide PAM options in fragment parser when compile time disabled
1097 - Support --test based on current system state
1098 - 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().
1099 - after deserializing sockets in socket.c we should reapply sockopts and things
1100 - drop PID 1 reloading, only do reexecing (difficult: Reload()
1101 currently is properly synchronous, Reexec() is weird, because we
1102 cannot delay the response properly until we are back, so instead of
1103 being properly synchronous we just keep open the fd and close it
1104 when done. That means clients do not get a successful method reply,
1105 but much rather a disconnect on success.
1106 - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
1107 - when a bus name of a service disappears from the bus make sure to queue further activation requests
1110 - allow port=0 in .socket units
1111 - maybe introduce ExecRestartPre=
1112 - add ReloadSignal= for configuring a reload signal to use
1113 - implement Register= switch in .socket units to enable registration
1114 in Avahi, RPC and other socket registration services.
1115 - allow Type=simple with PIDFile=
1116 https://bugzilla.redhat.com/show_bug.cgi?id=723942
1117 - allow writing multiple conditions in unit files on one line
1118 - introduce Type=pid-file
1119 - introduce mix of BindTo and Requisite
1120 - add a concept of RemainAfterExit= to scope units
1121 - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
1122 - add verification of [Install] section to systemd-analyze verify
1125 - Make sure ID_PATH is always exported and complete for
1126 network devices where possible, so we can safely rely
1130 - add support for more attribute types
1131 - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places
1134 - add more keys to [Route] and [Address] sections
1135 - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
1136 - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
1137 - add reduced [Link] support to .network files
1138 - add Scope= parsing option for [Network]
1139 - properly handle routerless dhcp leases
1140 - work with non-Ethernet devices
1141 - add support for more bond options
1142 - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from?
1143 - the DHCP lease data (such as NTP/DNS) is still made available when
1144 a carrier is lost on a link. It should be removed instantly.
1145 - expose in the API the following bits:
1146 - option 15, domain name and/or option 119, search list
1147 - option 12, host name and/or option 81, fqdn
1148 - option 123, 144, geolocation
1149 - option 252, configure http proxy (PAC/wpad)
1150 - provide a way to define a per-network interface default metric value
1151 for all routes to it. possibly a second default for DHCP routes.
1152 - allow Name= to be specified repeatedly in the [Match] section. Maybe also
1153 support Name=foo*|bar*|baz ?
1154 - duplicate address check for static IPs (like ARPCHECK in network-scripts)
1155 - allow DUID/IAID to be customized, see issue #394.
1156 - whenever uplink info changes, make DHCP server send out FORCERENEW
1158 * networkd-wait-online:
1159 - make operstates to wait for configurable?
1162 - figure out how much we can increase Maximum Message Size
1165 - add functions to set previously stored IPv6 addresses on startup and get
1166 them at shutdown; store them in client->ia_na
1167 - write more test cases
1168 - implement reconfigure support, see 5.3., 15.11. and 22.20.
1169 - implement support for temporary adressess (IA_TA)
1170 - implement dhcpv6 authentication
1171 - investigate the usefulness of Confirm messages; i.e. are there any
1172 situations where the link changes without any loss in carrier detection
1174 - some servers don't do rapid commit without a filled in IA_NA, verify
1181 - natively watch for dbus-*.service symlinks (PENDING)
1182 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
1184 * fix alsa mixer restore to not print error when no config is stored
1186 * make cryptsetup lower --iter-time
1188 * patch kernel for xattr support in /dev, /proc/, /sys?
1190 * kernel: add device_type = "fb", "fbcon" to class "graphics"
1192 * /usr/bin/service should actually show the new command line
1194 * 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=
1196 * fedora: F20: go timer units all the way, leave cron.daily for cron
1198 * neither pkexec nor sudo initialize environ[] from the PAM environment?
1200 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
1202 * register catalog database signature as file magic
1204 * zsh shell completion:
1205 - <command> <verb> -<TAB> should complete options, but currently does not
1206 - systemctl add-wants,add-requires
1208 * systemctl status should know about 'systemd-analyze calendar ... --iterations='
1209 * If timer has just OnInactiveSec=..., it should fire after a specified time
1210 after being started.
1214 * look for close() vs. close_nointr() vs. close_nointr_nofail()
1216 * check for strerror(r) instead of strerror(-r)
1220 * set_put(), hashmap_put() return values check. i.e. == 0 does not free()!
1222 * use secure_getenv() instead of getenv() where appropriate
1224 * link up selected blog stories from man pages and unit files Documentation= fields
1225 String is not UTF-8 clean, ignoring assignment