]>
Commit | Line | Data |
---|---|---|
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 | * stop importing kernel exported env variables. The utterly broken logic in | |
15 | the kernel exports every kernel command line option which is not recognized | |
16 | as a built-in module option as an env variable. Systemd should not pass-on | |
17 | that nonsense, a kernel command line option is a command line option not an | |
18 | env variable: | |
19 | $ cat /proc/252/environ | |
20 | initrd=\6a9857a393724b7a981ebb5b8495b9ea\3.10.0-2.fc20.x86_64\initrd | |
21 | ||
22 | Fedora 19: | |
23 | ||
24 | * external: maybe it is time to patch procps so that "ps" links to | |
25 | libsystemd-logind to print a pretty service name, seat name, session | |
26 | name in its output. Currently it only shows cgroup membership, but | |
27 | that's sometimes kinda hard to parse for a human. | |
28 | ||
29 | * cgroup attrs: | |
30 | - update dbus interface docs in wiki | |
31 | ||
32 | * localed: | |
33 | - localectl: support new converted x11→console keymaps | |
34 | ||
35 | * when installing fedora with yum --installroot /var/run is a directory, not a symlink | |
36 | ||
37 | Features: | |
38 | ||
39 | * when a kernel driver logs in a tight loop we should ratelimit that too. | |
40 | ||
41 | * journald: when we drop syslog messages because the syslog socket is | |
42 | full, make sure to write how many messages are lost as first thing | |
43 | to syslog when it works again. | |
44 | ||
45 | * prohibit Restart= set with Type=oneshot | |
46 | ||
47 | * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted. | |
48 | ||
49 | * load .d/*.conf dropins for device units | |
50 | ||
51 | * user@.service and session-*.scope should get posession of their own cgroups | |
52 | ||
53 | * move systemctl set-log-level to systemd-analyze? | |
54 | ||
55 | * fix killing spree logic in systemd-user-sessions | |
56 | ||
57 | * logind: implement session kill exceptions | |
58 | ||
59 | * fix machine regstration to forward property array | |
60 | ||
61 | * fix loginctl cgroup enumeration | |
62 | ||
63 | * move "systemctl dump" to systemd-analyze | |
64 | ||
65 | * introduce "mainpid" for scopes | |
66 | ||
67 | * add a fixed dbus path for "my own unit", "my own session", ... to PID1, logind, ... | |
68 | ||
69 | * add implicit slice for instantiated services | |
70 | ||
71 | * service_coldplug() appears to reinstall the wrong stop timeout watch? | |
72 | ||
73 | * transient units: allow creating auxiliary units with the same call | |
74 | ||
75 | * make BlockIODeviceWeight=, BlockIODeviceBandwidth= runtime settable | |
76 | ||
77 | * split up BlockIOWeight= and BlockIODeviceWeight= | |
78 | ||
79 | * introduce high-level settings for RT budget, swapiness | |
80 | ||
81 | * how to reset dynamically changed attributes sanely? | |
82 | ||
83 | * when reloading configuration, apply new cgroup configuration | |
84 | ||
85 | * implement system-wide DefaultCPUAccounting=1 switch (and similar for blockio, memory?) | |
86 | ||
87 | * implement per-slice CPUFairScheduling=1 switch | |
88 | ||
89 | * handle jointly mounted controllers correctly | |
90 | ||
91 | * journald: make sure ratelimit is actually really per-service with the new cgroup changes | |
92 | ||
93 | * libsystemd-logind: sd_session_is_active() and friends: verify | |
94 | validity of session name before appending it to a path | |
95 | ||
96 | * gparted needs to disable auto-activation of mount units somehow, or | |
97 | maybe we should stop doing auto-activiation of this after boot | |
98 | entirely. https://bugzilla.gnome.org/show_bug.cgi?id=701676 | |
99 | ||
100 | * when a service changes state make reflect that in the | |
101 | RUNNING/LISTENING states of its socket | |
102 | ||
103 | * when recursively showing the cgroup hierarchy, optionally also show | |
104 | the hierarchies of child processes | |
105 | ||
106 | * document logic of auto/noauto and fail/nofail in fstab in systemd.mount or systemd-fstab-generator man page | |
107 | ||
108 | * we should properly escape ssh hostnames we add into dbus server strings | |
109 | ||
110 | * something pulls in pcre as so dep into our daemons such as hostnamed. | |
111 | ||
112 | * cgroup-agent: downgrade error messages | |
113 | ||
114 | * document systemd-journal-flush.service properly | |
115 | ||
116 | * chane systemd-journal-flush into a service that stays around during | |
117 | boot, and causes the journal to be moved back to /run on shutdown, | |
118 | so that we don't keep /var busy. This needs to happen synchronously, | |
119 | hence doing this via signals is not going to work. | |
120 | ||
121 | * allow implementation of InaccessibleDirectories=/ plus | |
122 | ReadOnlyDirectories=... for whitelisting files for a service. | |
123 | ||
124 | * libsystemd-journal: | |
125 | - return ECHILD as soon as somebody tries to reuse a journal object across a fork() | |
126 | ||
127 | * libsystemd-bus: | |
128 | - default policy (allow uid == 0 and our own uid) | |
129 | - enforce alignment of pointers passed in | |
130 | - when kdbus doesn't take our message without memfds, try again with memfds | |
131 | - implement translator service | |
132 | - port systemd to new library | |
133 | - implement busname unit type in systemd | |
134 | - move to gvariant | |
135 | - merge busctl into systemctl or so? | |
136 | - synthesize sd_bus_message objects from kernel messages | |
137 | - properly implement name registry ioctls for kdbus | |
138 | - get rid of object hash table, use decision tree everyhwere instead? | |
139 | - implement monitor logic | |
140 | - object vtable logic | |
141 | - longer term: | |
142 | * priority queues | |
143 | * priority inheritance | |
144 | ||
145 | * in the final killing spree, detect processes from the root directory, and | |
146 | complain loudly if they have argv[0][0] == '@' set. | |
147 | https://bugzilla.redhat.com/show_bug.cgi?id=961044 | |
148 | ||
149 | * add an option to nspawn that uses seccomp to make socket(AF_NETLINK, | |
150 | SOCK_RAW, NETLINK_AUDIT) fail the the appropriate error code that | |
151 | makes the audit userspace to think auditing is not available in the | |
152 | kernel. | |
153 | ||
154 | * Introduce a way how we can kill the main process of a service with KillSignal, but all processes with SIGKILL later on | |
155 | https://bugzilla.redhat.com/show_bug.cgi?id=952634 | |
156 | ||
157 | * maybe add a warning to the unit file parses whern the acces mode of unit files is non-sensical. | |
158 | ||
159 | * investigate endianess issues of UUID vs. GUID | |
160 | ||
161 | * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we | |
162 | should be able to safely try another attempt when the bus call LoadUnit() is invoked. | |
163 | ||
164 | * if pam_systemd is invoked by su from a process that is outside of a | |
165 | any session we should probably just become a NOP, since that's | |
166 | usually not a real user session but just some system code that just | |
167 | needs setuid(). | |
168 | ||
169 | * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login. | |
170 | ||
171 | * add a pam module that on password changes updates any LUKS slot where the password matches | |
172 | ||
173 | * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases... | |
174 | ||
175 | * timedatctl, localectl: possibly make some commands work without the daemon, for chroot situations... | |
176 | ||
177 | * logind: add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly. | |
178 | ||
179 | * we need dynamic units | |
180 | ||
181 | * cgtop: make cgtop useful in a container | |
182 | ||
183 | * test/: | |
184 | - add 'set -e' to scripts in test/ | |
185 | - make stuff in test/ work with separate output dir | |
186 | - remove all the duplicated code in test/ | |
187 | ||
188 | * suppress log output on shutdown when "quiet" is used | |
189 | ||
190 | * systemctl delete x.snapshot leaves no trace in logs (at least at default level). | |
191 | ||
192 | * make the coredump collector tool move itself into the user's cgroup | |
193 | so that the coredump is properly written to the user's own journal | |
194 | file. | |
195 | ||
196 | * seems that when we follow symlinks to units we prefer the symlink | |
197 | destination path over /etc and /usr. We shouldn't do that. Instead | |
198 | /etc should always override /run+/usr and also any symlink | |
199 | destination. | |
200 | ||
201 | * remove duplicate default deps logic from fstab-generator vs. mount.c | |
202 | ||
203 | * when isolating, try to figure out a way how we implicitly can order | |
204 | all units we stop before the isolating unit... | |
205 | ||
206 | * teach udev + logind's uaccess to somehow handle the "dead" device nodes from: | |
207 | /lib/modules/$(uname -r)/modules.devname | |
208 | and apply ACLs to them if they have TAG=="uaccess" in udev rules. | |
209 | ||
210 | * add ConditionArchitecture= or so | |
211 | ||
212 | * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off}) | |
213 | ||
214 | * we should log capabilities too | |
215 | ||
216 | * Support SO_REUSEPORT with socket activation: | |
217 | - Let systemd maintain a pool of servers. | |
218 | - Use for seamless upgrades, by running the new server before stopping the | |
219 | old. | |
220 | ||
221 | * after all byte-wise realloc() is slow, even on glibc, so i guess we | |
222 | need manual exponential loops after all | |
223 | ||
224 | * BootLoaderSpec: drop allowing ext234 for $BOOT. Clarify that the | |
225 | kernel has to be in $BOOT. Clarify that the boot loader should be | |
226 | installed to the ESP. Define a way how an installer can figure out | |
227 | whether a BLS compliant boot loader is installed. | |
228 | ||
229 | * man: remove .include documentation, and instead push people to use .d/*.conf | |
230 | ||
231 | * think about requeuing jobs when daemon-reload is issued? usecase: | |
232 | the initrd issues a reload after fstab from the host is accessible | |
233 | and we might want to requeue the mounts local-fs acquired through | |
234 | that automatically. | |
235 | ||
236 | * rework specifier logic so that we can distuingish OOM errors from other errors | |
237 | ||
238 | * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep() | |
239 | ||
240 | * journal-or-kmsg is currently broken? See reverted commit 4a01181e460686d8b4a543b1dfa7f77c9e3c5ab8. | |
241 | ||
242 | * remove any syslog support from log.c -- we probably can't do this before split-off udev is gone for good | |
243 | ||
244 | * documentation: recommend to connect the timer units of a service to the service via Also= in [Install] | |
245 | ||
246 | * add a tool that lists active timer units plus their next elapstion and the time the units ran last | |
247 | ||
248 | * man: document the very specific env the shutdown drop-in tools live in | |
249 | ||
250 | * shutdown logging: store to EFI var, and store to USB stick? | |
251 | ||
252 | * man: extend runlevel(8) to mention that runlevels suck, and are dead. Maybe add runlevel(7) with a note about that too | |
253 | ||
254 | * systemctl: maybe add "systemctl add-wants" or so... | |
255 | ||
256 | * man: add more examples to man pages | |
257 | ||
258 | * man: maybe sort directives in man pages, and take sections from --help and apply them to man too | |
259 | ||
260 | * write UI tool that pops up emergency messages from the journal as notification | |
261 | ||
262 | * think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service | |
263 | ||
264 | * msgcatalog determine default language of entries from the catalog | |
265 | file name. i.e. foobar.de_DE.catalog would set the default entry | |
266 | name for the entries to de_DE if they aren't explicitly suffixed | |
267 | individually. | |
268 | ||
269 | * use "log level" rather than "log priority" everywhere | |
270 | ||
271 | * timedate: have global on/off switches for auto-time (NTP), and auto-timezone that connman can subscribe to. | |
272 | ||
273 | * Honour "-" prefix for InaccessibleDirectories= and ReadOnlyDirectories= to | |
274 | suppress errors of the specified path doesn't exist | |
275 | ||
276 | * dev-setup.c: when running in a container, create a tiny stub udev | |
277 | database with the systemd tag set for all network interfaces found, | |
278 | so that libudev reports them as present, and systemd's .device units | |
279 | will be activated. | |
280 | ||
281 | * merge unit_kill_common() and unit_kill_context() | |
282 | ||
283 | * introduce ExecCondition= in services | |
284 | ||
285 | * EFI: | |
286 | - fsck hookup for the ESP mount is missing | |
287 | - write man page for efi boot generator | |
288 | - honor language efi variables for default language selection (if there are any?) | |
289 | - honor timezone efi variables for default timezone selection (if there are any?) | |
290 | - introduce bootctl (backed by systemd-bootd) to control temporary and persistent default boot goal plus efi variables | |
291 | ||
292 | * maybe do not install getty@tty1.service symlink in /etc but in /usr? | |
293 | ||
294 | * re-enable "make check" for gtk-doc (broken for unknown reason) | |
295 | ||
296 | * fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points | |
297 | https://bugzilla.redhat.com/show_bug.cgi?id=812826 | |
298 | ||
299 | * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word | |
300 | ||
301 | * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units. | |
302 | ||
303 | * logind: | |
304 | - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around | |
305 | - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot | |
306 | - logind: wakelock/opportunistic suspend support | |
307 | - Add pretty name for seats in logind | |
308 | - logind: allow showing logout dialog from system? | |
309 | - logind: spawn user@..service on login | |
310 | - logind: non-local X11 server handling | |
311 | - logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API | |
312 | - pam: when leaving a session explicitly exclude the ReleaseSession() caller process from the killing spree | |
313 | - logind: GetSessionByPID() should accept 0 as PID value | |
314 | - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case | |
315 | ||
316 | * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty. | |
317 | ||
318 | * DeviceAllow/DeviceDeny: disallow everything by default, but whitelist /dev/zero, /dev/null and friends | |
319 | ||
320 | * service: watchdog logic: for testing purposes allow ping, but do not require pong | |
321 | ||
322 | * journal: | |
323 | - do not use magic msync() in src/journal/journal-file.c, just call fsync() | |
324 | <alxchk> poettering: looks like msync just calls vfs_fsync | |
325 | http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/mm/msync.c#n18 | |
326 | - import and delete pstore filesystem content at startup | |
327 | - journald: also get thread ID from client, plus thread name | |
328 | - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups | |
329 | - add API to close/reopen/get fd for journal client fd in libsystemd-journal. | |
330 | - fallback to /dev/log based logging in libsystemd-journal, if we can't log natively? | |
331 | - declare the local journal protocol stable in the wiki interface chart | |
332 | - journal: reuse XZ context | |
333 | - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg | |
334 | - journald: when dropping msgs due to ratelimit make sure to write | |
335 | "dropped %u messages" not only when we are about to print the next | |
336 | message that works, but alraedy after a short tiemout | |
337 | - check if we can make journalctl by default use --follow mode inside of less if called without args? | |
338 | - maybe add API to send pairs of iovecs via sd_journal_send | |
339 | - journal: when writing journal auto-rotate if time jumps backwards | |
340 | - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access | |
341 | - journactl: support negative filtering, i.e. FOOBAR!="waldo", | |
342 | and !FOOBAR for events without FOOBAR. | |
343 | - journal: when rotating, copy over old acls/access mode | |
344 | - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs. | |
345 | - 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. | |
346 | - journal: find a way to allow dropping history early, based on priority, other rules | |
347 | - journal: When used on NFS, check payload hashes | |
348 | - Introduce journalctl -b <nr> to show journal messages of a previous boot | |
349 | - journald: check whether it is OK if the client can still modify delivered journal entries | |
350 | - journal live copy, based on libneon (client) and libmicrohttpd (server) | |
351 | - journald: add kernel cmdline option to disable ratelimiting for debug purposes | |
352 | - refuse taking lower-case variable names in sd_journal_send() and friends. | |
353 | - journald: we currently rotate only after MaxUse+MaxFilesize has been reached. | |
354 | - journal: deal nicely with byte-by-byte copied files, especially regards header | |
355 | - journalctl: show multiline log messages sanely, expand tabs, and show all valid utf8 messages | |
356 | - journal: store euid in journal if it differs from uid | |
357 | - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit | |
358 | - Replace utmp, wtmp, btmp, and lastlog completely with journal | |
359 | - Port upower to use the journal for historical power information used in future calculations | |
360 | ||
361 | * document: | |
362 | - document unit_name_mangle() | |
363 | - document that deps in [Unit] sections ignore Alias= fileds in | |
364 | [Install] units of other units, unless those units are disabled | |
365 | - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets | |
366 | - Document word splitting syntax for ExecStart= and friends | |
367 | - document that units from /etc override those from /usr and /run | |
368 | - document the exit codes when services fail before they are exec()ed | |
369 | - document that service reload may be implemented as service reexec | |
370 | - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications | |
371 | ||
372 | * systemctl: | |
373 | - systemctl list-jobs - show dependencies | |
374 | - add systemctl switch to dump transaction without executing it | |
375 | - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done | |
376 | - "systemctl disable" on a static unit prints no message and does | |
377 | nothing. "systemctl enable" does nothing, and gives a bad message | |
378 | about it. Should fix both to print nice actionable messages. | |
379 | - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service | |
380 | - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible | |
381 | - systemctl enable: improve the success messages (i.e. more human readable, less shell-like) | |
382 | - systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards? | |
383 | - systemctl: "Journal has been rotated since unit was started." message is misleading | |
384 | - support "systemctl stop foobar@.service" to stop all units matching a certain template | |
385 | - Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files" | |
386 | ||
387 | * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations | |
388 | ||
389 | * deal with sendmail/postfix exclusivity | |
390 | ||
391 | * timer units: | |
392 | - configurable jitter for timer events | |
393 | - timer events with system resume | |
394 | - timer units should get the ability to trigger when: | |
395 | o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET) | |
396 | o DST changes | |
397 | - Support 2012-02~4 as syntax for specifying the fourth to last day of the month. | |
398 | ||
399 | * update the kernel's TZ (sys_tz) when DST changes | |
400 | ||
401 | * sync down the system time to the RTC when: | |
402 | - CLOCK_REALTIME makes jumps (the user explicitely requested a time set) | |
403 | - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp) | |
404 | This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime | |
405 | mode, it will never touch the RTC if the no reliable time source is active or the | |
406 | user did not request anything like it. | |
407 | ||
408 | * hwdb: | |
409 | - implement conditional properties (dmi matches) | |
410 | - hwdb --filter=ID_DRIVE_* | |
411 | ||
412 | * if booted in "quiet" mode, and an error happens, turn on status | |
413 | output again, so that the emergency mode isn't totally | |
414 | surprising. Also, terminate plymouth. | |
415 | ||
416 | * libunwind support for coredump pattern hook, and includes this in | |
417 | the message for coredumps. After all, libunwind is now capable to | |
418 | unwind coredumps since a few weeks ago. This probably requires that | |
419 | we have nice support for multi-line messages on display in | |
420 | logs-show.c. Alternatively: use libelfutil, which seems to be the | |
421 | better supported alternative. | |
422 | ||
423 | * figure out relation of --all and --full in the various tools | |
424 | ||
425 | * add libsystemd-password or so to query passwords during boot using the password agent logic | |
426 | ||
427 | * 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(). | |
428 | ||
429 | * fedup: add --unit to systemctl switch-root somehow | |
430 | * fedup: don't delete initrd on switch-root | |
431 | * fedup: generator | |
432 | ||
433 | * timedated: | |
434 | - timedated: refuse time changes when NTP is on | |
435 | ||
436 | * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed | |
437 | ||
438 | * introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields | |
439 | ||
440 | * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel | |
441 | ||
442 | * add "provisioning" instructions to setup an empty /etc + /var | |
443 | - used to setup a new container from a shared /usr | |
444 | - superset of tmpfiles model | |
445 | - instructions shipped by packages and stored in /usr/lib/ | |
446 | - compose /etc/passwd and /etc/group, copy files | |
447 | - able to create uid + gid used by packages, for file ownership | |
448 | ||
449 | * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without | |
450 | waiting for the timeout | |
451 | ||
452 | * high level net_prio setting in execution context | |
453 | ||
454 | * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ... | |
455 | ||
456 | * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't | |
457 | ||
458 | * refuse boot if /etc/machine-id is not useful (or set taint?) | |
459 | ||
460 | * nspawn: | |
461 | - nspawn: consider changing users for -u with su, so that NSS resolving works correctly | |
462 | - nspawn: implement personality changes a la linux32(8) | |
463 | - nspawn: --read-only is not applied recursively to submounts | |
464 | - nspawn: make use of device cgroup controller by default | |
465 | - bind mount read-only the cgroup tree higher than nspawn | |
466 | - nspawn: investigate whether we can support the same as LXC's | |
467 | lxc.network.type=phys mode, and pass through entire network | |
468 | interfaces to the container | |
469 | - nspawn: maybe add a way to drop additional caps, in addition to add additional caps | |
470 | - nspawn: maybe explicitly reset loginuid? | |
471 | - nspawn: make it work for dwalsh and shared /usr containers -- tmpfs mounts as command line parameters, selinux exec context | |
472 | ||
473 | * cryptsetup: | |
474 | - cryptsetup-generator: allow specification of passwords in crypttab itself | |
475 | - move cryptsetup key caching into kernel keyctl? | |
476 | https://bugs.freedesktop.org/show_bug.cgi?id=54982 | |
477 | - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator | |
478 | ||
479 | * move debug shell to tty6 and make sure this doesn't break the gettys on tty6 | |
480 | ||
481 | * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it | |
482 | https://bugs.freedesktop.org/show_bug.cgi?id=54712 | |
483 | ||
484 | * after deserializing sockets in socket.c we should reapply sockopts and things | |
485 | ||
486 | * make timer units go away after they elapsed | |
487 | ||
488 | * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html | |
489 | (network and remote-fs on shutdown) | |
490 | ||
491 | * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead | |
492 | ||
493 | * move PID 1 segfaults to /var/lib/systemd/coredump? | |
494 | ||
495 | * create /sbin/init symlinks from the build system | |
496 | ||
497 | * Query Paul Moore about relabelling socket fds while they are open | |
498 | ||
499 | * system.conf should have controls for cgroups | |
500 | ||
501 | * allow writing multiple conditions in unit files on one line | |
502 | ||
503 | * explore multiple service instances per listening socket idea | |
504 | ||
505 | * shutdown: don't read-only mount anything when running in container | |
506 | ||
507 | * MountFlags=shared acts as MountFlags=slave right now. | |
508 | ||
509 | * ReadOnlyDirectories= is not applied recursively to submounts | |
510 | ||
511 | * drop PID 1 reloading, only do reexecing (difficult: Reload() | |
512 | currently is properly synchronous, Reexec() is weird, because we | |
513 | can't delay the response properly until we are back, so instead of | |
514 | being properly synchronous we just keep open the fd and close it | |
515 | when done. That means clients don't get a successful method reply, | |
516 | but much rather a disconnect on success. | |
517 | ||
518 | * remember which condition failed for services, not just the fact that something failed | |
519 | ||
520 | * use opterr = 0 for all getopt tools | |
521 | ||
522 | * properly handle loop back mounts via fstab, especially regards to fsck/passno | |
523 | ||
524 | * allow services with no ExecStart= but with an ExecStop= | |
525 | ||
526 | * dracut-shutdown needs to be ordered before unmounting /boot | |
527 | ||
528 | * initialize the hostname from the fs label of /, if /etc/hostname does not exist? | |
529 | ||
530 | * rename "userspace" to "core-os" | |
531 | ||
532 | * syscall filter: | |
533 | - syscall filter: add knowledge about compat syscalls | |
534 | - syscall filter: don't enforce no new privs? | |
535 | - syscall filter: option to return EPERM rather than SIGSYS? | |
536 | - syscall filter: port to libseccomp | |
537 | - system-wide seccomp filter | |
538 | ||
539 | * .device aliases need to be implemented with the "following" logic, probably. | |
540 | ||
541 | * load-fragment: when loading a unit file via a chain of symlinks | |
542 | verify that it isn't masked via any of the names traversed. | |
543 | ||
544 | * introduce Type=pid-file | |
545 | ||
546 | * change Requires=basic.target to RequisiteOverride=basic.target | |
547 | ||
548 | * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr | |
549 | ||
550 | * move passno parsing to fstab generator | |
551 | ||
552 | * automount: implement expire: | |
553 | - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD | |
554 | - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD | |
555 | - every timeout/4 (original autofs logic) | |
556 | - blocking, needs a thread | |
557 | - run until -EAGAIN | |
558 | - receive expire packet on pipe if kernel tells the timeout is over | |
559 | - call umount | |
560 | - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD | |
561 | - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns | |
562 | ||
563 | * services which create their own subcgroups break cgroup-empty notification (needs to be fixed in the kernel) | |
564 | ||
565 | * ExecOnFailure=/usr/bin/foo | |
566 | ||
567 | * udev: | |
568 | - remove src/udev/udev-builtin-firmware.c (CONFIG_FW_LOADER_USER_HELPER=n) | |
569 | - move to LGPL | |
570 | - unify utf8 validator code with shared/ | |
571 | - kill scsi_id | |
572 | - add trigger --subsystem-match=usb/usb_device device | |
573 | ||
574 | * cleanup syslog 'priority' vs. 'level' wording | |
575 | ||
576 | * 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 | |
577 | ||
578 | * support container_ttys= | |
579 | ||
580 | * introduce mix of BindTo and Requisite | |
581 | ||
582 | * add DeleteSocketsOnStop=yes|no option to socket units | |
583 | ||
584 | * There's currently no way to cancel fsck (used to be possible via C-c or c on the console) | |
585 | ||
586 | * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/ | |
587 | ||
588 | * default unix qlen is too small (10). bump sysctl? add sockopt? | |
589 | ||
590 | * save coredump in Windows/Mozilla minidump format | |
591 | ||
592 | * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting) | |
593 | ||
594 | * clean up session cgroups that remain after logout (think sshd), but eventually run empty | |
595 | ||
596 | * when an instanced service exits, remove its parent cgroup too if possible. | |
597 | ||
598 | * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max | |
599 | ||
600 | * be able to specify a forced restart of service A where service B depends on, in case B | |
601 | needs to be auto-respawned? | |
602 | ||
603 | * when a bus name of a service disappears from the bus make sure to queue further activation requests | |
604 | ||
605 | * tmpfiles: apply "x" on "D" too (see patch from William Douglas) | |
606 | ||
607 | * for services: don't set $HOME in services unless requested | |
608 | ||
609 | * hide PAM/TCPWrap options in fragment parser when compile time disabled | |
610 | ||
611 | * when we automatically restart a service, ensure we restart its rdeps, too. | |
612 | ||
613 | * allow Type=simple with PIDFile= | |
614 | https://bugzilla.redhat.com/show_bug.cgi?id=723942 | |
615 | ||
616 | * move PAM code into its own binary | |
617 | ||
618 | * implement Register= switch in .socket units to enable registration | |
619 | in Avahi, RPC and other socket registration services. | |
620 | ||
621 | * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early | |
622 | ||
623 | * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination | |
624 | ||
625 | * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1) | |
626 | ||
627 | * verify that the AF_UNIX sockets of a service in the fs still exist | |
628 | when we start a service in order to avoid confusion when a user | |
629 | assumes starting a service is enough to make it accessible | |
630 | ||
631 | * Make it possible to set the keymap independently from the font on | |
632 | the kernel cmdline. Right now setting one resets also the other. | |
633 | ||
634 | * and a dbus call to generate target from current state | |
635 | ||
636 | * readahead: | |
637 | - drop /.readahead on bigger upgrades with yum | |
638 | - move readahead files into /var (look for them with .path units?) | |
639 | - readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO | |
640 | - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes | |
641 | - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/ | |
642 | ||
643 | * add support for /bin/mount -s | |
644 | ||
645 | * GC unreferenced jobs (such as .device jobs) | |
646 | ||
647 | * write blog stories about: | |
648 | - hwdb: what belongs into it, lsusb | |
649 | - enabling dbus services | |
650 | - status update | |
651 | - how to make changes to sysctl and sysfs attributes | |
652 | - remote access | |
653 | - how to pass throw-away units to systemd, or dynamically change properties of existing units | |
654 | - how to integrate cgconfig and suchlike with systemd | |
655 | - testing with Harald's awesome test kit | |
656 | - auto-restart | |
657 | - how to develop against journal browsing APIs | |
658 | - the journal HTTP iface | |
659 | - non-cgroup resource management | |
660 | - dynamic resource management with cgroups | |
661 | - refreshed, longer missions statement | |
662 | - calendar time events | |
663 | - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell | |
664 | - how to create your own target | |
665 | - instantiated apache, dovecot and so on | |
666 | - hooking a script into various stages of shutdown/rearly booot | |
667 | ||
668 | * allow port=0 in .socket units | |
669 | ||
670 | * support systemd.mask= on the kernel command line. | |
671 | ||
672 | * recreate systemd's D-Bus private socket file on SIGUSR2 | |
673 | ||
674 | * Support --test based on current system state | |
675 | ||
676 | * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support. | |
677 | ||
678 | * maybe introduce ExecRestartPre= | |
679 | ||
680 | * dot output for --test showing the 'initial transaction' | |
681 | ||
682 | * port over to LISTEN_FDS/LISTEN_PID: | |
683 | - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH | |
684 | - cups HAVEPATCH | |
685 | - postfix, saslauthd | |
686 | - apache/samba | |
687 | - libvirtd (/var/run/libvirt/libvirt-sock-ro) | |
688 | - bluetoothd (/var/run/sdp! @/org/bluez/audio!) | |
689 | - distccd | |
690 | ||
691 | * fingerprint.target, wireless.target, gps.target, netdevice.target | |
692 | ||
693 | * io priority during initialization | |
694 | ||
695 | * drop cap bounding set in readahead and other services | |
696 | ||
697 | * systemd-python: | |
698 | - figure out a simple way to wait for journal events in a way that | |
699 | works with ^C | |
700 | - add documentation to systemd.daemon | |
701 | ||
702 | * bootchart: | |
703 | - plot per-process IO utilization | |
704 | - group processes based on service association (cgroups) | |
705 | - document initcall_debug | |
706 | - kernel cmdline "bootchart" option for simplicity? | |
707 | ||
708 | External: | |
709 | ||
710 | * dbus: | |
711 | - dbus --user | |
712 | - natively watch for dbus-*.service symlinks (PENDING) | |
713 | - allow specification of socket mode/umask when allocating DBusServer | |
714 | - allow disabling of fd passing when connecting a AF_UNIX connection | |
715 | - allow disabling of UID passing for AUTH EXTERNAL | |
716 | - always pass cred data along each message | |
717 | - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service | |
718 | ||
719 | * fix alsa mixer restore to not print error when no config is stored | |
720 | ||
721 | * make cryptsetup lower --iter-time | |
722 | ||
723 | * patch kernel for xattr support in /dev, /proc/, /sys? | |
724 | ||
725 | * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only | |
726 | in an ~30 minutes window. It does not adjust larger differences. Find a way | |
727 | to tell the kernel, to always do a full time sync when the RTC is in UTC and | |
728 | we are in 11-minutes-mode. When we trust the system time to NTP we also want | |
729 | the RTC to sync up. | |
730 | ||
731 | * kernel: add device_type = "fb", "fbcon" to class "graphics" | |
732 | ||
733 | * drop accountsservice's StandardOutput=syslog and Type=dbus fields | |
734 | ||
735 | * dbus upstream still refers to dbus.target and shouldn't | |
736 | ||
737 | * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id | |
738 | ||
739 | * add "# export SYSTEMD_PAGER=" to bash login | |
740 | ||
741 | * /usr/bin/service should actually show the new command line | |
742 | ||
743 | * 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= | |
744 | ||
745 | * fedora: F20: go timer units all the way, leave cron.daily for cron | |
746 | ||
747 | * neither pkexec nor sudo initialize environ[] from the PAM environment? | |
748 | ||
749 | * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it | |
750 | ||
751 | Regularly: | |
752 | ||
753 | * look for close() vs. close_nointr() vs. close_nointr_nofail() | |
754 | ||
755 | * check for strerror(r) instead of strerror(-r) | |
756 | ||
757 | * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel | |
758 | ||
759 | * %m in printf() instead of strerror(errno); | |
760 | ||
761 | * pahole | |
762 | ||
763 | * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()! | |
764 | ||
765 | * use secure_getenv() instead of getenv() where appropriate | |
766 | ||
767 | * link up selected blog stories from man pages and unit files Documentation= fields | |
768 | ||
769 | Scheduled for removal or fixing: | |
770 | ||
771 | * xxxOverridable dependencies (probably: fix) |