]>
Commit | Line | Data |
---|---|---|
663996b3 | 1 | Bugfixes: |
14228c0d MB |
2 | |
3 | * enabling an instance unit creates pointless link, and | |
4 | the unit will be started with getty@getty.service: | |
5 | $ systemctl enable getty@.service | |
6 | ln -s '/usr/lib/systemd/system/getty@.service' '/etc/systemd/system/getty.target.wants/getty@.service' | |
663996b3 MS |
7 | |
8 | * check systemd-tmpfiles for selinux context hookup for mknod(), symlink() and similar | |
9 | ||
10 | * swap units that are activated by one name but shown in the kernel under another are semi-broken | |
11 | ||
12 | * Dangling symlinks of .automount unit files in .wants/ directories, set up | |
13 | automount points even when the original .automount file did not exist | |
14 | anymore. Only the .mount unit was still around. | |
15 | ||
16 | * make polkit checks async | |
17 | ||
18 | * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point. | |
19 | ||
14228c0d MB |
20 | * ellipsize_mem must take into account multi-byte unicode characters, and |
21 | - make the resulting line the requested number of *characters*, not *bytes*, | |
22 | - avoid truncuating multi-byte sequences in the middle. | |
23 | ||
24 | * When we detect invalid UTF-8, we cant't use it in an error message: | |
25 | log...("Path is not UTF-8 clean, ignoring assignment: %s", rvalue); | |
26 | ||
27 | * shorten the message to sane length: | |
663996b3 | 28 | |
14228c0d | 29 | Cannot add dependency job for unit display-manager.service, ignoring: Unit display-manager.service failed to load: No such file or directory. See system logs and 'systemctl status display-manager.service' for details. |
663996b3 | 30 | |
14228c0d MB |
31 | Fedora 20: |
32 | ||
33 | * external: ps should gain colums for slice and machine | |
663996b3 MS |
34 | |
35 | * localed: | |
36 | - localectl: support new converted x11→console keymaps | |
37 | ||
14228c0d MB |
38 | * when installing fedora with yum --installroot /var/run is a directory, not a symlink |
39 | https://bugzilla.redhat.com/show_bug.cgi?id=975864 | |
40 | ||
41 | CGroup Rework Completion: | |
42 | ||
43 | * introduce "mainpid" for scopes (or maybe not?) | |
44 | ||
45 | * implement system-wide DefaultCPUAccounting=1 switch (and similar for blockio, memory?) | |
46 | ||
47 | * implement per-slice CPUFairScheduling=1 switch | |
48 | ||
49 | * handle jointly mounted controllers correctly | |
50 | ||
51 | * introduce high-level settings for RT budget, swappiness | |
52 | ||
53 | * wiki: guidelines how to make use of new cgroup apis | |
54 | ||
663996b3 MS |
55 | Features: |
56 | ||
14228c0d MB |
57 | * set $SHELL where we set $HOME and $USER when User= is set of a service, drop its manual setting from user@.service |
58 | ||
59 | * we probably should replace the left-over uses of strv_append() and replace them by strv_push() or strv_extend() | |
60 | ||
61 | * move config_parse_path_strv() out of conf-parser.c | |
62 | ||
63 | * libdsystemd-bus should expose utf8 validation calls | |
64 | ||
65 | * After coming back from hibernation reset hibernation swap partition | |
66 | ||
67 | * If we try to find a unit via a dangling symlink generate a clean | |
68 | error. Currently we just ignore it and read the unit from the search | |
69 | path anyway. | |
70 | ||
71 | * When a Type=forking service fails and needed another service that | |
72 | service is not cleaned up again when it has StopWhenUnneeded=yes | |
73 | http://lists.freedesktop.org/archives/systemd-devel/2013-July/012141.html | |
74 | ||
75 | * backlight: properly handle multiple backlight devices for the same | |
76 | hardware: at shutdown we should only save the backlight setting for | |
77 | the "best" way to access the backlight. Strategy should be: at | |
78 | shutdown, ignore all backlights that are connected to a non-eDP or | |
79 | non-LVDS port, and then prefer the firmware device over platform | |
80 | device over raw device per-PCI card. Delete all old data. At boot | |
81 | simply apply whatever data we find. Also see | |
82 | http://cgit.freedesktop.org/libbacklight/tree/libbacklight.c#n194 | |
83 | ||
84 | * rfkill: save/restore soft rfkill status across reboots | |
85 | ||
86 | * refuse boot if /etc/os-release is missing or /etc/machine-id cannot be set up | |
87 | ||
88 | * ensure scope units may be started only a single time | |
89 | ||
90 | * document that in instead of FsckPassNo= people should just add a manual dep to systemd-fsck@.service to their mount units. | |
91 | ||
92 | * better error message if you run systemctl without systemd running | |
93 | ||
94 | * systemctl status output should should include list of triggering units and their status | |
95 | ||
96 | * for transient units, instead of writing out drop-ins for all properties consider serializing them in the normal serialization stream | |
97 | ||
98 | * logind: when logging out, remove user-owned sysv and posix IPC objects | |
99 | ||
100 | * session scopes/user unit: add RequiresMountsFor for the home directory of the user | |
101 | ||
102 | * add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr. | |
103 | ||
104 | * journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax? | |
105 | ||
106 | * given that logind/machined now let PID 1 do all nasty work we can | |
107 | probably reduce the capability set they retain substantially. | |
108 | ||
109 | * btrfs raid assembly: some .device jobs stay stuck in the queue | |
110 | ||
111 | * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros. | |
112 | ||
113 | * Fedora: post FPC ticket to move add %tmpfiles_create to the packaging guidelines | |
114 | ||
115 | * make sure gdm doesn't use multi-user-x but the new default X configuration file, and then remove multi-user-x from systemd | |
116 | ||
117 | * when parsing calendar timestamps support the UTC timezone (even if we won't support arbitrary timezone specs, support UTC itself certainly makes sense), also support syntaxes such as +0200 | |
118 | ||
119 | * when a kernel driver logs in a tight loop we should ratelimit that too. | |
120 | ||
121 | * "systemctl disable" of a unit instance removes all symlinks, should | |
122 | only remove the instance symlink (systemctl disable of a template | |
123 | unit however should remove them all). | |
124 | ||
125 | * journald: optionally, log debug messages to /run but everything else to /var | |
126 | ||
127 | * systemctl list-unit-files should list generated files (and probably with a new state "generated" for them, or so) | |
128 | ||
129 | * journald: when we drop syslog messages because the syslog socket is | |
130 | full, make sure to write how many messages are lost as first thing | |
131 | to syslog when it works again. | |
132 | ||
133 | * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted. | |
134 | ||
135 | * load .d/*.conf dropins for device units | |
136 | ||
137 | * add a fixed dbus path for "my own unit", "my own session", ... to PID1, logind, ... | |
138 | ||
139 | * service_coldplug() appears to reinstall the wrong stop timeout watch? | |
140 | ||
141 | * transient units: allow creating auxiliary units with the same call | |
142 | ||
143 | * how to reset dynamically changed attributes sanely? | |
144 | ||
145 | * when reloading configuration, apply new cgroup configuration | |
146 | ||
147 | * journald: make sure ratelimit is actually really per-service with the new cgroup changes | |
148 | ||
149 | * gparted needs to disable auto-activation of mount units somehow, or | |
150 | maybe we should stop doing auto-activation of this after boot | |
151 | entirely. https://bugzilla.gnome.org/show_bug.cgi?id=701676 | |
152 | Maybe take a BSD lock at the disk device node and teach udev to | |
153 | check for that and suppress event handling. | |
154 | ||
155 | * when recursively showing the cgroup hierarchy, optionally also show | |
156 | the hierarchies of child processes | |
157 | ||
158 | * document logic of auto/noauto and fail/nofail in fstab in systemd.mount or systemd-fstab-generator man page | |
159 | ||
160 | * we should properly escape ssh hostnames we add into dbus server strings | |
161 | ||
162 | * something pulls in pcre as so dep into our daemons such as hostnamed. | |
163 | ||
164 | * document systemd-journal-flush.service properly | |
165 | ||
166 | * change systemd-journal-flush into a service that stays around during | |
167 | boot, and causes the journal to be moved back to /run on shutdown, | |
168 | so that we don't keep /var busy. This needs to happen synchronously, | |
169 | hence doing this via signals is not going to work. | |
170 | ||
171 | * allow implementation of InaccessibleDirectories=/ plus | |
172 | ReadOnlyDirectories=... for whitelisting files for a service. | |
173 | ||
174 | * libsystemd-bus: | |
175 | - default policy (allow uid == 0 and our own uid) | |
176 | - enforce alignment of pointers passed in | |
177 | - when kdbus doesn't take our message without memfds, try again with memfds | |
178 | - implement translator service | |
179 | - port systemd to new library | |
180 | - implement busname unit type in systemd | |
181 | - move to gvariant | |
182 | - merge busctl into systemctl or so? | |
183 | - synthesize sd_bus_message objects from kernel messages | |
184 | - properly implement name registry ioctls for kdbus | |
185 | - get rid of object hash table, use decision tree everyhwere instead? | |
186 | - implement monitor logic | |
187 | - object vtable logic | |
188 | - longer term: | |
189 | * priority queues | |
190 | * priority inheritance | |
191 | ||
192 | * in the final killing spree, detect processes from the root directory, and | |
193 | complain loudly if they have argv[0][0] == '@' set. | |
194 | https://bugzilla.redhat.com/show_bug.cgi?id=961044 | |
195 | ||
196 | * add an option to nspawn that uses seccomp to make socket(AF_NETLINK, | |
197 | SOCK_RAW, NETLINK_AUDIT) fail the the appropriate error code that | |
198 | makes the audit userspace to think auditing is not available in the | |
199 | kernel. | |
200 | ||
663996b3 MS |
201 | * Introduce a way how we can kill the main process of a service with KillSignal, but all processes with SIGKILL later on |
202 | https://bugzilla.redhat.com/show_bug.cgi?id=952634 | |
203 | ||
14228c0d | 204 | * maybe add a warning to the unit file parses where the access mode of unit files is nonsensical. |
663996b3 | 205 | |
14228c0d | 206 | * investigate endianness issues of UUID vs. GUID |
663996b3 MS |
207 | |
208 | * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we | |
209 | should be able to safely try another attempt when the bus call LoadUnit() is invoked. | |
210 | ||
211 | * if pam_systemd is invoked by su from a process that is outside of a | |
212 | any session we should probably just become a NOP, since that's | |
213 | usually not a real user session but just some system code that just | |
214 | needs setuid(). | |
215 | ||
216 | * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login. | |
217 | ||
218 | * add a pam module that on password changes updates any LUKS slot where the password matches | |
219 | ||
220 | * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases... | |
221 | ||
222 | * timedatctl, localectl: possibly make some commands work without the daemon, for chroot situations... | |
223 | ||
224 | * logind: add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly. | |
225 | ||
663996b3 MS |
226 | * cgtop: make cgtop useful in a container |
227 | ||
228 | * test/: | |
229 | - add 'set -e' to scripts in test/ | |
230 | - make stuff in test/ work with separate output dir | |
231 | - remove all the duplicated code in test/ | |
232 | ||
663996b3 MS |
233 | * systemctl delete x.snapshot leaves no trace in logs (at least at default level). |
234 | ||
235 | * make the coredump collector tool move itself into the user's cgroup | |
236 | so that the coredump is properly written to the user's own journal | |
237 | file. | |
238 | ||
663996b3 MS |
239 | * seems that when we follow symlinks to units we prefer the symlink |
240 | destination path over /etc and /usr. We shouldn't do that. Instead | |
241 | /etc should always override /run+/usr and also any symlink | |
242 | destination. | |
243 | ||
663996b3 MS |
244 | * when isolating, try to figure out a way how we implicitly can order |
245 | all units we stop before the isolating unit... | |
246 | ||
663996b3 MS |
247 | * add ConditionArchitecture= or so |
248 | ||
249 | * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off}) | |
250 | ||
663996b3 MS |
251 | * Support SO_REUSEPORT with socket activation: |
252 | - Let systemd maintain a pool of servers. | |
253 | - Use for seamless upgrades, by running the new server before stopping the | |
254 | old. | |
255 | ||
256 | * after all byte-wise realloc() is slow, even on glibc, so i guess we | |
257 | need manual exponential loops after all | |
258 | ||
259 | * BootLoaderSpec: drop allowing ext234 for $BOOT. Clarify that the | |
260 | kernel has to be in $BOOT. Clarify that the boot loader should be | |
261 | installed to the ESP. Define a way how an installer can figure out | |
262 | whether a BLS compliant boot loader is installed. | |
263 | ||
264 | * man: remove .include documentation, and instead push people to use .d/*.conf | |
265 | ||
266 | * think about requeuing jobs when daemon-reload is issued? usecase: | |
267 | the initrd issues a reload after fstab from the host is accessible | |
268 | and we might want to requeue the mounts local-fs acquired through | |
269 | that automatically. | |
270 | ||
14228c0d | 271 | * rework specifier logic so that we can distinguish OOM errors from other errors |
663996b3 MS |
272 | |
273 | * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep() | |
274 | ||
275 | * journal-or-kmsg is currently broken? See reverted commit 4a01181e460686d8b4a543b1dfa7f77c9e3c5ab8. | |
276 | ||
277 | * remove any syslog support from log.c -- we probably can't do this before split-off udev is gone for good | |
278 | ||
279 | * documentation: recommend to connect the timer units of a service to the service via Also= in [Install] | |
280 | ||
14228c0d | 281 | * add a tool that lists active timer units plus their next elapse and the time the units ran last |
663996b3 MS |
282 | |
283 | * man: document the very specific env the shutdown drop-in tools live in | |
284 | ||
285 | * shutdown logging: store to EFI var, and store to USB stick? | |
286 | ||
14228c0d | 287 | * man: extend runlevel(8) to mention that runlevels suck, and are dead. Maybe add runlevel(7) with a note about that too |
663996b3 MS |
288 | |
289 | * systemctl: maybe add "systemctl add-wants" or so... | |
290 | ||
291 | * man: add more examples to man pages | |
292 | ||
293 | * man: maybe sort directives in man pages, and take sections from --help and apply them to man too | |
294 | ||
295 | * write UI tool that pops up emergency messages from the journal as notification | |
296 | ||
297 | * think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service | |
298 | ||
299 | * msgcatalog determine default language of entries from the catalog | |
300 | file name. i.e. foobar.de_DE.catalog would set the default entry | |
301 | name for the entries to de_DE if they aren't explicitly suffixed | |
302 | individually. | |
303 | ||
304 | * use "log level" rather than "log priority" everywhere | |
305 | ||
306 | * timedate: have global on/off switches for auto-time (NTP), and auto-timezone that connman can subscribe to. | |
307 | ||
663996b3 MS |
308 | * dev-setup.c: when running in a container, create a tiny stub udev |
309 | database with the systemd tag set for all network interfaces found, | |
310 | so that libudev reports them as present, and systemd's .device units | |
311 | will be activated. | |
312 | ||
313 | * merge unit_kill_common() and unit_kill_context() | |
314 | ||
315 | * introduce ExecCondition= in services | |
316 | ||
317 | * EFI: | |
318 | - fsck hookup for the ESP mount is missing | |
319 | - write man page for efi boot generator | |
320 | - honor language efi variables for default language selection (if there are any?) | |
321 | - honor timezone efi variables for default timezone selection (if there are any?) | |
14228c0d | 322 | - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables |
663996b3 MS |
323 | |
324 | * maybe do not install getty@tty1.service symlink in /etc but in /usr? | |
325 | ||
326 | * re-enable "make check" for gtk-doc (broken for unknown reason) | |
327 | ||
328 | * fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points | |
329 | https://bugzilla.redhat.com/show_bug.cgi?id=812826 | |
330 | ||
331 | * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word | |
332 | ||
333 | * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units. | |
334 | ||
335 | * logind: | |
336 | - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around | |
337 | - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot | |
338 | - logind: wakelock/opportunistic suspend support | |
339 | - Add pretty name for seats in logind | |
340 | - logind: allow showing logout dialog from system? | |
663996b3 MS |
341 | - logind: non-local X11 server handling |
342 | - logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API | |
343 | - pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree | |
344 | - logind: GetSessionByPID() should accept 0 as PID value | |
345 | - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case | |
346 | ||
347 | * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty. | |
348 | ||
663996b3 MS |
349 | * service: watchdog logic: for testing purposes allow ping, but do not require pong |
350 | ||
351 | * journal: | |
352 | - do not use magic msync() in src/journal/journal-file.c, just call fsync() | |
353 | <alxchk> poettering: looks like msync just calls vfs_fsync | |
354 | http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/mm/msync.c#n18 | |
355 | - import and delete pstore filesystem content at startup | |
356 | - journald: also get thread ID from client, plus thread name | |
357 | - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups | |
358 | - add API to close/reopen/get fd for journal client fd in libsystemd-journal. | |
359 | - fallback to /dev/log based logging in libsystemd-journal, if we can't log natively? | |
360 | - declare the local journal protocol stable in the wiki interface chart | |
361 | - journal: reuse XZ context | |
362 | - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg | |
363 | - journald: when dropping msgs due to ratelimit make sure to write | |
364 | "dropped %u messages" not only when we are about to print the next | |
365 | message that works, but alraedy after a short tiemout | |
366 | - check if we can make journalctl by default use --follow mode inside of less if called without args? | |
367 | - maybe add API to send pairs of iovecs via sd_journal_send | |
368 | - journal: when writing journal auto-rotate if time jumps backwards | |
369 | - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access | |
370 | - journactl: support negative filtering, i.e. FOOBAR!="waldo", | |
371 | and !FOOBAR for events without FOOBAR. | |
372 | - journal: when rotating, copy over old acls/access mode | |
373 | - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs. | |
374 | - 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. | |
375 | - journal: find a way to allow dropping history early, based on priority, other rules | |
376 | - journal: When used on NFS, check payload hashes | |
663996b3 MS |
377 | - journald: check whether it is OK if the client can still modify delivered journal entries |
378 | - journal live copy, based on libneon (client) and libmicrohttpd (server) | |
379 | - journald: add kernel cmdline option to disable ratelimiting for debug purposes | |
380 | - refuse taking lower-case variable names in sd_journal_send() and friends. | |
381 | - journald: we currently rotate only after MaxUse+MaxFilesize has been reached. | |
382 | - journal: deal nicely with byte-by-byte copied files, especially regards header | |
14228c0d | 383 | - journalctl: expand tabs |
663996b3 MS |
384 | - journal: store euid in journal if it differs from uid |
385 | - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit | |
386 | - Replace utmp, wtmp, btmp, and lastlog completely with journal | |
387 | - Port upower to use the journal for historical power information used in future calculations | |
388 | ||
389 | * document: | |
390 | - document unit_name_mangle() | |
391 | - document that deps in [Unit] sections ignore Alias= fileds in | |
392 | [Install] units of other units, unless those units are disabled | |
393 | - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets | |
394 | - Document word splitting syntax for ExecStart= and friends | |
395 | - document that units from /etc override those from /usr and /run | |
396 | - document the exit codes when services fail before they are exec()ed | |
397 | - document that service reload may be implemented as service reexec | |
398 | - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications | |
399 | ||
400 | * systemctl: | |
401 | - systemctl list-jobs - show dependencies | |
402 | - add systemctl switch to dump transaction without executing it | |
403 | - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done | |
404 | - "systemctl disable" on a static unit prints no message and does | |
405 | nothing. "systemctl enable" does nothing, and gives a bad message | |
406 | about it. Should fix both to print nice actionable messages. | |
407 | - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service | |
408 | - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible | |
409 | - systemctl enable: improve the success messages (i.e. more human readable, less shell-like) | |
410 | - systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards? | |
411 | - systemctl: "Journal has been rotated since unit was started." message is misleading | |
412 | - support "systemctl stop foobar@.service" to stop all units matching a certain template | |
413 | - Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files" | |
414 | ||
415 | * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations | |
416 | ||
417 | * deal with sendmail/postfix exclusivity | |
418 | ||
419 | * timer units: | |
420 | - configurable jitter for timer events | |
421 | - timer events with system resume | |
422 | - timer units should get the ability to trigger when: | |
423 | o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET) | |
424 | o DST changes | |
425 | - Support 2012-02~4 as syntax for specifying the fourth to last day of the month. | |
426 | ||
427 | * update the kernel's TZ (sys_tz) when DST changes | |
428 | ||
429 | * sync down the system time to the RTC when: | |
430 | - CLOCK_REALTIME makes jumps (the user explicitely requested a time set) | |
431 | - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp) | |
432 | This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime | |
433 | mode, it will never touch the RTC if the no reliable time source is active or the | |
434 | user did not request anything like it. | |
435 | ||
663996b3 MS |
436 | * if booted in "quiet" mode, and an error happens, turn on status |
437 | output again, so that the emergency mode isn't totally | |
438 | surprising. Also, terminate plymouth. | |
439 | ||
440 | * libunwind support for coredump pattern hook, and includes this in | |
441 | the message for coredumps. After all, libunwind is now capable to | |
442 | unwind coredumps since a few weeks ago. This probably requires that | |
443 | we have nice support for multi-line messages on display in | |
444 | logs-show.c. Alternatively: use libelfutil, which seems to be the | |
445 | better supported alternative. | |
446 | ||
447 | * figure out relation of --all and --full in the various tools | |
448 | ||
449 | * add libsystemd-password or so to query passwords during boot using the password agent logic | |
450 | ||
451 | * 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(). | |
452 | ||
453 | * fedup: add --unit to systemctl switch-root somehow | |
454 | * fedup: don't delete initrd on switch-root | |
455 | * fedup: generator | |
456 | ||
457 | * timedated: | |
458 | - timedated: refuse time changes when NTP is on | |
459 | ||
460 | * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed | |
461 | ||
462 | * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields | |
463 | ||
464 | * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel | |
465 | ||
466 | * add "provisioning" instructions to setup an empty /etc + /var | |
467 | - used to setup a new container from a shared /usr | |
468 | - superset of tmpfiles model | |
469 | - instructions shipped by packages and stored in /usr/lib/ | |
470 | - compose /etc/passwd and /etc/group, copy files | |
471 | - able to create uid + gid used by packages, for file ownership | |
472 | ||
473 | * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without | |
474 | waiting for the timeout | |
475 | ||
476 | * high level net_prio setting in execution context | |
477 | ||
478 | * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ... | |
479 | ||
480 | * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't | |
481 | ||
482 | * refuse boot if /etc/machine-id is not useful (or set taint?) | |
483 | ||
484 | * nspawn: | |
485 | - nspawn: consider changing users for -u with su, so that NSS resolving works correctly | |
486 | - nspawn: implement personality changes a la linux32(8) | |
487 | - nspawn: --read-only is not applied recursively to submounts | |
488 | - nspawn: make use of device cgroup controller by default | |
489 | - bind mount read-only the cgroup tree higher than nspawn | |
490 | - nspawn: investigate whether we can support the same as LXC's | |
491 | lxc.network.type=phys mode, and pass through entire network | |
492 | interfaces to the container | |
493 | - nspawn: maybe add a way to drop additional caps, in addition to add additional caps | |
494 | - nspawn: maybe explicitly reset loginuid? | |
495 | - nspawn: make it work for dwalsh and shared /usr containers -- tmpfs mounts as command line parameters, selinux exec context | |
496 | ||
497 | * cryptsetup: | |
498 | - cryptsetup-generator: allow specification of passwords in crypttab itself | |
499 | - move cryptsetup key caching into kernel keyctl? | |
500 | https://bugs.freedesktop.org/show_bug.cgi?id=54982 | |
501 | - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator | |
502 | ||
503 | * move debug shell to tty6 and make sure this doesn't break the gettys on tty6 | |
504 | ||
505 | * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it | |
506 | https://bugs.freedesktop.org/show_bug.cgi?id=54712 | |
507 | ||
508 | * after deserializing sockets in socket.c we should reapply sockopts and things | |
509 | ||
510 | * make timer units go away after they elapsed | |
511 | ||
512 | * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html | |
513 | (network and remote-fs on shutdown) | |
514 | ||
515 | * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead | |
516 | ||
517 | * move PID 1 segfaults to /var/lib/systemd/coredump? | |
518 | ||
519 | * create /sbin/init symlinks from the build system | |
520 | ||
521 | * Query Paul Moore about relabelling socket fds while they are open | |
522 | ||
663996b3 MS |
523 | * allow writing multiple conditions in unit files on one line |
524 | ||
525 | * explore multiple service instances per listening socket idea | |
526 | ||
527 | * shutdown: don't read-only mount anything when running in container | |
528 | ||
529 | * MountFlags=shared acts as MountFlags=slave right now. | |
530 | ||
531 | * ReadOnlyDirectories= is not applied recursively to submounts | |
532 | ||
533 | * drop PID 1 reloading, only do reexecing (difficult: Reload() | |
534 | currently is properly synchronous, Reexec() is weird, because we | |
535 | can't delay the response properly until we are back, so instead of | |
536 | being properly synchronous we just keep open the fd and close it | |
537 | when done. That means clients don't get a successful method reply, | |
538 | but much rather a disconnect on success. | |
539 | ||
663996b3 MS |
540 | * use opterr = 0 for all getopt tools |
541 | ||
542 | * properly handle loop back mounts via fstab, especially regards to fsck/passno | |
543 | ||
544 | * allow services with no ExecStart= but with an ExecStop= | |
545 | ||
546 | * dracut-shutdown needs to be ordered before unmounting /boot | |
547 | ||
548 | * initialize the hostname from the fs label of /, if /etc/hostname does not exist? | |
549 | ||
550 | * rename "userspace" to "core-os" | |
551 | ||
552 | * syscall filter: | |
553 | - syscall filter: add knowledge about compat syscalls | |
554 | - syscall filter: don't enforce no new privs? | |
555 | - syscall filter: option to return EPERM rather than SIGSYS? | |
556 | - syscall filter: port to libseccomp | |
557 | - system-wide seccomp filter | |
558 | ||
559 | * .device aliases need to be implemented with the "following" logic, probably. | |
560 | ||
561 | * load-fragment: when loading a unit file via a chain of symlinks | |
562 | verify that it isn't masked via any of the names traversed. | |
563 | ||
564 | * introduce Type=pid-file | |
565 | ||
566 | * change Requires=basic.target to RequisiteOverride=basic.target | |
567 | ||
568 | * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr | |
569 | ||
570 | * move passno parsing to fstab generator | |
571 | ||
572 | * automount: implement expire: | |
573 | - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD | |
574 | - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD | |
575 | - every timeout/4 (original autofs logic) | |
576 | - blocking, needs a thread | |
577 | - run until -EAGAIN | |
578 | - receive expire packet on pipe if kernel tells the timeout is over | |
579 | - call umount | |
580 | - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD | |
581 | - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns | |
582 | ||
583 | * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel) | |
584 | ||
585 | * ExecOnFailure=/usr/bin/foo | |
586 | ||
663996b3 MS |
587 | * udev: |
588 | - remove src/udev/udev-builtin-firmware.c (CONFIG_FW_LOADER_USER_HELPER=n) | |
589 | - move to LGPL | |
663996b3 MS |
590 | - kill scsi_id |
591 | - add trigger --subsystem-match=usb/usb_device device | |
592 | ||
593 | * cleanup syslog 'priority' vs. 'level' wording | |
594 | ||
595 | * 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 | |
596 | ||
597 | * support container_ttys= | |
598 | ||
599 | * introduce mix of BindTo and Requisite | |
600 | ||
601 | * add DeleteSocketsOnStop=yes|no option to socket units | |
602 | ||
603 | * There's currently no way to cancel fsck (used to be possible via C-c or c on the console) | |
604 | ||
605 | * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/ | |
606 | ||
607 | * default unix qlen is too small (10). bump sysctl? add sockopt? | |
608 | ||
609 | * save coredump in Windows/Mozilla minidump format | |
610 | ||
611 | * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting) | |
612 | ||
663996b3 MS |
613 | * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max |
614 | ||
615 | * be able to specify a forced restart of service A where service B depends on, in case B | |
616 | needs to be auto-respawned? | |
617 | ||
618 | * when a bus name of a service disappears from the bus make sure to queue further activation requests | |
619 | ||
620 | * tmpfiles: apply "x" on "D" too (see patch from William Douglas) | |
621 | ||
622 | * for services: don't set $HOME in services unless requested | |
623 | ||
624 | * hide PAM/TCPWrap options in fragment parser when compile time disabled | |
625 | ||
626 | * when we automatically restart a service, ensure we restart its rdeps, too. | |
627 | ||
628 | * allow Type=simple with PIDFile= | |
629 | https://bugzilla.redhat.com/show_bug.cgi?id=723942 | |
630 | ||
631 | * move PAM code into its own binary | |
632 | ||
633 | * implement Register= switch in .socket units to enable registration | |
634 | in Avahi, RPC and other socket registration services. | |
635 | ||
636 | * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early | |
637 | ||
638 | * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination | |
639 | ||
640 | * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1) | |
641 | ||
642 | * verify that the AF_UNIX sockets of a service in the fs still exist | |
643 | when we start a service in order to avoid confusion when a user | |
644 | assumes starting a service is enough to make it accessible | |
645 | ||
14228c0d MB |
646 | * support User= and Group= attributes for AF_UNIX sockets. |
647 | ||
663996b3 MS |
648 | * Make it possible to set the keymap independently from the font on |
649 | the kernel cmdline. Right now setting one resets also the other. | |
650 | ||
651 | * and a dbus call to generate target from current state | |
652 | ||
653 | * readahead: | |
654 | - drop /.readahead on bigger upgrades with yum | |
655 | - move readahead files into /var (look for them with .path units?) | |
656 | - readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO | |
657 | - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes | |
658 | - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/ | |
659 | ||
660 | * add support for /bin/mount -s | |
661 | ||
662 | * GC unreferenced jobs (such as .device jobs) | |
663 | ||
664 | * write blog stories about: | |
665 | - hwdb: what belongs into it, lsusb | |
666 | - enabling dbus services | |
667 | - status update | |
668 | - how to make changes to sysctl and sysfs attributes | |
669 | - remote access | |
670 | - how to pass throw-away units to systemd, or dynamically change properties of existing units | |
671 | - how to integrate cgconfig and suchlike with systemd | |
672 | - testing with Harald's awesome test kit | |
673 | - auto-restart | |
674 | - how to develop against journal browsing APIs | |
675 | - the journal HTTP iface | |
676 | - non-cgroup resource management | |
677 | - dynamic resource management with cgroups | |
678 | - refreshed, longer missions statement | |
679 | - calendar time events | |
680 | - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell | |
681 | - how to create your own target | |
682 | - instantiated apache, dovecot and so on | |
683 | - hooking a script into various stages of shutdown/rearly booot | |
684 | ||
685 | * allow port=0 in .socket units | |
686 | ||
687 | * support systemd.mask= on the kernel command line. | |
688 | ||
689 | * recreate systemd's D-Bus private socket file on SIGUSR2 | |
690 | ||
691 | * Support --test based on current system state | |
692 | ||
693 | * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support. | |
694 | ||
695 | * maybe introduce ExecRestartPre= | |
696 | ||
697 | * dot output for --test showing the 'initial transaction' | |
698 | ||
699 | * port over to LISTEN_FDS/LISTEN_PID: | |
700 | - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH | |
701 | - cups HAVEPATCH | |
702 | - postfix, saslauthd | |
703 | - apache/samba | |
704 | - libvirtd (/var/run/libvirt/libvirt-sock-ro) | |
705 | - bluetoothd (/var/run/sdp! @/org/bluez/audio!) | |
706 | - distccd | |
707 | ||
708 | * fingerprint.target, wireless.target, gps.target, netdevice.target | |
709 | ||
710 | * io priority during initialization | |
711 | ||
712 | * drop cap bounding set in readahead and other services | |
713 | ||
714 | * systemd-python: | |
715 | - figure out a simple way to wait for journal events in a way that | |
716 | works with ^C | |
717 | - add documentation to systemd.daemon | |
718 | ||
719 | * bootchart: | |
720 | - plot per-process IO utilization | |
721 | - group processes based on service association (cgroups) | |
722 | - document initcall_debug | |
723 | - kernel cmdline "bootchart" option for simplicity? | |
724 | ||
14228c0d MB |
725 | * systemd-run is missing zsh completion scripts |
726 | ||
663996b3 MS |
727 | External: |
728 | ||
729 | * dbus: | |
730 | - dbus --user | |
731 | - natively watch for dbus-*.service symlinks (PENDING) | |
732 | - allow specification of socket mode/umask when allocating DBusServer | |
733 | - allow disabling of fd passing when connecting a AF_UNIX connection | |
734 | - allow disabling of UID passing for AUTH EXTERNAL | |
735 | - always pass cred data along each message | |
736 | - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service | |
737 | ||
738 | * fix alsa mixer restore to not print error when no config is stored | |
739 | ||
740 | * make cryptsetup lower --iter-time | |
741 | ||
742 | * patch kernel for xattr support in /dev, /proc/, /sys? | |
743 | ||
744 | * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only | |
745 | in an ~30 minutes window. It does not adjust larger differences. Find a way | |
746 | to tell the kernel, to always do a full time sync when the RTC is in UTC and | |
747 | we are in 11-minutes-mode. When we trust the system time to NTP we also want | |
748 | the RTC to sync up. | |
749 | ||
750 | * kernel: add device_type = "fb", "fbcon" to class "graphics" | |
751 | ||
752 | * drop accountsservice's StandardOutput=syslog and Type=dbus fields | |
753 | ||
754 | * dbus upstream still refers to dbus.target and shouldn't | |
755 | ||
756 | * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id | |
757 | ||
758 | * add "# export SYSTEMD_PAGER=" to bash login | |
759 | ||
760 | * /usr/bin/service should actually show the new command line | |
761 | ||
762 | * 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= | |
763 | ||
764 | * fedora: F20: go timer units all the way, leave cron.daily for cron | |
765 | ||
766 | * neither pkexec nor sudo initialize environ[] from the PAM environment? | |
767 | ||
768 | * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it | |
769 | ||
770 | Regularly: | |
771 | ||
772 | * look for close() vs. close_nointr() vs. close_nointr_nofail() | |
773 | ||
774 | * check for strerror(r) instead of strerror(-r) | |
775 | ||
776 | * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel | |
777 | ||
778 | * %m in printf() instead of strerror(errno); | |
779 | ||
780 | * pahole | |
781 | ||
782 | * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()! | |
783 | ||
784 | * use secure_getenv() instead of getenv() where appropriate | |
785 | ||
786 | * link up selected blog stories from man pages and unit files Documentation= fields | |
787 | ||
788 | Scheduled for removal or fixing: | |
789 | ||
790 | * xxxOverridable dependencies (probably: fix) |