]>
Commit | Line | Data |
---|---|---|
1 | Bugfixes: | |
2 | * enabling an instance unit creates a pointless link, and | |
3 | the unit will be started with getty@getty.service: | |
4 | $ systemctl enable getty@.service | |
5 | ln -s '/usr/lib/systemd/system/getty@.service' '/etc/systemd/system/getty.target.wants/getty@.service' | |
6 | ||
7 | * Dangling symlinks of .automount unit files in .wants/ directories, set up | |
8 | automount points even when the original .automount file did not exist | |
9 | anymore. Only the .mount unit was still around. | |
10 | ||
11 | * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point. | |
12 | ||
13 | * shorten the message to sane length: | |
14 | ||
15 | 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. | |
16 | ||
17 | * systemctl --root=container/ set-default ... is totally borked. | |
18 | ||
19 | * sd_bus_unref() is broken regarding self-references and "pseudo thread-safety". | |
20 | See the comment in sd_bus_unref() for more.. | |
21 | ||
22 | External: | |
23 | ||
24 | * Fedora: when installing fedora with yum --installroot /var/run is a directory, not a symlink | |
25 | https://bugzilla.redhat.com/show_bug.cgi?id=975864 | |
26 | ||
27 | * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros. | |
28 | ||
29 | * Fedora: post FPC ticket to move add %tmpfiles_create to the packaging guidelines | |
30 | ||
31 | Features: | |
32 | ||
33 | * add bus api to query unit file's X fields. | |
34 | ||
35 | * consider adding RuntimeDirectoryUser= + RuntimeDirectoryGroup= | |
36 | ||
37 | * sd-event: define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ... | |
38 | ||
39 | * gpt-auto-generator: | |
40 | - Support LUKS for root devices | |
41 | - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap? | |
42 | - Make /home automount rather than mount? | |
43 | ||
44 | * improve journalctl performance by loading journal files | |
45 | lazily. Encode just enough information in the file name, so that we | |
46 | don't have to open it to know that it is not interesting for us, for | |
47 | the most common operations. | |
48 | ||
49 | * support transient mount units | |
50 | ||
51 | * add an "input" group to udev logic and add all input devices to it | |
52 | ||
53 | * machined: provide calls GetMachineByAddress() on the manager | |
54 | interface to get the machine for a local IP address, and | |
55 | GetAddress() on the Machine interface to get the Address for a | |
56 | machine. Implement via forking off child process which quickly joins | |
57 | the cotnainer and passes data to parent. Show this in "machinectl | |
58 | status", and use it to implement NSS module to provide automatic | |
59 | name resolution for containers. | |
60 | ||
61 | * add generator that pulls in systemd-network from containers when | |
62 | CAP_NET_ADMIN is set, more than the loopback device is defined, even | |
63 | when it is otherwise off | |
64 | ||
65 | * MessageQueueMessageSize= and RLimitFSIZE= (and suchlike) should use parse_iec_size(). | |
66 | ||
67 | * "busctl status" works only as root on dbus1, since we cannot read | |
68 | /proc/$PID/exe | |
69 | ||
70 | * systemctl: support --recursive for list-sockets, list-timers, | |
71 | ... too, not just for list-units. | |
72 | ||
73 | * implement Distribute= in socket units to allow running multiple | |
74 | service instances processing the listening socket, and open this up | |
75 | for ReusePort= | |
76 | ||
77 | * socket units: support creating sockets in different namespace, | |
78 | opening it up for JoinsNamespaceOf=. This would require to fork off | |
79 | a tiny process that joins the namespace and creates/binds the socket | |
80 | and passes this back to PID1 via SCM_RIGHTS. This also could be used | |
81 | to allow Chown/chgrp on sockets without requiring NSS in PID 1. | |
82 | ||
83 | * New service property: maximum CPU and wallclock runtime for a service | |
84 | ||
85 | * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze | |
86 | $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls | |
87 | should SIGSTOP all unit processes in a loop until all processes of | |
88 | it are fully stopped. This can later be used for app management by | |
89 | desktop UIs such as gnome-shell to freeze apps that are not visible | |
90 | on screen, not unlike how job control works on the shell | |
91 | ||
92 | * completions: | |
93 | - busctl zsh completion is outdated | |
94 | - systemd-nspawn -Z/-L/-q is missing for zsh | |
95 | - manager property enumeration was broken when systemd moved to /usr/lib/ | |
96 | ||
97 | * cgroups: | |
98 | - implement per-slice CPUFairScheduling=1 switch | |
99 | - handle jointly mounted controllers correctly | |
100 | - introduce high-level settings for RT budget, swappiness | |
101 | - how to reset dynamically changed unit cgroup attributes sanely? | |
102 | - when reloading configuration, apply new cgroup configuration | |
103 | - when recursively showing the cgroup hierarchy, optionally also show | |
104 | the hierarchies of child processes | |
105 | ||
106 | * transient units: | |
107 | - allow creating auxiliary units with the same call | |
108 | - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt | |
109 | - ensure scope units may be started only a single time | |
110 | ||
111 | * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk? | |
112 | ||
113 | * when we detect low battery and no AC on boot, show pretty splash and refuse boot | |
114 | ||
115 | * machined, localed: when we try to kill an empty cgroup, generate an ESRCH error over the bus | |
116 | ||
117 | * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops | |
118 | ||
119 | * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1 | |
120 | ||
121 | * unify dispatch table in systemctl_main() and friends | |
122 | ||
123 | * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk? | |
124 | ||
125 | * socket-proxyd: Use sd-resolve to resolve the server address | |
126 | ||
127 | * rfkill,backlight: we probably should run the load tools inside of the udev rules so that the state is properly initialized by the time other software sees it | |
128 | ||
129 | * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs | |
130 | ||
131 | * If we try to find a unit via a dangling symlink, generate a clean | |
132 | error. Currently, we just ignore it and read the unit from the search | |
133 | path anyway. | |
134 | ||
135 | * refuse boot if /etc/os-release is missing or /etc/machine-id cannot be set up | |
136 | ||
137 | * given that logind now lets PID 1 do all nasty work, we can | |
138 | probably reduce the capability set it retains substantially. | |
139 | (we need CAP_SYS_ADMIN for drmSetMaster(), so maybe not worth it) | |
140 | ||
141 | * btrfs raid assembly: some .device jobs stay stuck in the queue | |
142 | ||
143 | * make sure gdm doesn't use multi-user-x but the new default X configuration file, and then remove multi-user-x from systemd | |
144 | ||
145 | * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted. | |
146 | ||
147 | * load .d/*.conf dropins for device units | |
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 | * allow implementation of InaccessibleDirectories=/ plus | |
156 | ReadOnlyDirectories=... for whitelisting files for a service. | |
157 | ||
158 | * sd-bus: | |
159 | - when kdbus doesn't take our message without memfds, try again with memfds | |
160 | - systemd-bus-proxyd needs to enforce good old XML policy | |
161 | - port exit-on-idle logic to byebye ioctl | |
162 | - allow updating attach flags during runtime | |
163 | - pid1: peek into activating message when activating a service | |
164 | - introduce sd_bus_emit_object_added()/sd_bus_emit_object_removed() that automatically includes the build-in interfaces in the list | |
165 | - constructors for bus messages should probably not be OK with a NULL bus pointer | |
166 | - .busname units should not use get_user_creds()/get_cgroup_creds() but instead do NSS only in temporarily forked off child | |
167 | - see if we can drop more message validation on the sending side | |
168 | - add API to clone sd_bus_message objects | |
169 | - kdbus: matches against source or destination pids for an "strace -p"-like feel. Problem: The PID info needs to be available in userspace too... | |
170 | - longer term: priority inheritance | |
171 | - dbus spec updates: | |
172 | - kdbus mapping | |
173 | - NameLost/NameAcquired obsolete | |
174 | - GVariant | |
175 | - "const" properties (posted) | |
176 | - path escaping | |
177 | - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now | |
178 | - test bloom filter generation indexes | |
179 | - port to sd-resolve for connecting to TCP dbus servers | |
180 | ||
181 | * sd-event | |
182 | - allow multiple signal handlers per signal? | |
183 | ||
184 | * in the final killing spree, detect processes from the root directory, and | |
185 | complain loudly if they have argv[0][0] == '@' set. | |
186 | https://bugzilla.redhat.com/show_bug.cgi?id=961044 | |
187 | ||
188 | * investigate endianness issues of UUID vs. GUID | |
189 | ||
190 | * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we | |
191 | should be able to safely try another attempt when the bus call LoadUnit() is invoked. | |
192 | ||
193 | * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login. | |
194 | ||
195 | * add a pam module that on password changes updates any LUKS slot where the password matches | |
196 | ||
197 | * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases... | |
198 | ||
199 | * timedatctl, localectl: possibly make some commands work without the daemon, for chroot situations... | |
200 | ||
201 | * timedatectl: print a nicer message when enabling ntp fails because ntpd/chrony are not installed | |
202 | ||
203 | * cgtop: make cgtop useful in a container | |
204 | ||
205 | * test/: | |
206 | - add 'set -e' to scripts in test/ | |
207 | - make stuff in test/ work with separate output dir | |
208 | ||
209 | * systemctl delete x.snapshot leaves no trace in logs (at least at default level). | |
210 | ||
211 | * make the coredump collector tool move itself into the user's cgroup | |
212 | so that the coredump is properly written to the user's own journal | |
213 | file. | |
214 | ||
215 | * seems that when we follow symlinks to units we prefer the symlink | |
216 | destination path over /etc and /usr. We shouldn't do that. Instead | |
217 | /etc should always override /run+/usr and also any symlink | |
218 | destination. | |
219 | ||
220 | * when isolating, try to figure out a way how we implicitly can order | |
221 | all units we stop before the isolating unit... | |
222 | ||
223 | * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off}) | |
224 | ||
225 | * after all byte-wise realloc() is slow, even on glibc, so i guess we | |
226 | need manual exponential loops after all | |
227 | ||
228 | * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify | |
229 | that the boot loader should be installed to the ESP. Define a way | |
230 | how an installer can figure out whether a BLS compliant boot loader | |
231 | is installed. | |
232 | ||
233 | * think about requeuing jobs when daemon-reload is issued? usecase: | |
234 | the initrd issues a reload after fstab from the host is accessible | |
235 | and we might want to requeue the mounts local-fs acquired through | |
236 | that automatically. | |
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 | * shutdown logging: store to EFI var, and store to USB stick? | |
245 | ||
246 | * write UI tool that pops up emergency messages from the journal as notification | |
247 | ||
248 | * think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service | |
249 | ||
250 | * use "log level" rather than "log priority" everywhere | |
251 | ||
252 | * timedate: have global on/off switches for auto-time (NTP), and auto-timezone that connman can subscribe to. | |
253 | ||
254 | * merge unit_kill_common() and unit_kill_context() | |
255 | ||
256 | * introduce ExecCondition= in services | |
257 | ||
258 | * EFI: | |
259 | - write man page for efi boot generator | |
260 | - honor language efi variables for default language selection (if there are any?) | |
261 | - honor timezone efi variables for default timezone selection (if there are any?) | |
262 | - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables | |
263 | ||
264 | * maybe do not install getty@tty1.service symlink in /etc but in /usr? | |
265 | ||
266 | * re-enable "make check" for gtk-doc (broken for unknown reason) | |
267 | ||
268 | * fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points | |
269 | https://bugzilla.redhat.com/show_bug.cgi?id=812826 | |
270 | ||
271 | * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word | |
272 | ||
273 | * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units. | |
274 | ||
275 | * logind: | |
276 | - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around | |
277 | - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot | |
278 | - logind: wakelock/opportunistic suspend support | |
279 | - Add pretty name for seats in logind | |
280 | - logind: allow showing logout dialog from system? | |
281 | - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case | |
282 | - session scopes/user unit: add RequiresMountsFor for the home directory of the user | |
283 | - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly. | |
284 | - if pam_systemd is invoked by su from a process that is outside of a | |
285 | any session we should probably just become a NOP, since that's | |
286 | usually not a real user session but just some system code that just | |
287 | needs setuid(). | |
288 | ||
289 | * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty. | |
290 | ||
291 | * service: watchdog logic: for testing purposes allow ping, but do not require pong | |
292 | ||
293 | * journal: | |
294 | - import and delete pstore filesystem content at startup | |
295 | - journald: also get thread ID from client, plus thread name | |
296 | - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups | |
297 | - add API to close/reopen/get fd for journal client fd in libsystemd-journal. | |
298 | - fallback to /dev/log based logging in libsystemd-journal, if we can't log natively? | |
299 | - declare the local journal protocol stable in the wiki interface chart | |
300 | - journal: reuse XZ context | |
301 | - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg | |
302 | - journald: when dropping msgs due to ratelimit make sure to write | |
303 | "dropped %u messages" not only when we are about to print the next | |
304 | message that works, but alraedy after a short tiemout | |
305 | - check if we can make journalctl by default use --follow mode inside of less if called without args? | |
306 | - maybe add API to send pairs of iovecs via sd_journal_send | |
307 | - journal: when writing journal auto-rotate if time jumps backwards | |
308 | - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access | |
309 | - journactl: support negative filtering, i.e. FOOBAR!="waldo", | |
310 | and !FOOBAR for events without FOOBAR. | |
311 | - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs. | |
312 | - 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. | |
313 | - journal: find a way to allow dropping history early, based on priority, other rules | |
314 | - journal: When used on NFS, check payload hashes | |
315 | - journald: check whether it is OK if the client can still modify delivered journal entries | |
316 | - journal live copy, based on libneon (client) and libmicrohttpd (server) | |
317 | - journald: add kernel cmdline option to disable ratelimiting for debug purposes | |
318 | - refuse taking lower-case variable names in sd_journal_send() and friends. | |
319 | - journald: we currently rotate only after MaxUse+MaxFilesize has been reached. | |
320 | - journal: deal nicely with byte-by-byte copied files, especially regards header | |
321 | - journal: store euid in journal if it differs from uid | |
322 | - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit | |
323 | - Replace utmp, wtmp, btmp, and lastlog completely with journal | |
324 | - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax? | |
325 | - tmpfiles: when applying ownership to /run/log/journal, also do this for the journal fails contained in it | |
326 | - when a kernel driver logs in a tight loop, we should ratelimit that too. | |
327 | - journald: optionally, log debug messages to /run but everything else to /var | |
328 | - journald: when we drop syslog messages because the syslog socket is | |
329 | full, make sure to write how many messages are lost as first thing | |
330 | to syslog when it works again. | |
331 | - journald: make sure ratelimit is actually really per-service with the new cgroup changes | |
332 | - change systemd-journal-flush into a service that stays around during | |
333 | boot, and causes the journal to be moved back to /run on shutdown, | |
334 | so that we don't keep /var busy. This needs to happen synchronously, | |
335 | hence doing this via signals is not going to work. | |
336 | ||
337 | * document: | |
338 | - document that deps in [Unit] sections ignore Alias= fields in | |
339 | [Install] units of other units, unless those units are disabled | |
340 | - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets | |
341 | - document the exit codes when services fail before they are exec()ed | |
342 | - document that service reload may be implemented as service reexec | |
343 | - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications | |
344 | - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr. | |
345 | - document systemd-journal-flush.service properly | |
346 | - documentation: recommend to connect the timer units of a service to the service via Also= in [Install] | |
347 | - man: document the very specific env the shutdown drop-in tools live in | |
348 | - man: extend runlevel(8) to mention that runlevels suck, and are dead. Maybe add runlevel(7) with a note about that too | |
349 | - man: add more examples to man pages | |
350 | - man: maybe sort directives in man pages, and take sections from --help and apply them to man too | |
351 | ||
352 | * systemctl: | |
353 | - systemctl list-jobs - show dependencies | |
354 | - add systemctl switch to dump transaction without executing it | |
355 | - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done | |
356 | - "systemctl disable" on a static unit prints no message and does | |
357 | nothing. "systemctl enable" does nothing, and gives a bad message | |
358 | about it. Should fix both to print nice actionable messages. | |
359 | - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service | |
360 | - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible | |
361 | - systemctl enable: improve the success messages (i.e. more human readable, less shell-like) | |
362 | - systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards? | |
363 | - systemctl: "Journal has been rotated since unit was started." message is misleading | |
364 | - support "systemctl stop foobar@.service" to stop all units matching a certain template | |
365 | - Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files" | |
366 | - rework wait filter to not require match callback | |
367 | - better error message if you run systemctl without systemd running | |
368 | - systemctl status output should should include list of triggering units and their status | |
369 | ||
370 | * unit install: | |
371 | - "systemctl mask" should find all names by which a unit is accessible | |
372 | (i.e. by scanning for symlinks to it) and link them all to /dev/null | |
373 | - "systemctl disable" of a unit instance removes all symlinks, but should | |
374 | only remove the instance symlink (systemctl disable of a template | |
375 | unit however should remove them all). | |
376 | - systemctl list-unit-files should list generated files (and probably with a new state "generated" for them, or so) | |
377 | - systemctl: maybe add "systemctl add-wants" or so... | |
378 | ||
379 | * introduce ntp.service (or suchlike) as symlink that is used to arbitrate between various NTP implementations | |
380 | ||
381 | * deal with sendmail/postfix exclusivity | |
382 | ||
383 | * timer units: | |
384 | - timer units should get the ability to trigger when: | |
385 | o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET) | |
386 | o DST changes | |
387 | - Support 2012-02~4 as syntax for specifying the fourth to last day of the month. | |
388 | - calendarspec: support value ranges with ".." notation. Example: 2013-4..8-1 | |
389 | - 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 | |
390 | - Modulate timer frequency based on battery state | |
391 | ||
392 | * update the kernel's TZ (sys_tz) when DST changes | |
393 | ||
394 | * sync down the system time to the RTC when: | |
395 | - CLOCK_REALTIME makes jumps (the user explicitely requested a time set) | |
396 | - DST/timezone changes && ntp is active && RTC-in-localtime (never do it without ntp) | |
397 | This takes care of syncing ntpdate updates to the RTC, and DST updates for localtime | |
398 | mode, it will never touch the RTC if the no reliable time source is active or the | |
399 | user did not request anything like it. | |
400 | ||
401 | * libunwind support for coredump pattern hook, and includes this in | |
402 | the message for coredumps. After all, libunwind is now capable to | |
403 | unwind coredumps since a few weeks ago. This probably requires that | |
404 | we have nice support for multi-line messages on display in | |
405 | logs-show.c. Alternatively: use libelfutil, which seems to be the | |
406 | better supported alternative. | |
407 | ||
408 | * add libsystemd-password or so to query passwords during boot using the password agent logic | |
409 | ||
410 | * 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(). | |
411 | ||
412 | * fedup: add --unit to systemctl switch-root somehow | |
413 | * fedup: don't delete initrd on switch-root | |
414 | * fedup: generator | |
415 | ||
416 | * timedated: refuse time changes when NTP is on | |
417 | ||
418 | * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed | |
419 | ||
420 | * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel | |
421 | ||
422 | * add "provisioning" instructions to setup an empty /etc + /var | |
423 | - used to setup a new container from a shared /usr | |
424 | - superset of tmpfiles model | |
425 | - instructions shipped by packages and stored in /usr/lib/ | |
426 | - compose /etc/passwd and /etc/group, copy files | |
427 | - able to create uid + gid used by packages, for file ownership | |
428 | ||
429 | * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without | |
430 | waiting for the timeout | |
431 | ||
432 | * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ... | |
433 | ||
434 | * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't | |
435 | ||
436 | * nspawn: | |
437 | - nspawn: --read-only is not applied recursively to submounts | |
438 | - bind mount read-only the cgroup tree higher than nspawn | |
439 | - nspawn: make it work for dwalsh and shared /usr containers -- tmpfs mounts as command line parameters | |
440 | - refuses to boot containers without /etc/machine-id (OK?), and with empty /etc/machine-id (not OK). | |
441 | - support taking a btrfs snapshot at startup and dropping it afterwards | |
442 | - maybe: hookup nspawn and PrivateNetwork=yes with "ip netns" | |
443 | - generate predictable mac addresses for the netdev we create (host0), maybe based on guest's machine | |
444 | name and machine-id of host (?) | |
445 | ||
446 | * cryptsetup: | |
447 | - cryptsetup-generator: allow specification of passwords in crypttab itself | |
448 | - move cryptsetup key caching into kernel keyctl? | |
449 | https://bugs.freedesktop.org/show_bug.cgi?id=54982 | |
450 | - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator | |
451 | ||
452 | * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it | |
453 | https://bugs.freedesktop.org/show_bug.cgi?id=54712 | |
454 | ||
455 | * after deserializing sockets in socket.c we should reapply sockopts and things | |
456 | ||
457 | * make timer units go away after they elapsed | |
458 | ||
459 | * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html | |
460 | (network and remote-fs on shutdown) | |
461 | ||
462 | * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead | |
463 | ||
464 | * move PID 1 segfaults to /var/lib/systemd/coredump? | |
465 | ||
466 | * create /sbin/init symlinks from the build system | |
467 | ||
468 | * Query Paul Moore about relabelling socket fds while they are open | |
469 | ||
470 | * allow writing multiple conditions in unit files on one line | |
471 | ||
472 | * explore multiple service instances per listening socket idea | |
473 | ||
474 | * MountFlags=shared acts as MountFlags=slave right now. | |
475 | ||
476 | * ReadOnlyDirectories= is not applied recursively to submounts | |
477 | ||
478 | * drop PID 1 reloading, only do reexecing (difficult: Reload() | |
479 | currently is properly synchronous, Reexec() is weird, because we | |
480 | can't delay the response properly until we are back, so instead of | |
481 | being properly synchronous we just keep open the fd and close it | |
482 | when done. That means clients don't get a successful method reply, | |
483 | but much rather a disconnect on success. | |
484 | ||
485 | * properly handle loop back mounts via fstab, especially regards to fsck/passno | |
486 | ||
487 | * allow services with no ExecStart= but with an ExecStop= | |
488 | ||
489 | * dracut-shutdown needs to be ordered before unmounting /boot | |
490 | ||
491 | * initialize the hostname from the fs label of /, if /etc/hostname does not exist? | |
492 | ||
493 | * rename "userspace" to "core-os" | |
494 | ||
495 | * load-fragment: when loading a unit file via a chain of symlinks | |
496 | verify that it isn't masked via any of the names traversed. | |
497 | ||
498 | * introduce Type=pid-file | |
499 | ||
500 | * change Requires=basic.target to RequisiteOverride=basic.target | |
501 | ||
502 | * when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr | |
503 | ||
504 | * automount: implement expire: | |
505 | - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD | |
506 | - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD | |
507 | - every timeout/4 (original autofs logic) | |
508 | - blocking, needs a thread | |
509 | - run until -EAGAIN | |
510 | - receive expire packet on pipe if kernel tells the timeout is over | |
511 | - call umount | |
512 | - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD | |
513 | - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns | |
514 | ||
515 | * ExecOnFailure=/usr/bin/foo | |
516 | ||
517 | * udev: | |
518 | - remove src/udev/udev-builtin-firmware.c (CONFIG_FW_LOADER_USER_HELPER=n) | |
519 | - move to LGPL | |
520 | - kill scsi_id | |
521 | - add trigger --subsystem-match=usb/usb_device device | |
522 | - reimport udev db after MOVE events for devices without dev_t | |
523 | ||
524 | * 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 | |
525 | ||
526 | * introduce mix of BindTo and Requisite | |
527 | ||
528 | * add DeleteSocketsOnStop=yes|no option to socket units | |
529 | ||
530 | * There's currently no way to cancel fsck (used to be possible via C-c or c on the console) | |
531 | ||
532 | * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/ | |
533 | ||
534 | * default unix qlen is too small (10). bump sysctl? add sockopt? | |
535 | ||
536 | * save coredump in Windows/Mozilla minidump format | |
537 | ||
538 | * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting) | |
539 | ||
540 | * default to actual 32bit PIDs, via /proc/sys/kernel/pid_max | |
541 | ||
542 | * be able to specify a forced restart of service A where service B depends on, in case B | |
543 | needs to be auto-respawned? | |
544 | ||
545 | * when a bus name of a service disappears from the bus make sure to queue further activation requests | |
546 | ||
547 | * tmpfiles: | |
548 | - check systemd-tmpfiles for selinux context hookup for mknod(), symlink() and similar | |
549 | - apply "x" on "D" too (see patch from William Douglas) | |
550 | ||
551 | * for services: don't set $HOME in services unless requested | |
552 | ||
553 | * hide PAM/TCPWrap options in fragment parser when compile time disabled | |
554 | ||
555 | * when we automatically restart a service, ensure we restart its rdeps, too. | |
556 | ||
557 | * allow Type=simple with PIDFile= | |
558 | https://bugzilla.redhat.com/show_bug.cgi?id=723942 | |
559 | ||
560 | * move PAM code into its own binary | |
561 | ||
562 | * implement Register= switch in .socket units to enable registration | |
563 | in Avahi, RPC and other socket registration services. | |
564 | ||
565 | * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early | |
566 | ||
567 | * support sd_notify() style notification when reload begins (RELOADING=1), reload is finished (READY=1), and add ReloadSignal= then to use in combination | |
568 | ||
569 | * support sd_notify() style notification when shutting down, to make auto-exit bus services work (STOPPING=1) | |
570 | ||
571 | * verify that the AF_UNIX sockets of a service in the fs still exist | |
572 | when we start a service in order to avoid confusion when a user | |
573 | assumes starting a service is enough to make it accessible | |
574 | ||
575 | * support User= and Group= attributes for AF_UNIX sockets. (difficult, requires NSS from PID 1?) | |
576 | ||
577 | * Make it possible to set the keymap independently from the font on | |
578 | the kernel cmdline. Right now setting one resets also the other. | |
579 | ||
580 | * and a dbus call to generate target from current state | |
581 | ||
582 | * readahead: | |
583 | - drop /.readahead on bigger upgrades with yum | |
584 | - move readahead files into /var (look for them with .path units?) | |
585 | - readahead: use BTRFS_IOC_DEFRAG_RANGE instead of BTRFS_IOC_DEFRAG ioctl, with START_IO | |
586 | - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes | |
587 | - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/ | |
588 | ||
589 | * add support for /bin/mount -s | |
590 | ||
591 | * GC unreferenced jobs (such as .device jobs) | |
592 | ||
593 | * write blog stories about: | |
594 | - hwdb: what belongs into it, lsusb | |
595 | - enabling dbus services | |
596 | - status update | |
597 | - how to make changes to sysctl and sysfs attributes | |
598 | - remote access | |
599 | - how to pass throw-away units to systemd, or dynamically change properties of existing units | |
600 | - how to integrate cgconfig and suchlike with systemd | |
601 | - testing with Harald's awesome test kit | |
602 | - auto-restart | |
603 | - how to develop against journal browsing APIs | |
604 | - the journal HTTP iface | |
605 | - non-cgroup resource management | |
606 | - dynamic resource management with cgroups | |
607 | - refreshed, longer missions statement | |
608 | - calendar time events | |
609 | - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell | |
610 | - how to create your own target | |
611 | - instantiated apache, dovecot and so on | |
612 | - hooking a script into various stages of shutdown/rearly booot | |
613 | ||
614 | * allow port=0 in .socket units | |
615 | ||
616 | * support systemd.mask= on the kernel command line. | |
617 | ||
618 | * recreate systemd's D-Bus private socket file on SIGUSR2 | |
619 | ||
620 | * Support --test based on current system state | |
621 | ||
622 | * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support. | |
623 | ||
624 | * maybe introduce ExecRestartPre= | |
625 | ||
626 | * dot output for --test showing the 'initial transaction' | |
627 | ||
628 | * fingerprint.target, wireless.target, gps.target, netdevice.target | |
629 | ||
630 | * io priority during initialization | |
631 | ||
632 | * drop cap bounding set in readahead and other services | |
633 | ||
634 | * systemd-python: | |
635 | - figure out a simple way to wait for journal events in a way that | |
636 | works with ^C | |
637 | - add documentation to systemd.daemon | |
638 | ||
639 | * bootchart: | |
640 | - plot per-process IO utilization | |
641 | - group processes based on service association (cgroups) | |
642 | - document initcall_debug | |
643 | - kernel cmdline "bootchart" option for simplicity? | |
644 | ||
645 | * udev-link-config: | |
646 | - Make sure ID_PATH is always exported and complete for | |
647 | network devices where possible, so we can safely rely | |
648 | on Path= matching | |
649 | ||
650 | * sd-rtnl: | |
651 | - add support for exiting containers without reading them fully first | |
652 | - add support for more attribute types | |
653 | ||
654 | * networkd: | |
655 | - make sure RTM_NEWLINK messages match both the ifname and kind when setting the ifindex of a netdev | |
656 | - add more keys to [Route] and [Address] sections | |
657 | - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config) | |
658 | - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline) | |
659 | - add reduced [Link] support to .network files | |
660 | - add IPv4LL tests (inspire by DHCP) | |
661 | - add Scope= parsing option for [Network] | |
662 | - have smooth transition from LL to routable address, without disconnecting clients. | |
663 | ||
664 | * sd-network: | |
665 | - make sure ipv4ll and dhcp clients can handle changing mac addresses while running | |
666 | ||
667 | External: | |
668 | ||
669 | * dbus: | |
670 | - natively watch for dbus-*.service symlinks (PENDING) | |
671 | - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service | |
672 | ||
673 | * fix alsa mixer restore to not print error when no config is stored | |
674 | ||
675 | * make cryptsetup lower --iter-time | |
676 | ||
677 | * patch kernel for xattr support in /dev, /proc/, /sys? | |
678 | ||
679 | * NTP: the kernel's 11-minutes-mode syncs the system time to the RTC, but only | |
680 | in an ~30 minutes window. It does not adjust larger differences. Find a way | |
681 | to tell the kernel, to always do a full time sync when the RTC is in UTC and | |
682 | we are in 11-minutes-mode. When we trust the system time to NTP we also want | |
683 | the RTC to sync up. | |
684 | ||
685 | * kernel: add device_type = "fb", "fbcon" to class "graphics" | |
686 | ||
687 | * drop accountsservice's StandardOutput=syslog and Type=dbus fields | |
688 | ||
689 | * dbus upstream still refers to dbus.target and shouldn't | |
690 | ||
691 | * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id | |
692 | ||
693 | * add "# export SYSTEMD_PAGER=" to bash login | |
694 | ||
695 | * /usr/bin/service should actually show the new command line | |
696 | ||
697 | * 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= | |
698 | ||
699 | * fedora: F20: go timer units all the way, leave cron.daily for cron | |
700 | ||
701 | * neither pkexec nor sudo initialize environ[] from the PAM environment? | |
702 | ||
703 | * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it | |
704 | ||
705 | Regularly: | |
706 | ||
707 | * look for close() vs. close_nointr() vs. close_nointr_nofail() | |
708 | ||
709 | * check for strerror(r) instead of strerror(-r) | |
710 | ||
711 | * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel | |
712 | ||
713 | * pahole | |
714 | ||
715 | * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()! | |
716 | ||
717 | * use secure_getenv() instead of getenv() where appropriate | |
718 | ||
719 | * link up selected blog stories from man pages and unit files Documentation= fields | |
720 | ||
721 | Scheduled for removal or fixing: | |
722 | ||
723 | * xxxOverridable dependencies (probably: fix) |