ci: add dependencies for repart + cryptsetup's pkcs#11 support
[thirdparty/systemd.git] / TODO
CommitLineData
afe3ab58 1Bugfixes:
c343be28 2
54fcb619
ZJS
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.
6
f38afcd0 7External:
f85857df 8
f38afcd0 9* Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
bafb15ba 10
83a1ff25
ZJS
11* wiki: update journal format documentation for lz4 additions
12
5e524b40
LP
13Janitorial Clean-ups:
14
5e524b40
LP
15* Rearrange tests so that the various test-xyz.c match a specific src/basic/xyz.c again
16
bb527e11
LP
17* rework mount.c and swap.c to follow proper state enumeration/deserialization
18 semantics, like we do for device.c now
19
c6407108
LP
20Features:
21
27cc3c9d
LP
22* when dissecting images, warn about unrecognized partition flags
23
24* honour specifiers in unit files that resolve to some very basic
25 /etc/os-release data, such as ID, VERSION_ID, BUILD_ID, VARIANT_ID.
26
91fc013f
LP
27* socket units: allow creating a udev monitor socket with ListenDevices= or so,
28 with matches, then actviate app thorugh that passing socket oveer
29
0e7e8544
LP
30* kill zenata, all hail weblate?
31
173c7873
LP
32* move discoverable partitions spec into markdown and our tree
33
34* sd-event: add native support for P_ALL waitid() watching, then move PID 1 to
35 it fo reaping assigned but unknown children. This needs to some special care
36 to operate somewhat sensibly in light of priorities: P_ALL will return
37 arbitrary processes, regardless of the priority we want to watch them with,
38 hence on each event loop iteration check all processes which we shall watch
39 with higher prio explicitly, and then watch the entire rest with P_ALL.
40
41* tweak sd-event's child watching: keep a prioq of children to watch and use
42 waitid() only on the children with the highest priority until one is waitable
43 and ignore all lower-prio ones from that point on
44
173c7873
LP
45* maybe introduce xattrs that can be set on the root dir of the root fs
46 partition that declare the volatility mode to use the image in. Previously I
47 thought marking this via GPT partition flags but that's not ideal since
48 that's outside of the LUKS encryption/verity verification, and we probably
49 shouldn't operate in a volatile mode unless we got told so from a trusted
50 source.
51
173c7873
LP
52* figure out automatic partition discovery when combining writable root dir
53 with immutable /usr
54
97e5cc88
LP
55* coredump: maybe when coredumping read a new xattr from /proc/$PID/exe that
56 may be used to mark a whole binary as non-coredumpable. Would fix:
57 https://bugs.freedesktop.org/show_bug.cgi?id=69447
58
f3e361c1
LP
59* teach parse_timestamp() timezones like the calendar spec already knows it
60
9ef3376b
LP
61* beef up hibernation to optionally do swapon/swapoff immediately before/after
62 the hibernation
63
c846b233
LP
64* beef up s2h to implement a battery watch loop: instead of entering
65 hibernation unconditionally after coming back from resume make a decision
66 based on the battery load level: if battery level is above a specific
67 threshold, go to suspend again, only hibernate if below it. This means we'd
68 stick to suspend usually, but fall back to hibernation only when battery runs
69 empty (well, subject to our sampling interval). Related to this, check if we
70 can make ACPI _BTP (i.e. /sys/class/power_supply/*/alarm) work for us too,
71 i.e. see if it can wake up machines from suspend, so that we could resume
72 automatically when the system is low on power and move automatically to
73 hibernation mode. (see
74 https://uefi.org/sites/default/files/resources/ACPI%206_2_A_Sept29.pdf
75 section 10.2.2.8 and
76 https://docs.microsoft.com/en-us/windows-hardware/design/device-experiences/modern-standby-wake-sources
77 at the end).
78
f293fc51
LP
79* add an explicit "vertical" mode to format-table, so that "systemctl
80 status"-like outputs (i.e. with a series of field names left and values
81 right) become genuine first class citizens, and we gain automatic, sane JSON
82 output for them.
83
84* dissector: invoke fsck on the file systems we encounter, after all ext4 is
85 still pretty popular (and we mount the ESP too with it after all, which is
86 fat)
87
88* systemd-firstboot: teach it dissector magic, so that you can point it to some
89 disk image and it will just set everything in it all behind the scenes.
90
91* systemd-firstboot: add --force mode that replaces existing configuration.
92
9c230b8f
LP
93* We should probably replace /var/log/README, /etc/rc.d/README with symlinks
94 that are linked to these places instead of copied. After all they are
95 constant vendor data.
96
312dc153
LP
97* maybe add kernel cmdline params: 1) to force first-boot mode + 2) to force
98 random seed crediting
99
100* nspawn: on cgroupsv1 issue cgroup empty handler process based on host events,
101 so that we make cgroup agent logic safe
102
103* nspawn/machined: add API to invoke binary in container, then use that as
104 fallback in "machinectl shell"
72bd8525 105
ac5dca64
LP
106* logind: rework pam_logind to also do a bus call in case of invocation from
107 user@.service, which returns the XDG_RUNTIME_DIR value, and make this
108 behaviour selectable via pam module option.
109
110* introduce a new per-process uuid, similar to the boot id, the machine id, the
111 invocation id, that is derived from process creds, specifically a hashed
112 combination of AT_RANDOM + getpid() + the starttime from
113 /proc/self/status. Then add these ids implicitly when logging. Deriving this
114 uuid from these three things has the benefit that it can be derived easily
115 from /proc/$PID/ in a stable, and unique way that changes on both fork() and
116 exec().
117
118* let's not GC a unit while its ratelimits are still pending
119
5daeeecf
LP
120* when killing due to service watchdog timeout maybe detect whether target
121 process is under ptracing and then log loudly and continue instead.
122
5802d977
LP
123* introduce a new group to own TPM devices
124
125* make rfkill uaccess controllable by default, i.e. steal rule from
126 gnome-bluetooth and friends
127
d84520e9
LP
128* tweak journald context caching. In addition to caching per-process attributes
129 keyed by PID, cache per-cgroup attributes (i.e. the various xattrs we read)
130 keyed by cgroup path, and guarded by ctime changes. This should provide us
131 with a nice speed-up on services that have many processes running in the same
132 cgroup.
133
03ccc7f0
LP
134* make MAINPID= message reception checks even stricter: if service uses User=,
135 then check sending UID and ignore message if it doesn't match the user or
136 root.
137
138* maybe trigger a uevent "change" on a device if "systemctl reload xyz.device"
139 is issued.
140
db3cea22
LP
141* when importing an fs tree with machined, optionally apply userns-rec-chown
142
143* when importing an fs tree with machined, complain if image is not an OS
144
707b3fbd
LP
145* Maybe introduce a helper safe_exec() or so, which is to execve() which
146 safe_fork() is to fork(). And then make revert the RLIMIT_NOFILE soft limit
147 to 1K implicitly, unless explicitly opted-out.
d7b659ef 148
d238709c 149* rework seccomp/nnp logic that even if User= is used in combination with
d96c081a
LP
150 a seccomp option we don't have to set NNP. For that, change uid first whil
151 keeping CAP_SYS_ADMIN, then apply seccomp, the drop cap.
152
06bd149a
LP
153* add a concept for automatically loading per-unit secrets off disk and
154 inserting them into the kernel keyring. Maybe SecretsDirectory= similar to
155 ConfigurationDirectory=.
156
8a7cf157
LP
157* when no locale is configured, default to UEFI's PlatformLang variable
158
d3584ec4 159* bootctl,sd-boot: actually honour the "architecture" key
fe1bff15 160
d96c081a
LP
161* add a new syscall group "@esoteric" for more esoteric stuff such as bpf() and
162 usefaultd() and make systemd-analyze check for it.
163
06898123 164* paranoia: whenever we process passwords, call mlock() on the memory
309c6b19 165 first. i.e. look for all places we use free_and_erasep() and
9ae4ef49 166 augment them with mlock(). Also use MADV_DONTDUMP.
06898123 167
32875617
LP
168* Move RestrictAddressFamily= to the new cgroup create socket
169
170* support the bind/connect/sendmsg cgroup stuff for sandboxing, and possibly
171 patching around
172
76853293
LP
173* maybe implicitly attach monotonic+realtime timestamps to outgoing messages in
174 log.c and sd-journal-send
175
06898123
LP
176* optionally: turn on cgroup delegation for per-session scope units
177
d3aeddb8
LP
178* introduce per-unit (i.e. per-slice, per-service) journal log size limits.
179
af560fc1
LP
180* optionally, if a per-partition GPT flag is set for the root/home/… partitions
181 format the partition on next boot and unset the flag, in order to implement
182 factory reset. also, add a second flag that simply indicates whether such a
183 scheme is supported. then, add a tool (or maybe beef up systemd-dissect) to
184 show state of these flags, and optionally trigger such a factory reset on
185 next boot by setting the flag.
186
42e18088
LP
187* sd-boot: automatically load EFI modules from some drop-in dir, so that people
188 can add in file system drivers and such
189
42e18088
LP
190* sd-boot: optionally, show boot menu when previous default boot item has
191 non-zero "tries done" count
192
8f2eb730
LP
193* introduce an option (or replacement) for "systemctl show" that outputs all
194 properties as JSON, similar to busctl's new JSON output. In contrast to that
195 it should skip the variant type string though.
196
197* augment CODE_FILE=, CODE_LINE= with something like CODE_BASE= or so which
198 contains some identifier for the project, which allows us to include
199 clickable links to source files generating these log messages. The identifier
200 could be some abberviated URL prefix or so (taking inspiration from Go
201 imports). For example, for systemd we could use
202 CODE_BASE=github.com/systemd/systemd/blob/98b0b1123cc or so which is
203 sufficient to build a link by prefixing "http://" and suffixing the
204 CODE_FILE.
205
8f2eb730
LP
206* Augment MESSAGE_ID with MESSAGE_BASE, in a similar fashion so that we can
207 make clickable links from log messages carrying a MESSAGE_ID, that lead to
208 some explanatory text online.
209
7bd4bcf7
LP
210* maybe extend .path units to expose fanotify() per-mount change events
211
a33ffd15
LP
212* Add a "systemctl list-units --by-slice" mode or so, which rearranges the
213 output of "systemctl list-units" slightly by showing the tree structure of
214 the slices, and the units attached to them.
215
20d4b101
LP
216* the a-posteriori stopping of units bound to units that disappeared logic
217 should be reworked: there should be a queue of units, and we should only
218 enqeue stop jobs from a defer event that processes queue instead of
219 right-away when we find a unit that is bound to one that doesn't exist
220 anymore. (similar to how the stop-unneeded queue has been reworked the same
221 way)
4e8113bb 222
126708fa
LP
223* nspawn: make nspawn suitable for shell pipelines: instead of triggering a
224 hangup when input is finished, send ^D, which synthesizes an EOF. Then wait
225 for hangup or ^D before passing on the EOF.
226
bb527e11
LP
227* When reloading configuration PID 1 should reset all its properties to the
228 original defaults before calling parse_config()
229
72d711ef 230* nspawn: greater control over selinux label?
c7db2a9d 231
c633b0a6
LP
232* hibernate/s2h: make this robust and safe to enable in Fedora by default.
233 Specifically:
234
235 1. add resume_offset support to the resume code (i.e. support swap files
236 properly)
e83419d0 237 2. check if swap is on weird storage and refuse if so
46c41478 238 3. add auto-detection of hibernation images
c633b0a6 239
070d0ac9
LP
240* cgroups: use inotify to get notified when somebody else modifies cgroups
241 owned by us, then log a friendly warning.
242
e44924f5
LP
243* beef up log.c with support for stripping ANSI sequences from strings, so that
244 it is OK to include them in log strings. This would be particularly useful so
245 that our log messages could contain clickable links for example for unit
246 files and suchlike we operate on.
070d0ac9 247
8c4c2dfc
LP
248* add support for "portablectl attach http://foobar.com/waaa.raw (i.e. importd integration)
249
250* add attach --enable and attach --now (for attach+enable+start)
251
252* sync dynamic uids/gids between host+portable srvice (i.e. if DynamicUser=1 is set for a service, make sure that the
253 selected user is resolvable in the service even if it ships its own /etc/passwd)
254
5da19043 255* Fix DECIMAL_STR_MAX or DECIMAL_STR_WIDTH. One includes a trailing NUL, the
5238e957 256 other doesn't. What a disaster. Probably to exclude it. Also
5da19043
LP
257 DECIMAL_STR_WIDTH should probably add an extra "-" into account for negative
258 numbers.
259
16270697
LP
260* Check that users of inotify's IN_DELETE_SELF flag are using it properly, as
261 usually IN_ATTRIB is the right way to watch deleted files, as the former only
262 fires when a file is actually removed from disk, i.e. the link count drops to
263 zero and is not open anymore, while the latter happens when a file is
264 unlinked from any dir.
265
bd1b3f75 266* port systemctl, busctl, … over to format-table.[ch]'s table formatters
5da19043 267
7bc756ff
LP
268* pid1: lock image configured with RootDirectory=/RootImage= using the usual nspawn semantics while the unit is up
269
270* add --vacuum-xyz options to coredumpctl, matching those journalctl already has.
271
53c70a27
LP
272* introduce Ephemeral= unit file switch, that creates an ephemeral copy of all
273 files and directories that are left writable for a unit, and which are
274 removed after the unit goes down again. A bit like --ephemeral for
275 systemd-nspawn but for system services. If used together with RootImage= this
276 should reflink the image file itself.
277
278 Related: add Ephemeral=<path1> <path2> … which would allow marking
279 specific paths only like this.
280
53c70a27 281* add CopyFile= or so as unit file setting that may be used to copy files or
5238e957 282 directory trees from the host to the services RootImage= and RootDirectory=
53c70a27
LP
283 environment. Which we can use for /etc/machine-id and in particular
284 /etc/resolv.conf. Should be smart and do something useful on read-only
285 images, for example fallback to read-only bind mounting the file instead.
286
d9b50610
LP
287* show invocation ID in systemd-run output
288
289* bypass SIGTERM state in unit files if KillSignal is SIGKILL
290
9711b1ad
LP
291* add proper dbus APIs for the various sd_notify() commands, such as MAINPID=1
292 and so on, which would mean we could report errors and such.
293
1e784321
LP
294* teach tmpfiles.d q/Q logic something sensible in the context of XFS/ext4
295 project quota
0d545655 296
6b7b0f39
LP
297* introduce DefaultSlice= or so in system.conf that allows changing where we
298 place our units by default, i.e. change system.slice to something
299 else. Similar, ManagerSlice= should exist so that PID1's own scope unit could
300 be moved somewhere else too. Finally machined and logind should get similar
301 options so that it is possible to move user session scopes and machines to a
302 different slice too by default. Usecase: people who want to put resources on
303 the entire system, with the exception of one specific service. See:
304 https://lists.freedesktop.org/archives/systemd-devel/2018-February/040369.html
305
aa79f932
LP
306* maybe rework get_user_creds() to query the user database if $SHELL is used
307 for root, but only then.
586a8e93 308
46099c9e
LP
309* be stricter with fds we receive for the fdstore: close them asynchronously
310
46099c9e
LP
311* calenderspec: add support for week numbers and day numbers within a
312 year. This would allow us to define "bi-weekly" triggers safely.
313
c3cd7cc9
LP
314* sd-bus: add vtable flag, that may be used to request client creds implicitly
315 and asynchronously before dispatching the operation
316
06345858
LP
317* make use of ethtool veth peer info in machined, for automatically finding out
318 host-side interface pointing to the container.
319
48f1b5e5
LP
320* add some special mode to LogsDirectory=/StateDirectory=… that allows
321 declaring these directories without necessarily pulling in deps for them, or
322 creating them when starting up. That way, we could declare that
323 systemd-journald writes to /var/log/journal, which could be useful when we
324 doing disk usage calculations and so on.
325
cb77e122 326* taint systemd if there are fewer than 65536 users assigned (userns) to the system.
c6009ff0 327
899feb72 328* deprecate RootDirectoryStartOnly= in favour of a new ExecStart= prefix char
c6009ff0 329
3d80d454
LP
330* add a new RuntimeDirectoryPreserve= mode that defines a similar lifecycle for
331 the runtime dir as we maintain for the fdstore: i.e. keep it around as long
332 as the unit is running or has a job queued.
333
5f7ecd61 334* support projid-based quota in machinectl for containers
5962e9db 335
17b6f896
LP
336* maybe use SOURCE_DATE_EPOCH (i.e. the env var the reproducible builds folks
337 introduced) as the RTC epoch, instead of the mtime of NEWS.
338
17b6f896
LP
339* add a way to lock down cgroup migration: a boolean, which when set for a unit
340 makes sure the processes in it can never migrate out of it
341
17b6f896
LP
342* blog about fd store and restartable services
343
344* document Environment=SYSTEMD_LOG_LEVEL=debug drop-in in debugging document
345
370f9c21
LP
346* rework ExecOutput and ExecInput enums so that EXEC_OUTPUT_NULL loses its
347 magic meaning and is no longer upgraded to something else if set explicitly.
348
17b6f896
LP
349* in the long run: permit a system with /etc/machine-id linked to /dev/null, to
350 make it lose its identity, i.e. be anonymous. For this we'd have to patch
351 through the whole tree to make all code deal with the case where no machine
352 ID is available.
353
354* optionally, collect cgroup resource data, and store it in per-unit RRD files,
355 suitable for processing with rrdtool. Add bus API to access this data, and
356 possibly implement a CPULoad property based on it.
357
e7e4a258
LP
358* beef up pam_systemd to take unit file settings such as cgroups properties as
359 parameters
360
6fc373ee
LP
361* maybe hook of xfs/ext4 quotactl() with services? i.e. automatically manage
362 the quota of a the user indicated in User= via unit file settings, like the
5962e9db
LP
363 other resource management concepts. Would mix nicely with DynamicUser=1. Or
364 alternatively, do this with projids, so that we can also cover services
365 running as root. Quota should probably cover all the special dirs such as
366 StateDirectory=, LogsDirectory=, CacheDirectory=, as well as RootDirectory= if it
367 is set, plus the whole disk space any image configured with RootImage=.
6fc373ee 368
9a92a97a
LP
369* In DynamicUser= mode: before selecting a UID, use disk quota APIs on relevant
370 disks to see if the UID is already in use.
371
2368e974
LP
372* add "systemctl wait" or so, which does what "systemd-run --wait" does, but
373 for all units. It should be both a way to pin units into memory as well as a
374 wait to retrieve their exit data.
375
fa991fb7
LP
376* expose IO accounting data on the bus, show it in systemd-run --wait and log
377 about it in the resource log message
378
fa991fb7
LP
379* show whether a service has out-of-date configuration in "systemctl status" by
380 using mtime data of ConfigurationDirectory=.
381
d73b607d
LP
382* Add AddUser= setting to unit files, similar to DynamicUser=1 which however
383 creates a static, persistent user rather than a dynamic, transient user. We
384 can leverage code from sysusers.d for this.
385
fd63f36c
LP
386* add some optional flag to ReadWritePaths= and friends, that has the effect
387 that we create the dir in question when the service is started. Example:
388
389 ReadWritePaths=:/var/lib/foobar
390
3576dddf
LP
391* maybe add call sd_journal_set_block_timeout() or so to set SO_SNDTIMEO for
392 the sd-journal logging socket, and, if the timeout is set to 0, sets
393 O_NONBLOCK on it. That way people can control if and when to block for
394 logging.
395
fe6a0235
LP
396* hostnamed: populate form factor data from a new hwdb database, so that old
397 yogas can be recognized as "convertible" too, even if they predate the DMI
398 "convertible" form factor
399
08422e63
LP
400* Maybe add a small tool invoked early at boot, that adds in or resizes
401 partitions automatically, to be used when the media used is actually larger
402 than the image written onto it is.
403
d543b9f5
LP
404* Maybe add PrivatePIDs= as new unit setting, and do minimal PID namespacing
405 after all. Be strict however, only support the equivalent of nspawn's
406 --as-pid2 switch, and sanely proxy sd_notify() messages dropping stuff such
407 as MAINPID.
408
f59d1da8
LP
409* Add ExecMonitor= setting. May be used multiple times. Forks off a process in
410 the service cgroup, which is supposed to monitor the service, and when it
411 exits the service is considered failed by its monitor.
412
33bac67b
LP
413* track the per-service PAM process properly (i.e. as an additional control
414 process), so that it may be queried on the bus and everything.
415
416* add a new "debug" job mode, that is propagated to unit_start() and for
417 services results in two things: we raise SIGSTOP right before invoking
418 execve() and turn off watchdog support. Then, use that to implement
419 "systemd-gdb" for attaching to the start-up of any system service in its
420 natural habitat.
421
8eb7383b
LP
422* maybe add gpt-partition-based user management: each user gets his own
423 LUKS-encrypted GPT partition with a new GPT type. A small nss module
424 enumerates users via udev partition enumeration. UIDs are assigned in a fixed
425 way: the partition index is added as offset to some fixed base uid. User name
426 is stored in GPT partition name. A PAM module authenticates the user via the
427 LUKS partition password. Benefits: strong per-user security, compatibility
428 with stateless/read-only/verity-enabled root. (other idea: do this based on
429 loopback files in /home, without GPT involvement)
430
8eb7383b
LP
431* gpt-auto logic: related to the above, maybe support a "secondary" root
432 partition, that is mounted to / and is writable, and where the actual root's
433 /usr is mounted into.
434
8eb7383b
LP
435* 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
436
67234d21
LP
437* drop nss-myhostname in favour of nss-resolve?
438
b5bdbcd5
LP
439* add a percentage syntax for TimeoutStopSec=, e.g. TimeoutStopSec=150%, and
440 then use that for the setting used in user@.service. It should be understood
441 relative to the configured default value.
442
d21494ea
LP
443* in networkd, when matching device types, fix up DEVTYPE rubbish the kernel passes to us
444
445* enable LockMLOCK to take a percentage value relative to physical memory
446
04397464 447* Permit masking specific netlink APIs with RestrictAddressFamily=
d82047be 448
d82047be
LP
449* nspawn: support that /proc, /sys/, /dev are pre-mounted
450
d82047be
LP
451* define gpt header bits to select volatility mode
452
42d61ded
LP
453* ProtectClock= (drops CAP_SYS_TIMES, adds seecomp filters for settimeofday, adjtimex), sets DeviceAllow o /dev/rtc
454
04397464 455* ProtectTracing= (drops CAP_SYS_PTRACE, blocks ptrace syscall, makes /sys/kernel/tracing go away)
42d61ded 456
04397464 457* ProtectMount= (drop mount/umount/pivot_root from seccomp, disallow fuse via DeviceAllow, imply Mountflags=slave)
563a69f4 458
04397464 459* ProtectKeyRing= to take keyring calls away
2c5f2958 460
8ce9b83a 461* RemoveKeyRing= to remove all keyring entries of the specified user
a46eac1b
LP
462
463* ProtectReboot= that masks reboot() and kexec_load() syscalls, prohibits kill
464 on PID 1 with the relevant signals, and makes relevant files in /sys and
465 /proc (such as the sysrq stuff) unavailable
466
e40a326c 467* make sure the ratelimit object can deal with USEC_INFINITY as way to turn off things
89f193fa 468
6809de5b
LP
469* journalctl: make sure -f ends when the container indicated by -M terminates
470
2c5f2958
LP
471* mount: automatically search for "main" partition of an image has multiple
472 partitions
473
8ce9b83a
LP
474* in nss-systemd, if we run inside of RootDirectory= with PrivateUsers= set,
475 find a way to map the User=/Group= of the service to the right name. This way
476 a user/group for a service only has to exist on the host for the right
477 mapping to work.
478
2c5f2958
LP
479* add bus API for creating unit files in /etc, reusing the code for transient units
480
481* add bus API to remove unit files from /etc
482
483* add bus API to retrieve current unit file contents (i.e. implement "systemctl cat" on the bus only)
484
b8c7afdf
LP
485* rework fopen_temporary() to make use of open_tmpfile_linkable() (problem: the
486 kernel doesn't support linkat() that replaces existing files, currently)
487
1e555cb5
LP
488* transient units: don't bother with actually setting unit properties, we
489 reload the unit file anyway
490
1e555cb5
LP
491* journald: sigbus API via a signal-handler safe function that people may call
492 from the SIGBUS handler
493
c8048350
LP
494* optionally, also require WATCHDOG=1 notifications during service start-up and shutdown
495
648b1220
LP
496* delay activation of logind until somebody logs in, or when /dev/tty0 pulls it
497 in or lingering is on (so that containers don't bother with it until PAM is used). also exit-on-idle
498
3d39e6e5
LP
499* cache sd_event_now() result from before the first iteration...
500
bd098bce
LP
501* PID1: find a way how we can reload unit file configuration for
502 specific units only, without reloading the whole of systemd
503
f9bf1b8f 504* add an explicit parser for LimitRTPRIO= that verifies
de7399eb 505 the specified range and generates sane error messages for incorrect
f9bf1b8f 506 specifications.
de7399eb 507
3efc8c72
LP
508* when we detect that there are waiting jobs but no running jobs, do something
509
5ebbb45b 510* push CPUAffinity= also into the "cpuset" cgroup controller
03364e47 511
e6a26d8c
LP
512* PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn)
513
a2088fd0 514* there's probably something wrong with having user mounts below /sys,
5238e957 515 as we have for debugfs. for example, src/core/mount.c handles mounts
a2088fd0
LP
516 prefixed with /sys generally special.
517 http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html
518
8aa20381
LP
519* fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline
520
b18d23d7
LP
521* docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date
522
60d17b74
LP
523* add a job mode that will fail if a transaction would mean stopping
524 running units. Use this in timedated to manage the NTP service
525 state.
526 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html
527
477e75ef
LP
528* The udev blkid built-in should expose a property that reflects
529 whether media was sensed in USB CF/SD card readers. This should then
530 be used to control SYSTEMD_READY=1/0 so that USB card readers aren't
531 picked up by systemd unless they contain a medium. This would mirror
532 the behaviour we already have for CD drives.
533
2bc39683
LP
534* networkd/udev: implement SR_IOV configuration in .link files:
535 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html
536
c3a0d00d
LP
537* hostnamectl: show root image uuid
538
d2f81fb0
LP
539* Find a solution for SMACK capabilities stuff:
540 http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html
98cd2651 541
2a441c8a
LP
542* "systemctl preset-all" should probably order the unit files it
543 operates on lexicographically before starting to work, in order to
544 ensure deterministic behaviour if two unit files conflict (like DMs
545 do, for example)
546
0a86c1a9
LP
547* synchronize console access with BSD locks:
548 http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html
549
e031c227 550* as soon as we have sender timestamps, revisit coalescing multiple parallel daemon reloads:
0a86c1a9
LP
551 http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
552
8514b677
LP
553* figure out when we can use the coarse timers
554
7b0fce61
LP
555* add "systemctl start -v foobar.service" that shows logs of a service
556 while the start command runs. This is non-trivial to do without
557 races though, since we should flush out all journal messages before
558 returning from the "systemctl stop".
559
b568ef14 560* 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
c2d5b3c9 561
d48b7bd2
LP
562* maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment=
563
8483d73f
LP
564* maybe allow timer units with an empty Units= setting, so that they
565 can be used for resuming the system but nothing else.
566
25e773ee 567* what to do about udev db binary stability for apps? (raw access is not an option)
b857e042 568
279f0366 569* systemctl: if some operation fails, show log output?
533bb267 570
5238e957 571* systemctl edit: use equivalent of cat() to insert existing config as a comment, prepended with #.
b42de08a 572 Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc.
202aea45 573
720652b3 574* exponential backoff in timesyncd when we cannot reach a server
42aeb14a 575
720652b3 576* timesyncd: add ugly bus calls to set NTP servers per-interface, for usage by NM
e25b5a8d 577
563b1bdc
LP
578* merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share....
579
5dcf9838
LP
580* systemd.show_status= should probably have a mode where only failed
581 units are shown.
582
6bd7941e
TG
583* add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL
584 (throughout the codebase, not only PID1)
585
9d6db739 586* resolved:
9d6db739 587 - mDNS/DNS-SD
ccc3e8a1
LP
588 - service registration
589 - service/domain/types browsing
0f47ed0a 590 - avahi compat
9d6db739 591 - DNS-SD service registration from socket units
e25b5a8d
DH
592 - resolved should optionally register additional per-interface LLMNR
593 names, so that for the container case we can establish the same name
594 (maybe "host") for referencing the server, everywhere.
720652b3 595 - allow clients to request DNSSEC for a single lookup even if DNSSEC is off (?)
3efb871a 596 - hook up resolved with machined-based address resolution
3f77a1b1 597
e25b5a8d 598* refcounting in sd-resolve is borked
e2a69298 599
2834ffe7
LP
600* Add a new verb "systemctl top"
601
a940778f
LP
602* add new gpt type for btrfs volumes
603
a01647e5 604* support empty /etc boots nicely:
3c779fa5 605 - nspawn/gpt-generator: introduce new gpt partition type for /usr
a01647e5 606
3de03738
LP
607* generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
608
37efac5d
LP
609* a way for container managers to turn off getty starting via $container_headless= or so...
610
7348b3ad
LP
611* figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
612
81429136
KS
613* For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
614 they run added to the initial transaction and thus confuse Type=idle.
e107ed18 615
edb2935c
LP
616* add bus api to query unit file's X fields.
617
6a3f892a 618* gpt-auto-generator:
2a781fc9 619 - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap?
6a3f892a 620 - Make /home automount rather than mount?
6a3f892a 621
65026403
LP
622* add generator that pulls in systemd-network from containers when
623 CAP_NET_ADMIN is set, more than the loopback device is defined, even
624 when it is otherwise off
f8901862 625
f9bf1b8f 626* MessageQueueMessageSize= (and suchlike) should use parse_iec_size().
eda8f067 627
af1082b0
LP
628* implement Distribute= in socket units to allow running multiple
629 service instances processing the listening socket, and open this up
630 for ReusePort=
631
09ecd746
LP
632* introduce bus call FreezeUnit(s, b), as well as "systemctl freeze
633 $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls
634 should SIGSTOP all unit processes in a loop until all processes of
635 it are fully stopped. This can later be used for app management by
636 desktop UIs such as gnome-shell to freeze apps that are not visible
637 on screen, not unlike how job control works on the shell
638
f38afcd0 639* cgroups:
f38afcd0 640 - implement per-slice CPUFairScheduling=1 switch
f38afcd0
LP
641 - introduce high-level settings for RT budget, swappiness
642 - how to reset dynamically changed unit cgroup attributes sanely?
643 - when reloading configuration, apply new cgroup configuration
644 - when recursively showing the cgroup hierarchy, optionally also show
645 the hierarchies of child processes
0bee65f0 646
f38afcd0 647* transient units:
f38afcd0 648 - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
ebcf1f97 649
718db961
LP
650* when we detect low battery and no AC on boot, show pretty splash and refuse boot
651
718db961
LP
652* libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
653
966204e0
LP
654* be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
655
41644622
LP
656* 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
657
7f79cd71 658* After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
0aafd43d 659
19aadacf
JE
660* If we try to find a unit via a dangling symlink, generate a clean
661 error. Currently, we just ignore it and read the unit from the search
df5f6971
LP
662 path anyway.
663
04397464 664* refuse boot if /usr/lib/os-release is missing or /etc/machine-id cannot be set up
fcba531e 665
bdeeb6b5
LP
666* man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
667
668* load .d/*.conf dropins for device units
669
6bb648a1 670* sd-bus:
0d43ffef 671 - EBADSLT handling
fa607802 672 - GetAllProperties() on a non-existing object does not result in a failure currently
7212c608 673 - port to sd-resolve for connecting to TCP dbus servers
57f2a947 674 - 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
f38afcd0 675 - see if we can drop more message validation on the sending side
f38afcd0 676 - add API to clone sd_bus_message objects
ab9716c2 677 - longer term: priority inheritance
df98a87b 678 - dbus spec updates:
df98a87b
LP
679 - NameLost/NameAcquired obsolete
680 - GVariant
2ecfc64e 681 - path escaping
6a8b5fa4 682 - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
f38afcd0
LP
683
684* sd-event
12179984 685 - allow multiple signal handlers per signal?
0536ce5d 686 - document chaining of signal handler for SIGCHLD and child handlers
e25b5a8d 687 - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ...
eb01ba5d 688
69af4503 689* investigate endianness issues of UUID vs. GUID
e8a7a315 690
461bd8e4
LP
691* dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
692 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
693
ab8e074c
LP
694* add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
695
696* add a pam module that on password changes updates any LUKS slot where the password matches
697
fff87a35 698* test/:
20d52ab6 699 - add unit tests for config_parse_device_allow()
b8b4d3dd 700
b5c03638 701* seems that when we follow symlinks to units we prefer the symlink
d28315e4 702 destination path over /etc and /usr. We should not do that. Instead
b5c03638
LP
703 /etc should always override /run+/usr and also any symlink
704 destination.
705
eece8c6f
LP
706* when isolating, try to figure out a way how we implicitly can order
707 all units we stop before the isolating unit...
708
356ce991
LP
709* teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
710
c1446921
LP
711* BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify
712 that the boot loader should be installed to the ESP. Define a way
713 how an installer can figure out whether a BLS compliant boot loader
714 is installed.
795607b2 715
0be8342c
LP
716* think about requeuing jobs when daemon-reload is issued? usecase:
717 the initrd issues a reload after fstab from the host is accessible
718 and we might want to requeue the mounts local-fs acquired through
719 that automatically.
720
e5ec62c5 721* systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
54c31a79 722
ccddd104 723* remove any syslog support from log.c — we probably cannot do this before split-off udev is gone for good
826872b6 724
3679d112
LP
725* shutdown logging: store to EFI var, and store to USB stick?
726
356ce991 727* merge unit_kill_common() and unit_kill_context()
490b7e47 728
b44be3ec 729* EFI:
b44be3ec
LP
730 - honor language efi variables for default language selection (if there are any?)
731 - honor timezone efi variables for default timezone selection (if there are any?)
466784c8 732 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
631427d6 733* bootctl
631427d6 734 - recognize the case when not booted on EFI
e4181484 735
c0fe5db5 736* maybe do not install getty@tty1.service symlink in /etc but in /usr?
e9fd44b7 737
b44be3ec 738* print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
8ed20651 739
b44be3ec 740* mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
e6c6e7af 741
b44be3ec
LP
742* logind:
743 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
b44be3ec
LP
744 - logind: wakelock/opportunistic suspend support
745 - Add pretty name for seats in logind
746 - logind: allow showing logout dialog from system?
f38afcd0
LP
747 - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
748 - if pam_systemd is invoked by su from a process that is outside of a
749 any session we should probably just become a NOP, since that's
750 usually not a real user session but just some system code that just
751 needs setuid().
279f0366
LP
752 - logind: make the Suspend()/Hibernate() bus calls wait for the for
753 the job to be completed. before returning, so that clients can wait
754 for "systemctl suspend" to finish to know when the suspending is
755 complete.
756 - logind: when the power button is pressed short, just popup a
757 logout dialog. If it is pressed for 1s, do the usual
758 shutdown. Inspiration are Macs here.
279f0366 759 - expose "Locked" property on logind sesison objects
e25b5a8d
DH
760 - maybe allow configuration of the StopTimeout for session scopes
761 - rename session scope so that it includes the UID. THat way
762 the session scope can be arranged freely in slices and we don't have
763 make assumptions about their slice anymore.
764 - follow PropertiesChanged state more closely, to deal with quick logouts and
765 relogins
e673ad04 766
b44be3ec 767* journal:
57f2a947 768 - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields
b44be3ec
LP
769 - journald: also get thread ID from client, plus thread name
770 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
771 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
d28315e4 772 - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively?
b44be3ec 773 - declare the local journal protocol stable in the wiki interface chart
b44be3ec
LP
774 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
775 - journald: when dropping msgs due to ratelimit make sure to write
776 "dropped %u messages" not only when we are about to print the next
5238e957 777 message that works, but already after a short timeout
b44be3ec
LP
778 - check if we can make journalctl by default use --follow mode inside of less if called without args?
779 - maybe add API to send pairs of iovecs via sd_journal_send
f47ec8eb 780 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
b44be3ec
LP
781 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
782 and !FOOBAR for events without FOOBAR.
038cf334
ZJS
783 - journal: store timestamp of journal_file_set_offline() int he header,
784 so it is possible to display when the file was last synced.
b44be3ec
LP
785 - 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.
786 - journal: find a way to allow dropping history early, based on priority, other rules
787 - journal: When used on NFS, check payload hashes
b44be3ec
LP
788 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
789 - refuse taking lower-case variable names in sd_journal_send() and friends.
790 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
791 - journal: deal nicely with byte-by-byte copied files, especially regards header
b44be3ec 792 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
601d9d6f 793 - Replace utmp, wtmp, btmp, and lastlog completely with journal
f38afcd0 794 - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
f38afcd0
LP
795 - when a kernel driver logs in a tight loop, we should ratelimit that too.
796 - journald: optionally, log debug messages to /run but everything else to /var
797 - journald: when we drop syslog messages because the syslog socket is
798 full, make sure to write how many messages are lost as first thing
799 to syslog when it works again.
f38afcd0
LP
800 - change systemd-journal-flush into a service that stays around during
801 boot, and causes the journal to be moved back to /run on shutdown,
d28315e4 802 so that we do not keep /var busy. This needs to happen synchronously,
f38afcd0 803 hence doing this via signals is not going to work.
279f0366
LP
804 - optionally support running journald from the command line for testing purposes in external projects
805 - journald: allow per-priority and per-service retention times when rotating/vacuuming
806 - journald: make use of uid-range.h to managed uid ranges to split
807 journals in.
808 - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
809 - improve journalctl performance by loading journal files
810 lazily. Encode just enough information in the file name, so that we
811 do not have to open it to know that it is not interesting for us, for
812 the most common operations.
e25b5a8d 813 - man: document that corrupted journal files is nothing to act on
e25b5a8d
DH
814 - rework journald sigbus stuff to use mutex
815 - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our
816 services that run under their own user ids, and use User= (but only
817 in a world where userns is ubiquitous since otherwise we cannot
818 invoke those daemons on the host AND in a container anymore). Also,
819 if LimitNPROC= is used without User= we should warn and refuse
820 operation.
821 - journalctl --verify: don't show files that are currently being
822 written to as FAIL, but instead show that their are being written to.
823 - add journalctl -H that talks via ssh to a remote peer and passes through
824 binary logs data
e25b5a8d 825 - add a version of --merge which also merges /var/log/journal/remote
e25b5a8d
DH
826 - journalctl: -m should access container journals directly by enumerating
827 them via machined, and also watch containers coming and going.
828 Benefit: nspawn --ephemeral would start working nicely with the journal.
829 - assign MESSAGE_ID to log messages about failed services
b44be3ec 830
590171d1
ZJS
831* add a test if all entries in the catalog are properly formatted.
832 (Adding dashes in a catalog entry currently results in the catalog entry
833 being silently skipped. journalctl --update-catalog must warn about this,
834 and we should also have a unit test to check that all our message are OK.)
835
b44be3ec 836* document:
8b8f2591 837 - document that deps in [Unit] sections ignore Alias= fields in
b44be3ec
LP
838 [Install] units of other units, unless those units are disabled
839 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
b44be3ec 840 - document that service reload may be implemented as service reexec
f38afcd0
LP
841 - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
842 - document systemd-journal-flush.service properly
f38afcd0
LP
843 - documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
844 - man: document the very specific env the shutdown drop-in tools live in
f38afcd0
LP
845 - man: add more examples to man pages
846 - man: maybe sort directives in man pages, and take sections from --help and apply them to man too
17ec531f 847 - document root=gpt-auto properly
b44be3ec
LP
848
849* systemctl:
b44be3ec
LP
850 - add systemctl switch to dump transaction without executing it
851 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
852 - "systemctl disable" on a static unit prints no message and does
853 nothing. "systemctl enable" does nothing, and gives a bad message
854 about it. Should fix both to print nice actionable messages.
855 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
856 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
d28315e4 857 - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
b44be3ec 858 - systemctl: "Journal has been rotated since unit was started." message is misleading
61233823 859 - systemctl status output should include list of triggering units and their status
f38afcd0
LP
860
861* unit install:
862 - "systemctl mask" should find all names by which a unit is accessible
863 (i.e. by scanning for symlinks to it) and link them all to /dev/null
8b04b925 864
b44be3ec 865* timer units:
b44be3ec 866 - timer units should get the ability to trigger when:
b44be3ec 867 o DST changes
90e071d1 868 - Modulate timer frequency based on battery state
d2e83c23 869
a19554ed
LP
870* add libsystemd-password or so to query passwords during boot using the password agent logic
871
a7a3f28b
LP
872* clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
873
b44be3ec 874* on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
08f95888 875
f59d1da8 876* make repeated alt-ctrl-del presses printing a dump
dcfc4b2e 877
71ef24d0
LP
878* hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
879
d28315e4 880* currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
1b89884b 881
b44be3ec 882* nspawn:
e25b5a8d
DH
883 - emulate /dev/kmsg using CUSE and turn off the syslog syscall
884 with seccomp. That should provide us with a useful log buffer that
885 systemd can log to during early boot, and disconnect container logs
886 from the kernel's logs.
887 - as soon as networkd has a bus interface, hook up --network-interface=,
888 --network-bridge= with networkd, to trigger netdev creation should an
889 interface be missing
e25b5a8d
DH
890 - a nice way to boot up without machine id set, so that it is set at boot
891 automatically for supporting --ephemeral. Maybe hash the host machine id
892 together with the machine name to generate the machine id for the container
893 - fix logic always print a final newline on output.
894 https://github.com/systemd/systemd/pull/272#issuecomment-113153176
895 - should optionally support receiving WATCHDOG=1 messages from its payload
896 PID 1...
e25b5a8d
DH
897 - optionally automatically add FORWARD rules to iptables whenever nspawn is
898 running, remove them when shut down.
e25b5a8d 899
347a17d5
ZJS
900* dissect
901 - refuse mounting over a mount point
902 - automatically discover .roothash files in dissect, similarly to nspawn
903
e25b5a8d 904* machined:
e25b5a8d
DH
905 - add an API so that libvirt-lxc can inform us about network interfaces being
906 removed or added to an existing machine
907 - "machinectl migrate" or similar to copy a container from or to a
908 difference host, via ssh
e25b5a8d
DH
909 - introduce systemd-nspawn-ephemeral@.service, and hook it into
910 "machinectl start" with a new --ephemeral switch
911 - "machinectl status" should also show internal logs of the container in
912 question
e25b5a8d
DH
913 - "machinectl history"
914 - "machinectl diff"
915 - "machinectl commit" that takes a writable snapshot of a tree, invokes a
916 shell in it, and marks it read-only after use
917
b44be3ec 918* cryptsetup:
b44be3ec 919 - cryptsetup-generator: allow specification of passwords in crypttab itself
5f1dac6b 920 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
f8019684 921
424a19f8 922* hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
41f9172f 923 https://bugs.freedesktop.org/show_bug.cgi?id=54712
424a19f8 924
83fefc88
ZJS
925* add a dependency on standard-conf.xml and other included files to man pages
926
68f16003 927* MountFlags=shared acts as MountFlags=slave right now.
0790b9fe 928
fd4d89b2
LP
929* properly handle loop back mounts via fstab, especially regards to fsck/passno
930
47ae7201
LP
931* initialize the hostname from the fs label of /, if /etc/hostname does not exist?
932
abd55b16 933* udev:
abd55b16 934 - move to LGPL
abd55b16
KS
935 - kill scsi_id
936 - add trigger --subsystem-match=usb/usb_device device
e8d842a0 937 - reimport udev db after MOVE events for devices without dev_t
b8217b7b 938
7f110ff9
LP
939* There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
940
d0e5a333
LP
941* add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
942
e25b5a8d
DH
943* coredump:
944 - save coredump in Windows/Mozilla minidump format
73a99163 945 - when truncating coredumps, also log the full size that the process had, and make a metadata field so we can report truncated coredumps
87a8baa3
LP
946
947* support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
948
f957632b
KS
949* be able to specify a forced restart of service A where service B depends on, in case B
950 needs to be auto-respawned?
951
f38afcd0 952* tmpfiles:
f38afcd0 953 - apply "x" on "D" too (see patch from William Douglas)
614cc34f 954 - instead of ignoring unknown fields, reject them.
e25b5a8d
DH
955 - creating new directories/subvolumes/fifos/device nodes
956 should not follow symlinks. None of the other adjustment or creation
957 calls follow symlinks.
1258097c 958
a4c279f8
LP
959* make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
960
f28f1daf
LP
961* verify that the AF_UNIX sockets of a service in the fs still exist
962 when we start a service in order to avoid confusion when a user
963 assumes starting a service is enough to make it accessible
964
88a07670
LP
965* Make it possible to set the keymap independently from the font on
966 the kernel cmdline. Right now setting one resets also the other.
967
71092d70 968* and a dbus call to generate target from current state
916abb21 969
68c7d001 970* write blog stories about:
796b06c2 971 - hwdb: what belongs into it, lsusb
68c7d001 972 - enabling dbus services
7a2a0b90 973 - how to make changes to sysctl and sysfs attributes
253ee27a 974 - remote access
5d0fcd7c 975 - how to pass throw-away units to systemd, or dynamically change properties of existing units
8bbabc44 976 - testing with Harald's awesome test kit
f530371f 977 - auto-restart
e707c494 978 - how to develop against journal browsing APIs
a19554ed 979 - the journal HTTP iface
a73d88fa 980 - non-cgroup resource management
a74a8793 981 - dynamic resource management with cgroups
3040728b 982 - refreshed, longer missions statement
3679d112
LP
983 - calendar time events
984 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
985 - how to create your own target
986 - instantiated apache, dovecot and so on
987 - hooking a script into various stages of shutdown/rearly booot
73090dc8 988
21bdae12
LP
989* investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
990
59fee421
LP
991* dot output for --test showing the 'initial transaction'
992
e25b5a8d 993* pid1:
e25b5a8d
DH
994 - When logging about multiple units (stopping BoundTo units, conflicts, etc.),
995 log both units as UNIT=, so that journalctl -u triggers on both.
996 - generate better errors when people try to set transient properties
997 that are not supported...
998 http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html
999 - maybe introduce WantsMountsFor=? Usecase:
1000 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html
1001 - recreate systemd's D-Bus private socket file on SIGUSR2
e25b5a8d
DH
1002 - move PAM code into its own binary
1003 - when we automatically restart a service, ensure we restart its rdeps, too.
e25b5a8d
DH
1004 - hide PAM options in fragment parser when compile time disabled
1005 - Support --test based on current system state
1006 - 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().
1007 - after deserializing sockets in socket.c we should reapply sockopts and things
e25b5a8d
DH
1008 - drop PID 1 reloading, only do reexecing (difficult: Reload()
1009 currently is properly synchronous, Reexec() is weird, because we
1010 cannot delay the response properly until we are back, so instead of
1011 being properly synchronous we just keep open the fd and close it
1012 when done. That means clients do not get a successful method reply,
1013 but much rather a disconnect on success.
1014 - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
1015 - when a bus name of a service disappears from the bus make sure to queue further activation requests
1016
1017* unit files:
1018 - allow port=0 in .socket units
1019 - maybe introduce ExecRestartPre=
1020 - add ReloadSignal= for configuring a reload signal to use
1021 - implement Register= switch in .socket units to enable registration
1022 in Avahi, RPC and other socket registration services.
1023 - allow Type=simple with PIDFile=
1024 https://bugzilla.redhat.com/show_bug.cgi?id=723942
1025 - allow writing multiple conditions in unit files on one line
e25b5a8d 1026 - introduce Type=pid-file
e25b5a8d
DH
1027 - introduce mix of BindTo and Requisite
1028 - add a concept of RemainAfterExit= to scope units
e25b5a8d 1029 - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
e5171296 1030 - add verification of [Install] section to systemd-analyze verify
e25b5a8d 1031
af6f0d42
TG
1032* udev-link-config:
1033 - Make sure ID_PATH is always exported and complete for
1034 network devices where possible, so we can safely rely
1035 on Path= matching
af6f0d42 1036
88e4d1d7 1037* sd-rtnl:
88e4d1d7 1038 - add support for more attribute types
c589a0e6 1039 - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places
88e4d1d7 1040
0a4b9a07 1041* networkd:
c74ecd71
TG
1042 - add more keys to [Route] and [Address] sections
1043 - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
0a4b9a07 1044 - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
e8d842a0 1045 - add reduced [Link] support to .network files
5c1d3fc9 1046 - add Scope= parsing option for [Network]
798e174a 1047 - properly handle routerless dhcp leases
a8eaaee7 1048 - work with non-Ethernet devices
c9fa3670 1049 - add support for more bond options
e25b5a8d 1050 - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from?
e25b5a8d
DH
1051 - the DHCP lease data (such as NTP/DNS) is still made available when
1052 a carrier is lost on a link. It should be removed instantly.
1053 - expose in the API the following bits:
1054 - option 15, domain name and/or option 119, search list
1055 - option 12, host name and/or option 81, fqdn
1056 - option 123, 144, geolocation
1057 - option 252, configure http proxy (PAC/wpad)
1058 - provide a way to define a per-network interface default metric value
1059 for all routes to it. possibly a second default for DHCP routes.
1060 - allow Name= to be specified repeatedly in the [Match] section. Maybe also
1061 support Name=foo*|bar*|baz ?
1062 - duplicate address check for static IPs (like ARPCHECK in network-scripts)
1063 - allow DUID/IAID to be customized, see issue #394.
e25b5a8d 1064 - whenever uplink info changes, make DHCP server send out FORCERENEW
155e8b9a 1065
473dfd7b
TG
1066* networkd-wait-online:
1067 - make operstates to wait for configurable?
8f5675cc 1068
ac976532 1069* dhcp:
424a8732 1070 - figure out how much we can increase Maximum Message Size
ac976532 1071
37d8b536
PF
1072* dhcp6:
1073 - add functions to set previously stored IPv6 addresses on startup and get
1074 them at shutdown; store them in client->ia_na
1075 - write more test cases
37d8b536 1076 - implement reconfigure support, see 5.3., 15.11. and 22.20.
37d8b536 1077 - implement support for temporary adressess (IA_TA)
37d8b536
PF
1078 - implement dhcpv6 authentication
1079 - investigate the usefulness of Confirm messages; i.e. are there any
1080 situations where the link changes without any loss in carrier detection
1081 or interface down
1082 - some servers don't do rapid commit without a filled in IA_NA, verify
1083 this behavior
4a77c53d 1084 - RouteTable= ?
37d8b536 1085
2f8cd170 1086External:
74fe1fe3 1087
cc9784c6 1088* dbus:
3377af3e 1089 - natively watch for dbus-*.service symlinks (PENDING)
b44be3ec 1090 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
cc9784c6 1091
71092d70
LP
1092* fix alsa mixer restore to not print error when no config is stored
1093
74fe1fe3
LP
1094* make cryptsetup lower --iter-time
1095
9096ad36 1096* patch kernel for xattr support in /dev, /proc/, /sys?
06ae4bfe 1097
71092d70 1098* kernel: add device_type = "fb", "fbcon" to class "graphics"
14bd37fe 1099
8973790e
LP
1100* /usr/bin/service should actually show the new command line
1101
d2bbca68 1102* 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=
8973790e
LP
1103
1104* fedora: F20: go timer units all the way, leave cron.daily for cron
1105
1106* neither pkexec nor sudo initialize environ[] from the PAM environment?
b1ce67bf 1107
4641a16b 1108* fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
19aa7c4f 1109
baf167ee
ZJS
1110* register catalog database signature as file magic
1111
1cf3c30c
ZJS
1112* zsh shell completion:
1113 - <command> <verb> -<TAB> should complete options, but currently does not
409886c4 1114 - systemctl add-wants,add-requires
1cf3c30c 1115
90137b62 1116* systemctl status should know about 'systemd-analyze calendar ... --iterations='
90137b62
ZJS
1117* If timer has just OnInactiveSec=..., it should fire after a specified time
1118 after being started.
2c12a402 1119
129126f3 1120Regularly:
874aa2cc 1121
129126f3 1122* look for close() vs. close_nointr() vs. close_nointr_nofail()
5021be21 1123
129126f3 1124* check for strerror(r) instead of strerror(-r)
5b6319dc 1125
444a79d3
LP
1126* pahole
1127
d28315e4 1128* set_put(), hashmap_put() return values check. i.e. == 0 does not free()!
a3a3e5b6 1129
4db17f29 1130* use secure_getenv() instead of getenv() where appropriate
88fae6e0 1131
d05c556b 1132* link up selected blog stories from man pages and unit files Documentation= fields
62353f69 1133String is not UTF-8 clean, ignoring assignment