]> git.ipfire.org Git - thirdparty/systemd.git/blob - TODO
journalctl: highlight MESSAGE= in verbose output
[thirdparty/systemd.git] / TODO
1 Bugfixes:
2 * check systemd-tmpfiles for selinux context hookup for mknod(), symlink() and similar
3
4 * swap units that are activated by one name but shown in the kernel under another are semi-broken
5
6 * Dangling symlinks of .automount unit files in .wants/ directories, set up
7 automount points even when the original .automount file did not exist
8 anymore. Only the .mount unit was still around.
9
10 * make polkit checks async
11
12 * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point.
13
14 Fedora 19:
15
16 * external: maybe it is time to patch procps so that "ps" links to
17 libsystemd-logind to print a pretty service name, seat name, session
18 name in its output. Currently it only shows cgroup membership, but
19 that's sometimes kinda hard to parse for a human.
20
21 * cgroup attrs:
22 - update dbus interface docs in wiki
23
24 * localed:
25 - localectl: support new converted x11→console keymaps
26
27 * when installing fedora with yum --installroot /var/run is a directory, not a symlink
28
29 Features:
30
31 * journald: make sure ratelimit is actually really per-service with the new cgroup changes
32
33 * when creating a session or machine, automatically move the process into the root cgroup for all other hierarchies
34
35 * maybe reintroduce nspawn -C?
36
37 * move systemctl dump to systemd-analyze
38
39 * libsystemd-logind: sd_session_is_active() and friends: verify
40 validity of session name before appending it to a path
41
42 * logind: when a PAM client calls ReleaseSession() start a timeout and
43 kill the session entirely after that is reached.
44
45 * gparted needs to disable auto-activation of mount units somehow, or
46 maybe we should stop doing auto-activiation of this after boot
47 entirely. https://bugzilla.gnome.org/show_bug.cgi?id=701676
48
49 * when a service changes state make reflect that in the
50 RUNNING/LISTENING states of its socket
51
52 * slices:
53 - add option to pam_systemd to move login session into a slice (?)
54 - remove ControlGroup= setting
55 - in sd_pid_get_owner_uid() fallback to query session file
56 - add api to determine slice of unit
57
58 * when recursively showing the cgroup hierarchy, optionally also show
59 the hierarchies of child processes
60
61 * document logic of auto/noauto and fail/nofail in fstab in systemd.mount or systemd-fstab-generator man page
62
63 * we should properly escape ssh hostnames we add into dbus server strings
64
65 * something pulls in pcre as so dep into our daemons such as hostnamed.
66
67 * cgroup-agent: downgrade error messages
68
69 * document systemd-journal-flush.service properly
70
71 * chane systemd-journal-flush into a service that stays around during
72 boot, and causes the journal to be moved back to /run on shutdown,
73 so that we don't keep /var busy. This needs to happen synchronously,
74 hence doing this via signals is not going to work.
75
76 * allow implementation of InaccessibleDirectories=/ plus
77 ReadOnlyDirectories=... for whitelisting files for a service.
78
79 * libsystemd-journal:
80 - return ECHILD as soon as somebody tries to reuse a journal object across a fork()
81
82 * libsystemd-bus:
83 - default policy (allow uid == 0 and our own uid)
84 - enforce alignment of pointers passed in
85 - when kdbus doesn't take our message without memfds, try again with memfds
86 - implement translator service
87 - port systemd to new library
88 - implement busname unit type in systemd
89 - move to gvariant
90 - merge busctl into systemctl or so?
91 - synthesize sd_bus_message objects from kernel messages
92 - properly implement name registry ioctls for kdbus
93 - get rid of object hash table, use decision tree everyhwere instead?
94 - implement monitor logic
95 - object vtable logic
96 - longer term:
97 * priority queues
98 * priority inheritance
99
100 * in the final killing spree, detect processes from the root directory, and
101 complain loudly if they have argv[0][0] == '@' set.
102 https://bugzilla.redhat.com/show_bug.cgi?id=961044
103
104 * add an option to nspawn that uses seccomp to make socket(AF_NETLINK,
105 SOCK_RAW, NETLINK_AUDIT) fail the the appropriate error code that
106 makes the audit userspace to think auditing is not available in the
107 kernel.
108
109 * Introduce a way how we can kill the main process of a service with KillSignal, but all processes with SIGKILL later on
110 https://bugzilla.redhat.com/show_bug.cgi?id=952634
111
112 * maybe add a warning to the unit file parses whern the acces mode of unit files is non-sensical.
113
114 * investigate endianess issues of UUID vs. GUID
115
116 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
117 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
118
119 * if pam_systemd is invoked by su from a process that is outside of a
120 any session we should probably just become a NOP, since that's
121 usually not a real user session but just some system code that just
122 needs setuid().
123
124 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
125
126 * add a pam module that on password changes updates any LUKS slot where the password matches
127
128 * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases...
129
130 * timedatctl, localectl: possibly make some commands work without the daemon, for chroot situations...
131
132 * logind: add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
133
134 * we need dynamic units
135
136 * cgtop: make cgtop useful in a container
137
138 * test/:
139 - add 'set -e' to scripts in test/
140 - make stuff in test/ work with separate output dir
141 - remove all the duplicated code in test/
142
143 * suppress log output on shutdown when "quiet" is used
144
145 * systemctl delete x.snapshot leaves no trace in logs (at least at default level).
146
147 * make the coredump collector tool move itself into the user's cgroup
148 so that the coredump is properly written to the user's own journal
149 file.
150
151 * move /usr/lib/modules/$(uname -r)/modules.devname parsing from udevd to
152 kmod static-nodes
153 call kmod as an early service, and drop CAP_MKNOD from udevd.service
154
155 * seems that when we follow symlinks to units we prefer the symlink
156 destination path over /etc and /usr. We shouldn't do that. Instead
157 /etc should always override /run+/usr and also any symlink
158 destination.
159
160 * remove duplicate default deps logic from fstab-generator vs. mount.c
161
162 * when isolating, try to figure out a way how we implicitly can order
163 all units we stop before the isolating unit...
164
165 * teach udev + logind's uaccess to somehow handle the "dead" device nodes from:
166 /lib/modules/$(uname -r)/modules.devname
167 and apply ACLs to them if they have TAG=="uaccess" in udev rules.
168
169 * add ConditionArchitecture= or so
170
171 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
172
173 * we should log capabilities too
174
175 * Support SO_REUSEPORT with socket activation:
176 - Let systemd maintain a pool of servers.
177 - Use for seamless upgrades, by running the new server before stopping the
178 old.
179
180 * after all byte-wise realloc() is slow, even on glibc, so i guess we
181 need manual exponential loops after all
182
183 * BootLoaderSpec: drop allowing ext234 for $BOOT. Clarify that the
184 kernel has to be in $BOOT. Clarify that the boot loader should be
185 installed to the ESP. Define a way how an installer can figure out
186 whether a BLS compliant boot loader is installed.
187
188 * man: remove .include documentation, and instead push people to use .d/*.conf
189
190 * think about requeuing jobs when daemon-reload is issued? usecase:
191 the initrd issues a reload after fstab from the host is accessible
192 and we might want to requeue the mounts local-fs acquired through
193 that automatically.
194
195 * rework specifier logic so that we can distuingish OOM errors from other errors
196
197 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
198
199 * journal-or-kmsg is currently broken? See reverted commit 4a01181e460686d8b4a543b1dfa7f77c9e3c5ab8.
200
201 * remove any syslog support from log.c -- we probably can't do this before split-off udev is gone for good
202
203 * documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
204
205 * add a tool that lists active timer units plus their next elapstion and the time the units ran last
206
207 * man: document the very specific env the shutdown drop-in tools live in
208
209 * shutdown logging: store to EFI var, and store to USB stick?
210
211 * man: extend runlevel(8) to mention that runlevels suck, and are dead. Maybe add runlevel(7) with a note about that too
212
213 * systemctl: maybe add "systemctl add-wants" or so...
214
215 * man: add more examples to man pages
216
217 * man: maybe sort directives in man pages, and take sections from --help and apply them to man too
218
219 * write UI tool that pops up emergency messages from the journal as notification
220
221 * think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service
222
223 * msgcatalog determine default language of entries from the catalog
224 file name. i.e. foobar.de_DE.catalog would set the default entry
225 name for the entries to de_DE if they aren't explicitly suffixed
226 individually.
227
228 * use "log level" rather than "log priority" everywhere
229
230 * timedate: have global on/off switches for auto-time (NTP), and auto-timezone that connman can subscribe to.
231
232 * Honour "-" prefix for InaccessibleDirectories= and ReadOnlyDirectories= to
233 suppress errors of the specified path doesn't exist
234
235 * dev-setup.c: when running in a container, create a tiny stub udev
236 database with the systemd tag set for all network interfaces found,
237 so that libudev reports them as present, and systemd's .device units
238 will be activated.
239
240 * merge unit_kill_common() and unit_kill_context()
241
242 * introduce ExecCondition= in services
243
244 * EFI:
245 - fsck hookup for the ESP mount is missing
246 - write man page for efi boot generator
247 - honor language efi variables for default language selection (if there are any?)
248 - honor timezone efi variables for default timezone selection (if there are any?)
249 - introduce bootctl (backed by systemd-bootd) to control temporary and persistent default boot goal plus efi variables
250
251 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
252
253 * re-enable "make check" for gtk-doc (broken for unknown reason)
254
255 * fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points
256 https://bugzilla.redhat.com/show_bug.cgi?id=812826
257
258 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
259
260 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
261
262 * logind:
263 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
264 - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
265 - logind: wakelock/opportunistic suspend support
266 - Add pretty name for seats in logind
267 - logind: allow showing logout dialog from system?
268 - logind: spawn user@..service on login
269 - logind: non-local X11 server handling
270 - logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API
271 - pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree
272 - logind: GetSessionByPID() should accept 0 as PID value
273 - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case
274
275 * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
276
277 * DeviceAllow/DeviceDeny: disallow everything by default, but whitelist /dev/zero, /dev/null and friends
278
279 * service: watchdog logic: for testing purposes allow ping, but do not require pong
280
281 * journal:
282 - do not use magic msync() in src/journal/journal-file.c, just call fsync()
283 <alxchk> poettering: looks like msync just calls vfs_fsync
284 http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/mm/msync.c#n18
285 - import and delete pstore filesystem content at startup
286 - journald: also get thread ID from client, plus thread name
287 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
288 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
289 - fallback to /dev/log based logging in libsystemd-journal, if we can't log natively?
290 - declare the local journal protocol stable in the wiki interface chart
291 - journal: reuse XZ context
292 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
293 - journald: when dropping msgs due to ratelimit make sure to write
294 "dropped %u messages" not only when we are about to print the next
295 message that works, but alraedy after a short tiemout
296 - check if we can make journalctl by default use --follow mode inside of less if called without args?
297 - maybe add API to send pairs of iovecs via sd_journal_send
298 - journal: when writing journal auto-rotate if time jumps backwards
299 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
300 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
301 and !FOOBAR for events without FOOBAR.
302 - journal: when rotating, copy over old acls/access mode
303 - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs.
304 - 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.
305 - journal: find a way to allow dropping history early, based on priority, other rules
306 - journal: When used on NFS, check payload hashes
307 - Introduce journalctl -b <nr> to show journal messages of a previous boot
308 - journald: check whether it is OK if the client can still modify delivered journal entries
309 - journal live copy, based on libneon (client) and libmicrohttpd (server)
310 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
311 - refuse taking lower-case variable names in sd_journal_send() and friends.
312 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
313 - journal: deal nicely with byte-by-byte copied files, especially regards header
314 - journalctl: show multiline log messages sanely, expand tabs, and show all valid utf8 messages
315 - journal: store euid in journal if it differs from uid
316 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
317 - Replace utmp, wtmp, btmp, and lastlog completely with journal
318 - Port upower to use the journal for historical power information used in future calculations
319
320 * document:
321 - document unit_name_mangle()
322 - document that deps in [Unit] sections ignore Alias= fileds in
323 [Install] units of other units, unless those units are disabled
324 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
325 - Document word splitting syntax for ExecStart= and friends
326 - document that units from /etc override those from /usr and /run
327 - document the exit codes when services fail before they are exec()ed
328 - document that service reload may be implemented as service reexec
329 - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications
330
331 * systemctl:
332 - systemctl list-jobs - show dependencies
333 - add systemctl switch to dump transaction without executing it
334 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
335 - "systemctl disable" on a static unit prints no message and does
336 nothing. "systemctl enable" does nothing, and gives a bad message
337 about it. Should fix both to print nice actionable messages.
338 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
339 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
340 - systemctl enable: improve the success messages (i.e. more human readable, less shell-like)
341 - systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards?
342 - systemctl: "Journal has been rotated since unit was started." message is misleading
343 - support "systemctl stop foobar@.service" to stop all units matching a certain template
344 - Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files"
345
346 * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations
347
348 * deal with sendmail/postfix exclusivity
349
350 * timer units:
351 - configurable jitter for timer events
352 - timer events with system resume
353 - timer units should get the ability to trigger when:
354 o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
355 o DST changes
356 - Support 2012-02~4 as syntax for specifying the fourth to last day of the month.
357
358 * update the kernel's TZ (sys_tz) when DST changes
359
360 * sync down the system time to the RTC when:
361 - CLOCK_REALTIME makes jumps (the user explicitely requested a time set)
362 - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp)
363 This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime
364 mode, it will never touch the RTC if the no reliable time source is active or the
365 user did not request anything like it.
366
367 * hwdb:
368 - implement conditional properties (dmi matches)
369 - hwdb --filter=ID_DRIVE_*
370
371 * if booted in "quiet" mode, and an error happens, turn on status
372 output again, so that the emergency mode isn't totally
373 surprising. Also, terminate plymouth.
374
375 * libunwind support for coredump pattern hook, and includes this in
376 the message for coredumps. After all, libunwind is now capable to
377 unwind coredumps since a few weeks ago. This probably requires that
378 we have nice support for multi-line messages on display in
379 logs-show.c. Alternatively: use libelfutil, which seems to be the
380 better supported alternative.
381
382 * figure out relation of --all and --full in the various tools
383
384 * add libsystemd-password or so to query passwords during boot using the password agent logic
385
386 * 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().
387
388 * fedup: add --unit to systemctl switch-root somehow
389 * fedup: don't delete initrd on switch-root
390 * fedup: generator
391
392 * timedated:
393 - timedated: refuse time changes when NTP is on
394
395 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
396
397 * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields
398
399 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
400
401 * add "provisioning" instructions to setup an empty /etc + /var
402 - used to setup a new container from a shared /usr
403 - superset of tmpfiles model
404 - instructions shipped by packages and stored in /usr/lib/
405 - compose /etc/passwd and /etc/group, copy files
406 - able to create uid + gid used by packages, for file ownership
407
408 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
409 waiting for the timeout
410
411 * high level net_prio setting in execution context
412
413 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
414
415 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't
416
417 * refuse boot if /etc/machine-id is not useful (or set taint?)
418
419 * nspawn:
420 - nspawn: consider changing users for -u with su, so that NSS resolving works correctly
421 - nspawn: implement personality changes a la linux32(8)
422 - nspawn: --read-only is not applied recursively to submounts
423 - nspawn: make use of device cgroup controller by default
424 - bind mount read-only the cgroup tree higher than nspawn
425 - nspawn: investigate whether we can support the same as LXC's
426 lxc.network.type=phys mode, and pass through entire network
427 interfaces to the container
428 - nspawn: maybe add a way to drop additional caps, in addition to add additional caps
429 - nspawn: maybe explicitly reset loginuid?
430 - nspawn: make it work for dwalsh and shared /usr containers -- tmpfs mounts as command line parameters, selinux exec context
431
432 * cryptsetup:
433 - cryptsetup-generator: allow specification of passwords in crypttab itself
434 - move cryptsetup key caching into kernel keyctl?
435 https://bugs.freedesktop.org/show_bug.cgi?id=54982
436 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
437
438 * move debug shell to tty6 and make sure this doesn't break the gettys on tty6
439
440 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
441 https://bugs.freedesktop.org/show_bug.cgi?id=54712
442
443 * after deserializing sockets in socket.c we should reapply sockopts and things
444
445 * make timer units go away after they elapsed
446
447 * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html
448 (network and remote-fs on shutdown)
449
450 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
451
452 * move PID 1 segfaults to /var/lib/systemd/coredump?
453
454 * create /sbin/init symlinks from the build system
455
456 * Query Paul Moore about relabelling socket fds while they are open
457
458 * system.conf should have controls for cgroups
459
460 * allow writing multiple conditions in unit files on one line
461
462 * explore multiple service instances per listening socket idea
463
464 * shutdown: don't read-only mount anything when running in container
465
466 * MountFlags=shared acts as MountFlags=slave right now.
467
468 * ReadOnlyDirectories= is not applied recursively to submounts
469
470 * drop PID 1 reloading, only do reexecing (difficult: Reload()
471 currently is properly synchronous, Reexec() is weird, because we
472 can't delay the response properly until we are back, so instead of
473 being properly synchronous we just keep open the fd and close it
474 when done. That means clients don't get a successful method reply,
475 but much rather a disconnect on success.
476
477 * remember which condition failed for services, not just the fact that something failed
478
479 * use opterr = 0 for all getopt tools
480
481 * properly handle loop back mounts via fstab, especially regards to fsck/passno
482
483 * allow services with no ExecStart= but with an ExecStop=
484
485 * dracut-shutdown needs to be ordered before unmounting /boot
486
487 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
488
489 * rename "userspace" to "core-os"
490
491 * syscall filter:
492 - syscall filter: add knowledge about compat syscalls
493 - syscall filter: don't enforce no new privs?
494 - syscall filter: option to return EPERM rather than SIGSYS?
495 - syscall filter: port to libseccomp
496 - system-wide seccomp filter
497
498 * .device aliases need to be implemented with the "following" logic, probably.
499
500 * load-fragment: when loading a unit file via a chain of symlinks
501 verify that it isn't masked via any of the names traversed.
502
503 * introduce Type=pid-file
504
505 * change Requires=basic.target to RequisiteOverride=basic.target
506
507 * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
508
509 * move passno parsing to fstab generator
510
511 * automount: implement expire:
512 - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD
513 - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD
514 - every timeout/4 (original autofs logic)
515 - blocking, needs a thread
516 - run until -EAGAIN
517 - receive expire packet on pipe if kernel tells the timeout is over
518 - call umount
519 - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD
520 - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns
521
522 * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel)
523
524 * ExecOnFailure=/usr/bin/foo
525
526 * udev:
527 - remove src/udev/udev-builtin-firmware.c (CONFIG_FW_LOADER_USER_HELPER=n)
528 - move to LGPL
529 - unify utf8 validator code with shared/
530 - kill scsi_id
531 - add trigger --subsystem-match=usb/usb_device device
532
533 * cleanup syslog 'priority' vs. 'level' wording
534
535 * 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
536
537 * support container_ttys=
538
539 * introduce mix of BindTo and Requisite
540
541 * add DeleteSocketsOnStop=yes|no option to socket units
542
543 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
544
545 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
546
547 * default unix qlen is too small (10). bump sysctl? add sockopt?
548
549 * save coredump in Windows/Mozilla minidump format
550
551 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
552
553 * clean up session cgroups that remain after logout (think sshd), but eventually run empty
554
555 * when an instanced service exits, remove its parent cgroup too if possible.
556
557 * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max
558
559 * be able to specify a forced restart of service A where service B depends on, in case B
560 needs to be auto-respawned?
561
562 * when a bus name of a service disappears from the bus make sure to queue further activation requests
563
564 * tmpfiles: apply "x" on "D" too (see patch from William Douglas)
565
566 * for services: don't set $HOME in services unless requested
567
568 * hide PAM/TCPWrap options in fragment parser when compile time disabled
569
570 * when we automatically restart a service, ensure we restart its rdeps, too.
571
572 * allow Type=simple with PIDFile=
573 https://bugzilla.redhat.com/show_bug.cgi?id=723942
574
575 * move PAM code into its own binary
576
577 * implement Register= switch in .socket units to enable registration
578 in Avahi, RPC and other socket registration services.
579
580 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
581
582 * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination
583
584 * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1)
585
586 * verify that the AF_UNIX sockets of a service in the fs still exist
587 when we start a service in order to avoid confusion when a user
588 assumes starting a service is enough to make it accessible
589
590 * Make it possible to set the keymap independently from the font on
591 the kernel cmdline. Right now setting one resets also the other.
592
593 * and a dbus call to generate target from current state
594
595 * readahead:
596 - drop /.readahead on bigger upgrades with yum
597 - move readahead files into /var (look for them with .path units?)
598 - readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO
599 - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
600 - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
601
602 * add support for /bin/mount -s
603
604 * GC unreferenced jobs (such as .device jobs)
605
606 * write blog stories about:
607 - hwdb: what belongs into it, lsusb
608 - enabling dbus services
609 - status update
610 - how to make changes to sysctl and sysfs attributes
611 - remote access
612 - how to pass throw-away units to systemd, or dynamically change properties of existing units
613 - how to integrate cgconfig and suchlike with systemd
614 - testing with Harald's awesome test kit
615 - auto-restart
616 - how to develop against journal browsing APIs
617 - the journal HTTP iface
618 - non-cgroup resource management
619 - dynamic resource management with cgroups
620 - refreshed, longer missions statement
621 - calendar time events
622 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
623 - how to create your own target
624 - instantiated apache, dovecot and so on
625 - hooking a script into various stages of shutdown/rearly booot
626
627 * allow port=0 in .socket units
628
629 * support systemd.mask= on the kernel command line.
630
631 * recreate systemd's D-Bus private socket file on SIGUSR2
632
633 * Support --test based on current system state
634
635 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
636
637 * maybe introduce ExecRestartPre=
638
639 * dot output for --test showing the 'initial transaction'
640
641 * port over to LISTEN_FDS/LISTEN_PID:
642 - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH
643 - cups HAVEPATCH
644 - postfix, saslauthd
645 - apache/samba
646 - libvirtd (/var/run/libvirt/libvirt-sock-ro)
647 - bluetoothd (/var/run/sdp! @/org/bluez/audio!)
648 - distccd
649
650 * fingerprint.target, wireless.target, gps.target, netdevice.target
651
652 * io priority during initialization
653
654 * drop cap bounding set in readahead and other services
655
656 * systemd-python:
657 - figure out a simple way to wait for journal events in a way that
658 works with ^C
659 - add documentation to systemd.daemon
660
661 * bootchart:
662 - plot per-process IO utilization
663 - group processes based on service association (cgroups)
664 - document initcall_debug
665 - kernel cmdline "bootchart" option for simplicity?
666
667 External:
668
669 * dbus:
670 - dbus --user
671 - natively watch for dbus-*.service symlinks (PENDING)
672 - allow specification of socket mode/umask when allocating DBusServer
673 - allow disabling of fd passing when connecting a AF_UNIX connection
674 - allow disabling of UID passing for AUTH EXTERNAL
675 - always pass cred data along each message
676 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
677
678 * fix alsa mixer restore to not print error when no config is stored
679
680 * make cryptsetup lower --iter-time
681
682 * patch kernel for xattr support in /dev, /proc/, /sys?
683
684 * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only
685 in an ~30 minutes window. It does not adjust larger differences. Find a way
686 to tell the kernel, to always do a full time sync when the RTC is in UTC and
687 we are in 11-minutes-mode. When we trust the system time to NTP we also want
688 the RTC to sync up.
689
690 * kernel: add device_type = "fb", "fbcon" to class "graphics"
691
692 * drop accountsservice's StandardOutput=syslog and Type=dbus fields
693
694 * dbus upstream still refers to dbus.target and shouldn't
695
696 * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
697
698 * add "# export SYSTEMD_PAGER=" to bash login
699
700 * /usr/bin/service should actually show the new command line
701
702 * 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=
703
704 * fedora: F20: go timer units all the way, leave cron.daily for cron
705
706 * neither pkexec nor sudo initialize environ[] from the PAM environment?
707
708 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
709
710 Regularly:
711
712 * look for close() vs. close_nointr() vs. close_nointr_nofail()
713
714 * check for strerror(r) instead of strerror(-r)
715
716 * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
717
718 * %m in printf() instead of strerror(errno);
719
720 * pahole
721
722 * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()!
723
724 * use secure_getenv() instead of getenv() where appropriate
725
726 * link up selected blog stories from man pages and unit files Documentation= fields
727
728 Scheduled for removal or fixing:
729
730 * xxxOverridable dependencies (probably: fix)