]> git.ipfire.org Git - thirdparty/systemd.git/blob - TODO
journal: coalesce ftruncate()s in 250ms windows
[thirdparty/systemd.git] / TODO
1 Bugfixes:
2
3 * Should systemctl status \* work on all unit types, not just .service?
4
5 * Dangling symlinks of .automount unit files in .wants/ directories, set up
6 automount points even when the original .automount file did not exist
7 anymore. Only the .mount unit was still around.
8
9 * ExecStart with unicode characters fails in strv_split_extract:
10
11 [Service]
12 Environment=ONE='one' "TWO='two two' too" THREE=
13 ExecStart=/bin/python3 -c 'import sys;print(sys.argv)' $ONE $TWO $THREE
14
15 * When systemctl --host is used, underlying ssh connection can remain open.
16 bus_close does not kill children?
17
18 External:
19
20 * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
21
22 * wiki: update journal format documentation for lz4 additions
23
24 Janitorial Clean-ups:
25
26 * code cleanup: retire FOREACH_WORD_QUOTED, port to extract_first_word() loops instead
27
28 * replace manual readdir() loops with FOREACH_DIRENT or FOREACH_DIRENT_ALL
29
30 * Get rid of the last strerror() invocations in favour of %m and strerror_r()
31
32 * Rearrange tests so that the various test-xyz.c match a specific src/basic/xyz.c again
33
34 Features:
35
36 * PID1: find a way how we can reload unit file configuration for
37 specific units only, without reloading the whole of systemd
38
39 * add an explicit parser for LimitNICE= and LimitRTPRIO= that verifies
40 the specified range and generates sane error messages for incorrect
41 specifications. Also, for LimitNICE= maybe introduce a syntax such
42 as "+5" or "-7" in order to make the limits more readable as they
43 are otherwise shifted by 20.
44
45 * do something about "/control" subcgroups in the unified cgroup hierarchy
46
47 * when we detect that there are waiting jobs but no running jobs, do something
48
49 * push CPUAffinity= also into the "cpuset" cgroup controller (only after the cpuset controller got ported to the unified hierarchy)
50
51 * add a new command "systemctl revert" or so, that removes all dropin
52 snippets in /run and /etc, and all unit files with counterparts in
53 /usr, and thus undoes what "systemctl set-property" and "systemctl
54 edit" create. Maybe even add "systemctl revert -a" to do this for
55 all units.
56
57 * PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn)
58
59 * consider throwing a warning if a service declares it wants to be "Before=" a .device unit.
60
61 * "systemctl edit" should know a mode to create a new unit file
62
63 * there's probably something wrong with having user mounts below /sys,
64 as we have for debugfs. for exmaple, src/core/mount.c handles mounts
65 prefixed with /sys generally special.
66 http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html
67
68 * man: document that unless you use StandardError=null the shell >/dev/stderr won't work in shell scripts in services
69
70 * "systemctl daemon-reload" should result in /etc/systemd/system.conf being reloaded by systemd
71
72 * install: include generator dirs in unit file search paths
73
74 * rework C11 utf8.[ch] to use char32_t instead of uint32_t when referring
75 to unicode chars, to make things more expressive.
76
77 * fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline
78
79 * docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date
80
81 * mounting and unmounting mount points manually with different source
82 devices will result in collected collected on all devices used.
83 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030225.html
84
85 * add a job mode that will fail if a transaction would mean stopping
86 running units. Use this in timedated to manage the NTP service
87 state.
88 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html
89
90 * Maybe add support for the equivalent of "ethtool advertise" to .link files?
91 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030112.html
92
93 * The udev blkid built-in should expose a property that reflects
94 whether media was sensed in USB CF/SD card readers. This should then
95 be used to control SYSTEMD_READY=1/0 so that USB card readers aren't
96 picked up by systemd unless they contain a medium. This would mirror
97 the behaviour we already have for CD drives.
98
99 * networkd/udev: implement SR_IOV configuration in .link files:
100 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html
101
102 * Rework systemctl's GetAll property parsing to use the generic bus_map_all_properties() API
103
104 * implement a per-service firewall based on net_cls
105
106 * Port various tools to make use of verbs.[ch], where applicable
107
108 * hostnamectl: show root image uuid
109
110 * sysfs set api in libudev is not const
111
112 * Find a solution for SMACK capabilities stuff:
113 http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html
114
115 * "systemctl preset-all" should probably order the unit files it
116 operates on lexicographically before starting to work, in order to
117 ensure deterministic behaviour if two unit files conflict (like DMs
118 do, for example)
119
120 * synchronize console access with BSD locks:
121 http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html
122
123 * as soon as we have kdbus, and sender timestamps, revisit coalescing multiple parallel daemon reloads:
124 http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
125
126 * the install state probably shouldn't get confused by generated units, think dbus1/kdbus compat!
127
128 * in systemctl list-unit-files: show the install value the presets would suggest for a service in a third column
129
130 * figure out when we can use the coarse timers
131
132 * add "systemctl start -v foobar.service" that shows logs of a service
133 while the start command runs. This is non-trivial to do without
134 races though, since we should flush out all journal messages before
135 returning from the "systemctl stop".
136
137 * 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
138
139 * add infrastructure to allocate dynamic/transient users and UID ranges, for use in user-namespaced containers, per-seat gdm login screens and gdm guest sessions
140
141 * maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment=
142
143 * introduce systemd-timesync-wait.service or so to sync on an NTP fix?
144
145 * systemd --user should issue sd_notify() upon reaching basic.target, not on becoming idle
146
147 * consider showing the unit names during boot up in the status output, not just the unit descriptions
148
149 * maybe allow timer units with an empty Units= setting, so that they
150 can be used for resuming the system but nothing else.
151
152 * what to do about udev db binary stability for apps? (raw access is not an option)
153
154 * maybe provide an API to allow migration of foreign PIDs into existing scopes.
155
156 * man: maybe use the word "inspect" rather than "introspect"?
157
158 * systemctl: if some operation fails, show log output?
159
160 * systemctl edit:
161 - allow creation of units from scratch
162 - use equvalent of cat() to insert existing config as a comment, prepended with #.
163 Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc.
164
165 * exponential backoff in timesyncd and resolved when we cannot reach a server
166
167 * timesyncd + resolved: add ugly bus calls to set NTP and DNS servers per-interface, for usage by NM
168
169 * extract_many_words() should probably be used by a lot of code that
170 currently uses FOREACH_WORD and friends. For example, most conf
171 parsing callbacks should use it.
172
173 * merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share....
174
175 * systemd.show_status= should probably have a mode where only failed
176 units are shown.
177
178 * add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL
179 (throughout the codebase, not only PID1)
180
181 * resolved:
182 - put networkd events and rtnl events at a higher priority, so that
183 we always process them before we process client requests
184 - DNSSEC
185 - add display of private key types (http://tools.ietf.org/html/rfc4034#appendix-A.1.1)?
186 - synthesize negative cache entries from NSEC/NSEC3 and drop explicit negative caching of authenticated answers
187 - mDNS/DNS-SD
188 - mDNS RR resolving
189 - service registration
190 - service/domain/types browsing
191 - avahi compat
192 - DNS-SD service registration from socket units
193 - resolved should optionally register additional per-interface LLMNR
194 names, so that for the container case we can establish the same name
195 (maybe "host") for referencing the server, everywhere.
196 - add API so NM can push DNS server info into resolved
197
198 * refcounting in sd-resolve is borked
199
200 * Add a new verb "systemctl top"
201
202 * add new gpt type for btrfs volumes
203
204 * support empty /etc boots nicely:
205 - nspawn/gpt-generator: introduce new gpt partition type for /usr
206 - fstab-generator: support systemd.volatile=yes|no|state on the kernel cmdline, too, similar to nspawn's --volatile=
207
208 * generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
209
210 * a way for container managers to turn off getty starting via $container_headless= or so...
211
212 * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
213
214 * mount_cgroup_controllers(): symlinks need to get the label applied
215
216 * For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
217 they run added to the initial transaction and thus confuse Type=idle.
218
219 * Run most system services with cgroupfs read-only and procfs with a more secure mode (doesn't work, since the hidepid= option is per-pid-namespace, not per-mount)
220
221 * add bus api to query unit file's X fields.
222
223 * gpt-auto-generator:
224 - Support LUKS for root devices
225 - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap?
226 - Make /home automount rather than mount?
227
228 * add generator that pulls in systemd-network from containers when
229 CAP_NET_ADMIN is set, more than the loopback device is defined, even
230 when it is otherwise off
231
232 * MessageQueueMessageSize= and RLimitFSIZE= (and suchlike) should use parse_iec_size().
233
234 * "busctl status" works only as root on dbus1, since we cannot read
235 /proc/$PID/exe
236
237 * implement Distribute= in socket units to allow running multiple
238 service instances processing the listening socket, and open this up
239 for ReusePort=
240
241 * socket units: support creating sockets in different namespace,
242 opening it up for JoinsNamespaceOf=. This would require to fork off
243 a tiny process that joins the namespace and creates/binds the socket
244 and passes this back to PID1 via SCM_RIGHTS. This also could be used
245 to allow Chown/chgrp on sockets without requiring NSS in PID 1.
246
247 * New service property: maximum CPU and wallclock runtime for a service
248
249 * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze
250 $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls
251 should SIGSTOP all unit processes in a loop until all processes of
252 it are fully stopped. This can later be used for app management by
253 desktop UIs such as gnome-shell to freeze apps that are not visible
254 on screen, not unlike how job control works on the shell
255
256 * cgroups:
257 - implement per-slice CPUFairScheduling=1 switch
258 - handle jointly mounted controllers correctly
259 - introduce high-level settings for RT budget, swappiness
260 - how to reset dynamically changed unit cgroup attributes sanely?
261 - when reloading configuration, apply new cgroup configuration
262 - when recursively showing the cgroup hierarchy, optionally also show
263 the hierarchies of child processes
264
265 * transient units:
266 - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
267 - ensure scope units may be started only a single time
268
269 * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk?
270
271 * when we detect low battery and no AC on boot, show pretty splash and refuse boot
272
273 * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
274
275 * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
276
277 * unify dispatch table in systemctl_main() and friends
278
279 * 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
280
281 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
282
283 * If we try to find a unit via a dangling symlink, generate a clean
284 error. Currently, we just ignore it and read the unit from the search
285 path anyway.
286
287 * refuse boot if /etc/os-release is missing or /etc/machine-id cannot be set up
288
289 * btrfs raid assembly: some .device jobs stay stuck in the queue
290
291 * make sure gdm does not use multi-user-x but the new default X configuration file, and then remove multi-user-x from systemd
292
293 * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
294
295 * load .d/*.conf dropins for device units
296
297 * allow implementation of InaccessibleDirectories=/ plus
298 ReadOnlyDirectories=... for whitelisting files for a service.
299
300 * sd-bus:
301 - EBADSLT handling
302 - GetAllProperties() on a non-existing object does not result in a failure currently
303 - kdbus: process fd=-1 for incoming msgs
304 - port to sd-resolve for connecting to TCP dbus servers
305 - kdbus: maybe add controlling tty metadata fields
306 - 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
307 - when kdbus does not take our message without memfds, try again with memfds
308 - see if we can drop more message validation on the sending side
309 - add API to clone sd_bus_message objects
310 - make AddMatch calls on dbus1 transports async?
311 - kdbus: matches against source or destination pids for an "strace -p"-like feel. Problem: The PID info needs to be available in userspace too...
312 - longer term: priority inheritance
313 - dbus spec updates:
314 - kdbus mapping
315 - NameLost/NameAcquired obsolete
316 - GVariant
317 - path escaping
318 - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
319 - test bloom filter generation indexes
320 - bus-proxy: when passing messages from kdbus, make sure we properly
321 handle the case where a large number of fds is appended that we
322 cannot pass into sendmsg() of the AF_UNIX sokcet (which only accepts
323 253 messages)
324 - kdbus: introduce a concept of "send-only" connections
325 - kdbus: add counter for refused unicast messages that is passed out via the RECV ioctl. SImilar to the counter for dropped multicast messages we already have.
326
327 * sd-event
328 - allow multiple signal handlers per signal?
329 - document chaining of signal handler for SIGCHLD and child handlers
330 - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ...
331 - generate a failure of a default event loop is executed out-of-thread
332 - maybe add support for inotify events
333
334 * in the final killing spree, detect processes from the root directory, and
335 complain loudly if they have argv[0][0] == '@' set.
336 https://bugzilla.redhat.com/show_bug.cgi?id=961044
337
338 * investigate endianness issues of UUID vs. GUID
339
340 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
341 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
342
343 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
344
345 * add a pam module that on password changes updates any LUKS slot where the password matches
346
347 * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases...
348
349 * cgtop: make cgtop useful in a container
350
351 * test/:
352 - add 'set -e' to scripts in test/
353 - make stuff in test/ work with separate output dir
354
355 * seems that when we follow symlinks to units we prefer the symlink
356 destination path over /etc and /usr. We should not do that. Instead
357 /etc should always override /run+/usr and also any symlink
358 destination.
359
360 * when isolating, try to figure out a way how we implicitly can order
361 all units we stop before the isolating unit...
362
363 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
364
365 * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify
366 that the boot loader should be installed to the ESP. Define a way
367 how an installer can figure out whether a BLS compliant boot loader
368 is installed.
369
370 * think about requeuing jobs when daemon-reload is issued? usecase:
371 the initrd issues a reload after fstab from the host is accessible
372 and we might want to requeue the mounts local-fs acquired through
373 that automatically.
374
375 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
376
377 * remove any syslog support from log.c -- we probably cannot do this before split-off udev is gone for good
378
379 * shutdown logging: store to EFI var, and store to USB stick?
380
381 * think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service
382
383 * merge unit_kill_common() and unit_kill_context()
384
385 * introduce ExecCondition= in services
386
387 * EFI:
388 - honor language efi variables for default language selection (if there are any?)
389 - honor timezone efi variables for default timezone selection (if there are any?)
390 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
391
392 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
393
394 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
395
396 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
397
398 * logind:
399 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
400 - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
401 - logind: wakelock/opportunistic suspend support
402 - Add pretty name for seats in logind
403 - logind: allow showing logout dialog from system?
404 - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case
405 - session scopes/user unit: add RequiresMountsFor for the home directory of the user
406 - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
407 - if pam_systemd is invoked by su from a process that is outside of a
408 any session we should probably just become a NOP, since that's
409 usually not a real user session but just some system code that just
410 needs setuid().
411 - logind: make the Suspend()/Hibernate() bus calls wait for the for
412 the job to be completed. before returning, so that clients can wait
413 for "systemctl suspend" to finish to know when the suspending is
414 complete.
415 - logind: when the power button is pressed short, just popup a
416 logout dialog. If it is pressed for 1s, do the usual
417 shutdown. Inspiration are Macs here.
418 - expose "Locked" property on logind sesison objects
419 - given that logind now lets PID 1 do all nasty work, we can
420 probably reduce the capability set it retains substantially.
421 (we need CAP_SYS_ADMIN for drmSetMaster(), so maybe not worth it)
422 - expose orientation sensors and tablet mode through logind
423 - maybe allow configuration of the StopTimeout for session scopes
424 - rename session scope so that it includes the UID. THat way
425 the session scope can be arranged freely in slices and we don't have
426 make assumptions about their slice anymore.
427 - follow PropertiesChanged state more closely, to deal with quick logouts and
428 relogins
429
430 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
431
432 * service: watchdog logic: for testing purposes allow ping, but do not require pong
433
434 * journal:
435 - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields
436 - import and delete pstore filesystem content at startup
437 - journald: also get thread ID from client, plus thread name
438 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
439 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
440 - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively?
441 - declare the local journal protocol stable in the wiki interface chart
442 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
443 - journald: when dropping msgs due to ratelimit make sure to write
444 "dropped %u messages" not only when we are about to print the next
445 message that works, but alraedy after a short tiemout
446 - check if we can make journalctl by default use --follow mode inside of less if called without args?
447 - maybe add API to send pairs of iovecs via sd_journal_send
448 - journal: when writing journal auto-rotate if time jumps backwards
449 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
450 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
451 and !FOOBAR for events without FOOBAR.
452 - journal: store timestamp of journal_file_set_offline() int he header,
453 so it is possible to display when the file was last synced.
454 - 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.
455 - journal: find a way to allow dropping history early, based on priority, other rules
456 - journal: When used on NFS, check payload hashes
457 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
458 - refuse taking lower-case variable names in sd_journal_send() and friends.
459 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
460 - journal: deal nicely with byte-by-byte copied files, especially regards header
461 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
462 - Replace utmp, wtmp, btmp, and lastlog completely with journal
463 - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
464 - when a kernel driver logs in a tight loop, we should ratelimit that too.
465 - journald: optionally, log debug messages to /run but everything else to /var
466 - journald: when we drop syslog messages because the syslog socket is
467 full, make sure to write how many messages are lost as first thing
468 to syslog when it works again.
469 - journald: make sure ratelimit is actually really per-service with the new cgroup changes
470 - change systemd-journal-flush into a service that stays around during
471 boot, and causes the journal to be moved back to /run on shutdown,
472 so that we do not keep /var busy. This needs to happen synchronously,
473 hence doing this via signals is not going to work.
474 - optionally support running journald from the command line for testing purposes in external projects
475 - journald: allow per-priority and per-service retention times when rotating/vacuuming
476 - journald: make use of uid-range.h to managed uid ranges to split
477 journals in.
478 - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
479 - improve journalctl performance by loading journal files
480 lazily. Encode just enough information in the file name, so that we
481 do not have to open it to know that it is not interesting for us, for
482 the most common operations.
483 - journal-or-kmsg is currently broken? See reverted
484 commit 4a01181e460686d8b4a543b1dfa7f77c9e3c5ab8.
485 - man: document that corrupted journal files is nothing to act on
486 - systemd-journal-upload (or a new, related tool): allow pushing out
487 journal messages onto the network in BSD syslog protocol,
488 continuously. Default to some link-local IP mcast group, to make this
489 useful as a one-stop debugging tool.
490 - rework journald sigbus stuff to use mutex
491 - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our
492 services that run under their own user ids, and use User= (but only
493 in a world where userns is ubiquitous since otherwise we cannot
494 invoke those daemons on the host AND in a container anymore). Also,
495 if LimitNPROC= is used without User= we should warn and refuse
496 operation.
497 - journalctl --verify: don't show files that are currently being
498 written to as FAIL, but instead show that their are being written to.
499 - add journalctl -H that talks via ssh to a remote peer and passes through
500 binary logs data
501 - change journalctl -M to acquire fd to journal directory via machined, and
502 then operate on that via openat() instead of absolute paths
503 - add a version of --merge which also merges /var/log/journal/remote
504 - log accumulated resource usage after each service invocation
505 - journalctl: -m should access container journals directly by enumerating
506 them via machined, and also watch containers coming and going.
507 Benefit: nspawn --ephemeral would start working nicely with the journal.
508 - assign MESSAGE_ID to log messages about failed services
509
510 * document:
511 - document that deps in [Unit] sections ignore Alias= fields in
512 [Install] units of other units, unless those units are disabled
513 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
514 - document the exit codes when services fail before they are exec()ed
515 - document that service reload may be implemented as service reexec
516 - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications
517 - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
518 - document systemd-journal-flush.service properly
519 - documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
520 - man: document the very specific env the shutdown drop-in tools live in
521 - man: add more examples to man pages
522 - man: maybe sort directives in man pages, and take sections from --help and apply them to man too
523
524 * systemctl:
525 - systemctl list-jobs - show dependencies
526 - add systemctl switch to dump transaction without executing it
527 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
528 - "systemctl disable" on a static unit prints no message and does
529 nothing. "systemctl enable" does nothing, and gives a bad message
530 about it. Should fix both to print nice actionable messages.
531 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
532 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
533 - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
534 - systemctl: "Journal has been rotated since unit was started." message is misleading
535 - better error message if you run systemctl without systemd running
536 - systemctl status output should should include list of triggering units and their status
537
538 * unit install:
539 - "systemctl mask" should find all names by which a unit is accessible
540 (i.e. by scanning for symlinks to it) and link them all to /dev/null
541 - systemctl list-unit-files should list generated files (and probably with a new state "generated" for them, or so)
542
543 * timer units:
544 - timer units should get the ability to trigger when:
545 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
546 o DST changes
547 - Support 2012-02~4 as syntax for specifying the fourth to last day of the month.
548 - calendarspec: support value ranges with ".." notation. Example: 2013-4..8-1
549 - Modulate timer frequency based on battery state
550
551 * add libsystemd-password or so to query passwords during boot using the password agent logic
552
553 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
554
555 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
556
557 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
558 waiting for the timeout
559
560 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
561
562 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
563
564 * nspawn:
565 - to allow "linking" of nspawn containers, extend --network-bridge= so
566 that it can dynamically create bridge interfaces that are refcounted
567 by the containers on them. For each group of containers to link together
568 - refuses to boot containers without /etc/machine-id (OK?), and with empty
569 /etc/machine-id (not OK).
570 - nspawn -x should support ephemeral instances of gpt images
571 - emulate /dev/kmsg using CUSE and turn off the syslog syscall
572 with seccomp. That should provide us with a useful log buffer that
573 systemd can log to during early boot, and disconnect container logs
574 from the kernel's logs.
575 - as soon as networkd has a bus interface, hook up --network-interface=,
576 --network-bridge= with networkd, to trigger netdev creation should an
577 interface be missing
578 - don't copy /etc/resolv.conf from host into container unless we are in
579 shared-network mode
580 - a nice way to boot up without machine id set, so that it is set at boot
581 automatically for supporting --ephemeral. Maybe hash the host machine id
582 together with the machine name to generate the machine id for the container
583 - fix logic always print a final newline on output.
584 https://github.com/systemd/systemd/pull/272#issuecomment-113153176
585 - should optionally support receiving WATCHDOG=1 messages from its payload
586 PID 1...
587 - should send out sd_notify("WATCHDOG=1") messages
588 - optionally automatically add FORWARD rules to iptables whenever nspawn is
589 running, remove them when shut down.
590 - add a logic for cleaning up read-only, hidden container images in
591 /var/lib/machines that are not ancestors of any non-hidden containers
592 - Improve error message when --bind= is used on a non-existing source
593 directory
594 - maybe make copying of /etc/resolv.conf optional, and skip it if --read-only
595 is used
596
597 * machined:
598 - "machinectl list" should probably show columns for OS version and IP
599 addresses
600 - add an API so that libvirt-lxc can inform us about network interfaces being
601 removed or added to an existing machine
602 - "machinectl migrate" or similar to copy a container from or to a
603 difference host, via ssh
604 - introduce systemd-nspawn-ephemeral@.service, and hook it into
605 "machinectl start" with a new --ephemeral switch
606 - "machinectl status" should also show internal logs of the container in
607 question
608 - "machinectl list-images" should show os-release data, as well as
609 machine-info data (including deployment level)
610 - "machinectl history"
611 - "machinectl diff"
612 - "machinectl commit" that takes a writable snapshot of a tree, invokes a
613 shell in it, and marks it read-only after use
614
615 * importd:
616 - generate a nice warning if mkfs.btrfs is missing
617
618 * cryptsetup:
619 - cryptsetup-generator: allow specification of passwords in crypttab itself
620 - move cryptsetup key caching into kernel keyctl?
621 https://bugs.freedesktop.org/show_bug.cgi?id=54982
622 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
623
624 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
625 https://bugs.freedesktop.org/show_bug.cgi?id=54712
626
627 * create /sbin/init symlinks from the build system
628
629 * MountFlags=shared acts as MountFlags=slave right now.
630
631 * properly handle loop back mounts via fstab, especially regards to fsck/passno
632
633 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
634
635 * rename "userspace" to "core-os"
636
637 * udev:
638 - move to LGPL
639 - kill scsi_id
640 - add trigger --subsystem-match=usb/usb_device device
641 - reimport udev db after MOVE events for devices without dev_t
642
643 * when a service has the same env var set twice we actually store it twice and return that in systemctl show -p... We should only show the last setting
644
645 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
646
647 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
648
649 * coredump:
650 - save coredump in Windows/Mozilla minidump format
651 - move PID 1 segfaults to /var/lib/systemd/coredump?
652 - make the handler check /proc/$PID/rlimits for RLIMIT_CORE,
653 and supress coredump if turned off. Then change RLIMIT_CORE to
654 infinity by default for all services. This then allows per-service
655 control of coredumping.
656
657 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
658
659 * default to actual 32-bit PIDs, via /proc/sys/kernel/pid_max
660
661 * be able to specify a forced restart of service A where service B depends on, in case B
662 needs to be auto-respawned?
663
664 * tmpfiles:
665 - apply "x" on "D" too (see patch from William Douglas)
666 - replace F with f+.
667 - instead of ignoring unknown fields, reject them.
668 - creating new directories/subvolumes/fifos/device nodes
669 should not follow symlinks. None of the other adjustment or creation
670 calls follow symlinks.
671
672 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
673
674 * verify that the AF_UNIX sockets of a service in the fs still exist
675 when we start a service in order to avoid confusion when a user
676 assumes starting a service is enough to make it accessible
677
678 * Make it possible to set the keymap independently from the font on
679 the kernel cmdline. Right now setting one resets also the other.
680
681 * and a dbus call to generate target from current state
682
683 * write blog stories about:
684 - hwdb: what belongs into it, lsusb
685 - enabling dbus services
686 - status update
687 - how to make changes to sysctl and sysfs attributes
688 - remote access
689 - how to pass throw-away units to systemd, or dynamically change properties of existing units
690 - testing with Harald's awesome test kit
691 - auto-restart
692 - how to develop against journal browsing APIs
693 - the journal HTTP iface
694 - non-cgroup resource management
695 - dynamic resource management with cgroups
696 - refreshed, longer missions statement
697 - calendar time events
698 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
699 - how to create your own target
700 - instantiated apache, dovecot and so on
701 - hooking a script into various stages of shutdown/rearly booot
702
703 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
704
705 * dot output for --test showing the 'initial transaction'
706
707 * fingerprint.target, wireless.target, gps.target, netdevice.target
708
709 * pid1:
710 - .timer units should optionally support CLOCK_BOOTTIME in addition to CLOCK_MONOTONIC
711 - When logging about multiple units (stopping BoundTo units, conflicts, etc.),
712 log both units as UNIT=, so that journalctl -u triggers on both.
713 - generate better errors when people try to set transient properties
714 that are not supported...
715 http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html
716 - maybe introduce WantsMountsFor=? Usecase:
717 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html
718 - recreate systemd's D-Bus private socket file on SIGUSR2
719 - GC unreferenced jobs (such as .device jobs)
720 - move PAM code into its own binary
721 - when we automatically restart a service, ensure we restart its rdeps, too.
722 - for services: do not set $HOME in services unless requested
723 - hide PAM options in fragment parser when compile time disabled
724 - Support --test based on current system state
725 - 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().
726 - after deserializing sockets in socket.c we should reapply sockopts and things
727 - drop PID 1 reloading, only do reexecing (difficult: Reload()
728 currently is properly synchronous, Reexec() is weird, because we
729 cannot delay the response properly until we are back, so instead of
730 being properly synchronous we just keep open the fd and close it
731 when done. That means clients do not get a successful method reply,
732 but much rather a disconnect on success.
733 - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
734 - when a bus name of a service disappears from the bus make sure to queue further activation requests
735
736 * unit files:
737 - allow port=0 in .socket units
738 - maybe introduce ExecRestartPre=
739 - add ReloadSignal= for configuring a reload signal to use
740 - implement Register= switch in .socket units to enable registration
741 in Avahi, RPC and other socket registration services.
742 - allow Type=simple with PIDFile=
743 https://bugzilla.redhat.com/show_bug.cgi?id=723942
744 - allow writing multiple conditions in unit files on one line
745 - load-fragment: when loading a unit file via a chain of symlinks
746 verify that it is not masked via any of the names traversed.
747 - introduce Type=pid-file
748 - ExecOnFailure=/usr/bin/foo
749 - introduce mix of BindTo and Requisite
750 - add a concept of RemainAfterExit= to scope units
751 - Set NoNewPrivileges= on all of our own services, where that makes sense
752 - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
753 - consider adding RuntimeDirectoryUser= + RuntimeDirectoryGroup=
754
755 * systemd-python:
756 - figure out a simple way to wait for journal events in a way that
757 works with ^C
758 - add documentation to systemd.daemon
759
760 * bootchart:
761 - plot per-process IO utilization
762 - group processes based on service association (cgroups)
763 - document initcall_debug
764 - kernel cmdline "bootchart" option for simplicity?
765
766 * udev-link-config:
767 - Make sure ID_PATH is always exported and complete for
768 network devices where possible, so we can safely rely
769 on Path= matching
770 - check MTUBytes parsing (expecting size_t but we are using unsigned)
771
772 * sd-rtnl:
773 - add support for more attribute types
774 - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places
775
776 * networkd:
777 - add more keys to [Route] and [Address] sections
778 - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
779 - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
780 - add reduced [Link] support to .network files
781 - add Scope= parsing option for [Network]
782 - properly handle routerless dhcp leases
783 - add more attribute support for SIT tunnel
784 - work with non-Ethernet devices
785 - add support for more bond options
786 - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from?
787 - add LLDP client side support
788 - the DHCP lease data (such as NTP/DNS) is still made available when
789 a carrier is lost on a link. It should be removed instantly.
790 - expose in the API the following bits:
791 - option 15, domain name and/or option 119, search list
792 - option 12, host name and/or option 81, fqdn
793 - option 123, 144, geolocation
794 - option 252, configure http proxy (PAC/wpad)
795 - provide a way to define a per-network interface default metric value
796 for all routes to it. possibly a second default for DHCP routes.
797 - allow Name= to be specified repeatedly in the [Match] section. Maybe also
798 support Name=foo*|bar*|baz ?
799 - duplicate address check for static IPs (like ARPCHECK in network-scripts)
800 - allow DUID/IAID to be customized, see issue #394.
801 - support configuration option for TSO (tcp segmentation offload)
802 - whenever uplink info changes, make DHCP server send out FORCERENEW
803
804 * networkd-wait-online:
805 - make operstates to wait for configurable?
806
807 * dhcp:
808 - figure out how much we can increase Maximum Message Size
809 - support RFC4702 (pass FQDN)
810
811 * dhcp6:
812 - add functions to set previously stored IPv6 addresses on startup and get
813 them at shutdown; store them in client->ia_na
814 - write more test cases
815 - implement reconfigure support, see 5.3., 15.11. and 22.20.
816 - implement support for temporary adressess (IA_TA)
817 - implement dhcpv6 authentication
818 - investigate the usefulness of Confirm messages; i.e. are there any
819 situations where the link changes without any loss in carrier detection
820 or interface down
821 - some servers don't do rapid commit without a filled in IA_NA, verify
822 this behavior
823
824 External:
825
826 * dbus:
827 - natively watch for dbus-*.service symlinks (PENDING)
828 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
829
830 * fix alsa mixer restore to not print error when no config is stored
831
832 * make cryptsetup lower --iter-time
833
834 * patch kernel for xattr support in /dev, /proc/, /sys?
835
836 * kernel: add device_type = "fb", "fbcon" to class "graphics"
837
838 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
839
840 * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
841
842 * /usr/bin/service should actually show the new command line
843
844 * 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=
845
846 * fedora: F20: go timer units all the way, leave cron.daily for cron
847
848 * neither pkexec nor sudo initialize environ[] from the PAM environment?
849
850 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
851
852 * register catalog database signature as file magic
853
854 * zsh shell completion:
855 - <command> <verb> -<TAB> should complete options, but currently does not
856 - systemctl add-wants,add-requires
857
858
859 Regularly:
860
861 * look for close() vs. close_nointr() vs. close_nointr_nofail()
862
863 * check for strerror(r) instead of strerror(-r)
864
865 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
866
867 * pahole
868
869 * set_put(), hashmap_put() return values check. i.e. == 0 does not free()!
870
871 * use secure_getenv() instead of getenv() where appropriate
872
873 * link up selected blog stories from man pages and unit files Documentation= fields