]>
Commit | Line | Data |
---|---|---|
1 | Bugfixes: | |
2 | ||
3 | * Many manager configuration settings that are only applicable to user | |
4 | manager or system manager can be always set. It would be better to reject | |
5 | them when parsing config. | |
6 | ||
7 | * busctl --user call org.freedesktop.systemd1 /org/freedesktop/systemd1 org.freedesktop.systemd1.Manager GetUnitProcesses "s" run-rbff1b85427b34ba3adf864281aeda8e7.service | |
8 | Failed to set address: No such file or directory | |
9 | ||
10 | → improve error message | |
11 | ||
12 | External: | |
13 | ||
14 | * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros. | |
15 | ||
16 | * wiki: update journal format documentation for lz4 additions | |
17 | ||
18 | Janitorial Clean-ups: | |
19 | ||
20 | * Rearrange tests so that the various test-xyz.c match a specific src/basic/xyz.c again | |
21 | ||
22 | * rework mount.c and swap.c to follow proper state enumeration/deserialization | |
23 | semantics, like we do for device.c now | |
24 | ||
25 | Features: | |
26 | ||
27 | * seed: check if first-boot and then don't do anything | |
28 | ||
29 | * logind: rework pam_logind to also do a bus call in case of invocation from | |
30 | user@.service, which returns the XDG_RUNTIME_DIR value, and make this | |
31 | behaviour selectable via pam module option. | |
32 | ||
33 | * introduce a new per-process uuid, similar to the boot id, the machine id, the | |
34 | invocation id, that is derived from process creds, specifically a hashed | |
35 | combination of AT_RANDOM + getpid() + the starttime from | |
36 | /proc/self/status. Then add these ids implicitly when logging. Deriving this | |
37 | uuid from these three things has the benefit that it can be derived easily | |
38 | from /proc/$PID/ in a stable, and unique way that changes on both fork() and | |
39 | exec(). | |
40 | ||
41 | * let's not GC a unit while its ratelimits are still pending | |
42 | ||
43 | * when killing due to service watchdog timeout maybe detect whether target | |
44 | process is under ptracing and then log loudly and continue instead. | |
45 | ||
46 | * introduce a new group to own TPM devices | |
47 | ||
48 | * make rfkill uaccess controllable by default, i.e. steal rule from | |
49 | gnome-bluetooth and friends | |
50 | ||
51 | * warn if udev rules files are marked executable (docker?) | |
52 | ||
53 | * tweak journald context caching. In addition to caching per-process attributes | |
54 | keyed by PID, cache per-cgroup attributes (i.e. the various xattrs we read) | |
55 | keyed by cgroup path, and guarded by ctime changes. This should provide us | |
56 | with a nice speed-up on services that have many processes running in the same | |
57 | cgroup. | |
58 | ||
59 | * make MAINPID= message reception checks even stricter: if service uses User=, | |
60 | then check sending UID and ignore message if it doesn't match the user or | |
61 | root. | |
62 | ||
63 | * maybe trigger a uevent "change" on a device if "systemctl reload xyz.device" | |
64 | is issued. | |
65 | ||
66 | * when importing an fs tree with machined, optionally apply userns-rec-chown | |
67 | ||
68 | * when importing an fs tree with machined, complain if image is not an OS | |
69 | ||
70 | * Maybe introduce a helper safe_exec() or so, which is to execve() which | |
71 | safe_fork() is to fork(). And then make revert the RLIMIT_NOFILE soft limit | |
72 | to 1K implicitly, unless explicitly opted-out. | |
73 | ||
74 | * rework seccomp/nnp logic that even if User= is used in combination with | |
75 | a seccomp option we don't have to set NNP. For that, change uid first whil | |
76 | keeping CAP_SYS_ADMIN, then apply seccomp, the drop cap. | |
77 | ||
78 | * add a concept for automatically loading per-unit secrets off disk and | |
79 | inserting them into the kernel keyring. Maybe SecretsDirectory= similar to | |
80 | ConfigurationDirectory=. | |
81 | ||
82 | * when no locale is configured, default to UEFI's PlatformLang variable | |
83 | ||
84 | * When logind.conf contains HandleLidSwitch=suspend-then-hibernate and we can't | |
85 | hibernate because the swap partition isn't large enough, still suspend | |
86 | ||
87 | * bootctl,sd-boot: actually honour the "architecture" key | |
88 | ||
89 | * add a new syscall group "@esoteric" for more esoteric stuff such as bpf() and | |
90 | usefaultd() and make systemd-analyze check for it. | |
91 | ||
92 | * paranoia: whenever we process passwords, call mlock() on the memory | |
93 | first. i.e. look for all places we use string_erase()/free_and_erasep() and | |
94 | augment them with mlock(). Also use MADV_DONTDUMP. | |
95 | ||
96 | * Move RestrictAddressFamily= to the new cgroup create socket | |
97 | ||
98 | * support the bind/connect/sendmsg cgroup stuff for sandboxing, and possibly | |
99 | patching around | |
100 | ||
101 | * maybe implicitly attach monotonic+realtime timestamps to outgoing messages in | |
102 | log.c and sd-journal-send | |
103 | ||
104 | * optionally: turn on cgroup delegation for per-session scope units | |
105 | ||
106 | * introduce per-unit (i.e. per-slice, per-service) journal log size limits. | |
107 | ||
108 | * optionally, if a per-partition GPT flag is set for the root/home/… partitions | |
109 | format the partition on next boot and unset the flag, in order to implement | |
110 | factory reset. also, add a second flag that simply indicates whether such a | |
111 | scheme is supported. then, add a tool (or maybe beef up systemd-dissect) to | |
112 | show state of these flags, and optionally trigger such a factory reset on | |
113 | next boot by setting the flag. | |
114 | ||
115 | * sd-boot: automatically load EFI modules from some drop-in dir, so that people | |
116 | can add in file system drivers and such | |
117 | ||
118 | * sd-boot: optionally, show boot menu when previous default boot item has | |
119 | non-zero "tries done" count | |
120 | ||
121 | * maybe set a special xattr on cgroups that have delegate=yes set, to make it | |
122 | easy to mark cut points | |
123 | ||
124 | * introduce an option (or replacement) for "systemctl show" that outputs all | |
125 | properties as JSON, similar to busctl's new JSON output. In contrast to that | |
126 | it should skip the variant type string though. | |
127 | ||
128 | * augment CODE_FILE=, CODE_LINE= with something like CODE_BASE= or so which | |
129 | contains some identifier for the project, which allows us to include | |
130 | clickable links to source files generating these log messages. The identifier | |
131 | could be some abberviated URL prefix or so (taking inspiration from Go | |
132 | imports). For example, for systemd we could use | |
133 | CODE_BASE=github.com/systemd/systemd/blob/98b0b1123cc or so which is | |
134 | sufficient to build a link by prefixing "http://" and suffixing the | |
135 | CODE_FILE. | |
136 | ||
137 | * when outputting log data with journalctl and the log data includes references | |
138 | to configuration files (CONFIG_FILE=), create a clickable link for it. | |
139 | ||
140 | * Augment MESSAGE_ID with MESSAGE_BASE, in a similar fashion so that we can | |
141 | make clickable links from log messages carrying a MESSAGE_ID, that lead to | |
142 | some explanatory text online. | |
143 | ||
144 | * maybe extend .path units to expose fanotify() per-mount change events | |
145 | ||
146 | * Add a "systemctl list-units --by-slice" mode or so, which rearranges the | |
147 | output of "systemctl list-units" slightly by showing the tree structure of | |
148 | the slices, and the units attached to them. | |
149 | ||
150 | * the a-posteriori stopping of units bound to units that disappeared logic | |
151 | should be reworked: there should be a queue of units, and we should only | |
152 | enqeue stop jobs from a defer event that processes queue instead of | |
153 | right-away when we find a unit that is bound to one that doesn't exist | |
154 | anymore. (similar to how the stop-unneeded queue has been reworked the same | |
155 | way) | |
156 | ||
157 | * nspawn: make nspawn suitable for shell pipelines: instead of triggering a | |
158 | hangup when input is finished, send ^D, which synthesizes an EOF. Then wait | |
159 | for hangup or ^D before passing on the EOF. | |
160 | ||
161 | * When reloading configuration PID 1 should reset all its properties to the | |
162 | original defaults before calling parse_config() | |
163 | ||
164 | * nspawn: greater control over selinux label? | |
165 | ||
166 | * hibernate/s2h: make this robust and safe to enable in Fedora by default. | |
167 | Specifically: | |
168 | ||
169 | 1. add resume_offset support to the resume code (i.e. support swap files | |
170 | properly) | |
171 | 2. check if swap is on weird storage and refuse if so | |
172 | 3. add autodetection of hibernation images | |
173 | ||
174 | * cgroups: use inotify to get notified when somebody else modifies cgroups | |
175 | owned by us, then log a friendly warning. | |
176 | ||
177 | * beef up log.c with support for stripping ANSI sequences from strings, so that | |
178 | it is OK to include them in log strings. This would be particularly useful so | |
179 | that our log messages could contain clickable links for example for unit | |
180 | files and suchlike we operate on. | |
181 | ||
182 | * add support for "portablectl attach http://foobar.com/waaa.raw (i.e. importd integration) | |
183 | ||
184 | * add attach --enable and attach --now (for attach+enable+start) | |
185 | ||
186 | * sync dynamic uids/gids between host+portable srvice (i.e. if DynamicUser=1 is set for a service, make sure that the | |
187 | selected user is resolvable in the service even if it ships its own /etc/passwd) | |
188 | ||
189 | * Fix DECIMAL_STR_MAX or DECIMAL_STR_WIDTH. One includes a trailing NUL, the | |
190 | other doesn't. What a disaster. Probably to exclude it. Also | |
191 | DECIMAL_STR_WIDTH should probably add an extra "-" into account for negative | |
192 | numbers. | |
193 | ||
194 | * Check that users of inotify's IN_DELETE_SELF flag are using it properly, as | |
195 | usually IN_ATTRIB is the right way to watch deleted files, as the former only | |
196 | fires when a file is actually removed from disk, i.e. the link count drops to | |
197 | zero and is not open anymore, while the latter happens when a file is | |
198 | unlinked from any dir. | |
199 | ||
200 | * port systemctl, busctl, … over to format-table.[ch]'s table formatters | |
201 | ||
202 | * pid1: lock image configured with RootDirectory=/RootImage= using the usual nspawn semantics while the unit is up | |
203 | ||
204 | * add --vacuum-xyz options to coredumpctl, matching those journalctl already has. | |
205 | ||
206 | * SuccessExitStatus= and friends should probably also accept symbolic exit | |
207 | codes names, i.e. error codes from the list maintained in exit-codes.[ch] | |
208 | ||
209 | * introduce Ephemeral= unit file switch, that creates an ephemeral copy of all | |
210 | files and directories that are left writable for a unit, and which are | |
211 | removed after the unit goes down again. A bit like --ephemeral for | |
212 | systemd-nspawn but for system services. If used together with RootImage= this | |
213 | should reflink the image file itself. | |
214 | ||
215 | Related: add Ephemeral=<path1> <path2> … which would allow marking | |
216 | specific paths only like this. | |
217 | ||
218 | * add CopyFile= or so as unit file setting that may be used to copy files or | |
219 | directory trees from the host to the services RootImage= and RootDirectory= | |
220 | environment. Which we can use for /etc/machine-id and in particular | |
221 | /etc/resolv.conf. Should be smart and do something useful on read-only | |
222 | images, for example fallback to read-only bind mounting the file instead. | |
223 | ||
224 | * show invocation ID in systemd-run output | |
225 | ||
226 | * bypass SIGTERM state in unit files if KillSignal is SIGKILL | |
227 | ||
228 | * tree-wide: ensure we always block the signals we hook into with | |
229 | sd_event_add_signal() first | |
230 | ||
231 | * add proper dbus APIs for the various sd_notify() commands, such as MAINPID=1 | |
232 | and so on, which would mean we could report errors and such. | |
233 | ||
234 | * teach tmpfiles.d q/Q logic something sensible in the context of XFS/ext4 | |
235 | project quota | |
236 | ||
237 | * introduce DefaultSlice= or so in system.conf that allows changing where we | |
238 | place our units by default, i.e. change system.slice to something | |
239 | else. Similar, ManagerSlice= should exist so that PID1's own scope unit could | |
240 | be moved somewhere else too. Finally machined and logind should get similar | |
241 | options so that it is possible to move user session scopes and machines to a | |
242 | different slice too by default. Usecase: people who want to put resources on | |
243 | the entire system, with the exception of one specific service. See: | |
244 | https://lists.freedesktop.org/archives/systemd-devel/2018-February/040369.html | |
245 | ||
246 | * maybe rework get_user_creds() to query the user database if $SHELL is used | |
247 | for root, but only then. | |
248 | ||
249 | * be stricter with fds we receive for the fdstore: close them asynchronously | |
250 | ||
251 | * calenderspec: add support for week numbers and day numbers within a | |
252 | year. This would allow us to define "bi-weekly" triggers safely. | |
253 | ||
254 | * sd-bus: add vtable flag, that may be used to request client creds implicitly | |
255 | and asynchronously before dispatching the operation | |
256 | ||
257 | * make use of ethtool veth peer info in machined, for automatically finding out | |
258 | host-side interface pointing to the container. | |
259 | ||
260 | * add some special mode to LogsDirectory=/StateDirectory=… that allows | |
261 | declaring these directories without necessarily pulling in deps for them, or | |
262 | creating them when starting up. That way, we could declare that | |
263 | systemd-journald writes to /var/log/journal, which could be useful when we | |
264 | doing disk usage calculations and so on. | |
265 | ||
266 | * taint systemd if there are fewer than 65536 users assigned (userns) to the system. | |
267 | ||
268 | * deprecate RootDirectoryStartOnly= in favour of a new ExecStart= prefix char | |
269 | ||
270 | * add a new RuntimeDirectoryPreserve= mode that defines a similar lifecycle for | |
271 | the runtime dir as we maintain for the fdstore: i.e. keep it around as long | |
272 | as the unit is running or has a job queued. | |
273 | ||
274 | * support projid-based quota in machinectl for containers | |
275 | ||
276 | * maybe use SOURCE_DATE_EPOCH (i.e. the env var the reproducible builds folks | |
277 | introduced) as the RTC epoch, instead of the mtime of NEWS. | |
278 | ||
279 | * add a way to lock down cgroup migration: a boolean, which when set for a unit | |
280 | makes sure the processes in it can never migrate out of it | |
281 | ||
282 | * blog about fd store and restartable services | |
283 | ||
284 | * document Environment=SYSTEMD_LOG_LEVEL=debug drop-in in debugging document | |
285 | ||
286 | * rework ExecOutput and ExecInput enums so that EXEC_OUTPUT_NULL loses its | |
287 | magic meaning and is no longer upgraded to something else if set explicitly. | |
288 | ||
289 | * in the long run: permit a system with /etc/machine-id linked to /dev/null, to | |
290 | make it lose its identity, i.e. be anonymous. For this we'd have to patch | |
291 | through the whole tree to make all code deal with the case where no machine | |
292 | ID is available. | |
293 | ||
294 | * optionally, collect cgroup resource data, and store it in per-unit RRD files, | |
295 | suitable for processing with rrdtool. Add bus API to access this data, and | |
296 | possibly implement a CPULoad property based on it. | |
297 | ||
298 | * beef up pam_systemd to take unit file settings such as cgroups properties as | |
299 | parameters | |
300 | ||
301 | * maybe hook of xfs/ext4 quotactl() with services? i.e. automatically manage | |
302 | the quota of a the user indicated in User= via unit file settings, like the | |
303 | other resource management concepts. Would mix nicely with DynamicUser=1. Or | |
304 | alternatively, do this with projids, so that we can also cover services | |
305 | running as root. Quota should probably cover all the special dirs such as | |
306 | StateDirectory=, LogsDirectory=, CacheDirectory=, as well as RootDirectory= if it | |
307 | is set, plus the whole disk space any image configured with RootImage=. | |
308 | ||
309 | * In DynamicUser= mode: before selecting a UID, use disk quota APIs on relevant | |
310 | disks to see if the UID is already in use. | |
311 | ||
312 | * add "systemctl wait" or so, which does what "systemd-run --wait" does, but | |
313 | for all units. It should be both a way to pin units into memory as well as a | |
314 | wait to retrieve their exit data. | |
315 | ||
316 | * expose IO accounting data on the bus, show it in systemd-run --wait and log | |
317 | about it in the resource log message | |
318 | ||
319 | * add "systemctl purge" for flushing out configuration, state, logs, ... of a | |
320 | unit when it is stopped | |
321 | ||
322 | * show whether a service has out-of-date configuration in "systemctl status" by | |
323 | using mtime data of ConfigurationDirectory=. | |
324 | ||
325 | * replace all remaining uses of fgets() + LINE_MAX by read_line() | |
326 | ||
327 | * Add AddUser= setting to unit files, similar to DynamicUser=1 which however | |
328 | creates a static, persistent user rather than a dynamic, transient user. We | |
329 | can leverage code from sysusers.d for this. | |
330 | ||
331 | * add some optional flag to ReadWritePaths= and friends, that has the effect | |
332 | that we create the dir in question when the service is started. Example: | |
333 | ||
334 | ReadWritePaths=:/var/lib/foobar | |
335 | ||
336 | * maybe add call sd_journal_set_block_timeout() or so to set SO_SNDTIMEO for | |
337 | the sd-journal logging socket, and, if the timeout is set to 0, sets | |
338 | O_NONBLOCK on it. That way people can control if and when to block for | |
339 | logging. | |
340 | ||
341 | * hostnamed: populate form factor data from a new hwdb database, so that old | |
342 | yogas can be recognized as "convertible" too, even if they predate the DMI | |
343 | "convertible" form factor | |
344 | ||
345 | * Maybe add a small tool invoked early at boot, that adds in or resizes | |
346 | partitions automatically, to be used when the media used is actually larger | |
347 | than the image written onto it is. | |
348 | ||
349 | * Maybe add PrivatePIDs= as new unit setting, and do minimal PID namespacing | |
350 | after all. Be strict however, only support the equivalent of nspawn's | |
351 | --as-pid2 switch, and sanely proxy sd_notify() messages dropping stuff such | |
352 | as MAINPID. | |
353 | ||
354 | * Add ExecMonitor= setting. May be used multiple times. Forks off a process in | |
355 | the service cgroup, which is supposed to monitor the service, and when it | |
356 | exits the service is considered failed by its monitor. | |
357 | ||
358 | * track the per-service PAM process properly (i.e. as an additional control | |
359 | process), so that it may be queried on the bus and everything. | |
360 | ||
361 | * add a new "debug" job mode, that is propagated to unit_start() and for | |
362 | services results in two things: we raise SIGSTOP right before invoking | |
363 | execve() and turn off watchdog support. Then, use that to implement | |
364 | "systemd-gdb" for attaching to the start-up of any system service in its | |
365 | natural habitat. | |
366 | ||
367 | * maybe introduce gpt auto discovery for /var/tmp? | |
368 | ||
369 | * maybe add gpt-partition-based user management: each user gets his own | |
370 | LUKS-encrypted GPT partition with a new GPT type. A small nss module | |
371 | enumerates users via udev partition enumeration. UIDs are assigned in a fixed | |
372 | way: the partition index is added as offset to some fixed base uid. User name | |
373 | is stored in GPT partition name. A PAM module authenticates the user via the | |
374 | LUKS partition password. Benefits: strong per-user security, compatibility | |
375 | with stateless/read-only/verity-enabled root. (other idea: do this based on | |
376 | loopback files in /home, without GPT involvement) | |
377 | ||
378 | * gpt-auto logic: introduce support for discovering /var matching an image. For | |
379 | that, use a partition type UUID that is hashed from the OS name (as encoded | |
380 | in /etc/os-release), the architecture, and 4 new bits from the gpt flags | |
381 | field of the root partition. This way can easily support multiple OS | |
382 | installations on the same GPT partition table, without problems with | |
383 | unmatched /var partitions. | |
384 | ||
385 | * gpt-auto logic: related to the above, maybe support a "secondary" root | |
386 | partition, that is mounted to / and is writable, and where the actual root's | |
387 | /usr is mounted into. | |
388 | ||
389 | * gpt-auto logic: support encrypted swap, add kernel cmdline option to force it, and honour a gpt bit about it, plus maybe a configuration file | |
390 | ||
391 | * drop nss-myhostname in favour of nss-resolve? | |
392 | ||
393 | * add a percentage syntax for TimeoutStopSec=, e.g. TimeoutStopSec=150%, and | |
394 | then use that for the setting used in user@.service. It should be understood | |
395 | relative to the configured default value. | |
396 | ||
397 | * in networkd, when matching device types, fix up DEVTYPE rubbish the kernel passes to us | |
398 | ||
399 | * enable LockMLOCK to take a percentage value relative to physical memory | |
400 | ||
401 | * Permit masking specific netlink APIs with RestrictAddressFamily= | |
402 | ||
403 | * nspawn: support that /proc, /sys/, /dev are pre-mounted | |
404 | ||
405 | * define gpt header bits to select volatility mode | |
406 | ||
407 | * ProtectKernelLogs= (drops CAP_SYSLOG, add seccomp for syslog() syscall, and DeviceAllow to /dev/kmsg) in service files | |
408 | ||
409 | * ProtectClock= (drops CAP_SYS_TIMES, adds seecomp filters for settimeofday, adjtimex), sets DeviceAllow o /dev/rtc | |
410 | ||
411 | * ProtectTracing= (drops CAP_SYS_PTRACE, blocks ptrace syscall, makes /sys/kernel/tracing go away) | |
412 | ||
413 | * ProtectMount= (drop mount/umount/pivot_root from seccomp, disallow fuse via DeviceAllow, imply Mountflags=slave) | |
414 | ||
415 | * ProtectKeyRing= to take keyring calls away | |
416 | ||
417 | * RemoveKeyRing= to remove all keyring entries of the specified user | |
418 | ||
419 | * ProtectReboot= that masks reboot() and kexec_load() syscalls, prohibits kill | |
420 | on PID 1 with the relevant signals, and makes relevant files in /sys and | |
421 | /proc (such as the sysrq stuff) unavailable | |
422 | ||
423 | * make sure the ratelimit object can deal with USEC_INFINITY as way to turn off things | |
424 | ||
425 | * journalctl: make sure -f ends when the container indicated by -M terminates | |
426 | ||
427 | * mount: automatically search for "main" partition of an image has multiple | |
428 | partitions | |
429 | ||
430 | * expose the "privileged" flag of ExecCommand on the bus, and open it up to | |
431 | transient units | |
432 | ||
433 | * in nss-systemd, if we run inside of RootDirectory= with PrivateUsers= set, | |
434 | find a way to map the User=/Group= of the service to the right name. This way | |
435 | a user/group for a service only has to exist on the host for the right | |
436 | mapping to work. | |
437 | ||
438 | * add bus API for creating unit files in /etc, reusing the code for transient units | |
439 | ||
440 | * add bus API to remove unit files from /etc | |
441 | ||
442 | * add bus API to retrieve current unit file contents (i.e. implement "systemctl cat" on the bus only) | |
443 | ||
444 | * rework fopen_temporary() to make use of open_tmpfile_linkable() (problem: the | |
445 | kernel doesn't support linkat() that replaces existing files, currently) | |
446 | ||
447 | * transient units: don't bother with actually setting unit properties, we | |
448 | reload the unit file anyway | |
449 | ||
450 | * journald: sigbus API via a signal-handler safe function that people may call | |
451 | from the SIGBUS handler | |
452 | ||
453 | * optionally, also require WATCHDOG=1 notifications during service start-up and shutdown | |
454 | ||
455 | * delay activation of logind until somebody logs in, or when /dev/tty0 pulls it | |
456 | in or lingering is on (so that containers don't bother with it until PAM is used). also exit-on-idle | |
457 | ||
458 | * cache sd_event_now() result from before the first iteration... | |
459 | ||
460 | * add systemctl stop --job-mode=triggering that follows TRIGGERED_BY deps and adds them to the same transaction | |
461 | ||
462 | * PID1: find a way how we can reload unit file configuration for | |
463 | specific units only, without reloading the whole of systemd | |
464 | ||
465 | * add an explicit parser for LimitRTPRIO= that verifies | |
466 | the specified range and generates sane error messages for incorrect | |
467 | specifications. | |
468 | ||
469 | * when we detect that there are waiting jobs but no running jobs, do something | |
470 | ||
471 | * push CPUAffinity= also into the "cpuset" cgroup controller (only after the cpuset controller got ported to the unified hierarchy) | |
472 | ||
473 | * PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn) | |
474 | ||
475 | * there's probably something wrong with having user mounts below /sys, | |
476 | as we have for debugfs. for example, src/core/mount.c handles mounts | |
477 | prefixed with /sys generally special. | |
478 | http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html | |
479 | ||
480 | * fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline | |
481 | ||
482 | * docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date | |
483 | ||
484 | * add a job mode that will fail if a transaction would mean stopping | |
485 | running units. Use this in timedated to manage the NTP service | |
486 | state. | |
487 | http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html | |
488 | ||
489 | * The udev blkid built-in should expose a property that reflects | |
490 | whether media was sensed in USB CF/SD card readers. This should then | |
491 | be used to control SYSTEMD_READY=1/0 so that USB card readers aren't | |
492 | picked up by systemd unless they contain a medium. This would mirror | |
493 | the behaviour we already have for CD drives. | |
494 | ||
495 | * networkd/udev: implement SR_IOV configuration in .link files: | |
496 | http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html | |
497 | ||
498 | * hostnamectl: show root image uuid | |
499 | ||
500 | * Find a solution for SMACK capabilities stuff: | |
501 | http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html | |
502 | ||
503 | * "systemctl preset-all" should probably order the unit files it | |
504 | operates on lexicographically before starting to work, in order to | |
505 | ensure deterministic behaviour if two unit files conflict (like DMs | |
506 | do, for example) | |
507 | ||
508 | * synchronize console access with BSD locks: | |
509 | http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html | |
510 | ||
511 | * as soon as we have sender timestamps, revisit coalescing multiple parallel daemon reloads: | |
512 | http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html | |
513 | ||
514 | * in systemctl list-unit-files: show the install value the presets would suggest for a service in a third column | |
515 | ||
516 | * figure out when we can use the coarse timers | |
517 | ||
518 | * add "systemctl start -v foobar.service" that shows logs of a service | |
519 | while the start command runs. This is non-trivial to do without | |
520 | races though, since we should flush out all journal messages before | |
521 | returning from the "systemctl stop". | |
522 | ||
523 | * firstboot: make it useful to be run immediately after yum --installroot to set up a machine. (most specifically, make --copy-root-password work even if /etc/passwd already exists | |
524 | ||
525 | * maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment= | |
526 | ||
527 | * consider showing the unit names during boot up in the status output, not just the unit descriptions | |
528 | ||
529 | * maybe allow timer units with an empty Units= setting, so that they | |
530 | can be used for resuming the system but nothing else. | |
531 | ||
532 | * what to do about udev db binary stability for apps? (raw access is not an option) | |
533 | ||
534 | * man: maybe use the word "inspect" rather than "introspect"? | |
535 | ||
536 | * systemctl: if some operation fails, show log output? | |
537 | ||
538 | * systemctl edit: use equivalent of cat() to insert existing config as a comment, prepended with #. | |
539 | Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc. | |
540 | ||
541 | * exponential backoff in timesyncd when we cannot reach a server | |
542 | ||
543 | * timesyncd: add ugly bus calls to set NTP servers per-interface, for usage by NM | |
544 | ||
545 | * merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share.... | |
546 | ||
547 | * systemd.show_status= should probably have a mode where only failed | |
548 | units are shown. | |
549 | ||
550 | * add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL | |
551 | (throughout the codebase, not only PID1) | |
552 | ||
553 | * resolved: | |
554 | - mDNS/DNS-SD | |
555 | - service registration | |
556 | - service/domain/types browsing | |
557 | - avahi compat | |
558 | - DNS-SD service registration from socket units | |
559 | - resolved should optionally register additional per-interface LLMNR | |
560 | names, so that for the container case we can establish the same name | |
561 | (maybe "host") for referencing the server, everywhere. | |
562 | - allow clients to request DNSSEC for a single lookup even if DNSSEC is off (?) | |
563 | - hook up resolved with machined-based address resolution | |
564 | ||
565 | * refcounting in sd-resolve is borked | |
566 | ||
567 | * Add a new verb "systemctl top" | |
568 | ||
569 | * add new gpt type for btrfs volumes | |
570 | ||
571 | * support empty /etc boots nicely: | |
572 | - nspawn/gpt-generator: introduce new gpt partition type for /usr | |
573 | ||
574 | * generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them. | |
575 | ||
576 | * a way for container managers to turn off getty starting via $container_headless= or so... | |
577 | ||
578 | * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit | |
579 | ||
580 | * For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services | |
581 | they run added to the initial transaction and thus confuse Type=idle. | |
582 | ||
583 | * add bus api to query unit file's X fields. | |
584 | ||
585 | * gpt-auto-generator: | |
586 | - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap? | |
587 | - Make /home automount rather than mount? | |
588 | ||
589 | * add generator that pulls in systemd-network from containers when | |
590 | CAP_NET_ADMIN is set, more than the loopback device is defined, even | |
591 | when it is otherwise off | |
592 | ||
593 | * MessageQueueMessageSize= (and suchlike) should use parse_iec_size(). | |
594 | ||
595 | * implement Distribute= in socket units to allow running multiple | |
596 | service instances processing the listening socket, and open this up | |
597 | for ReusePort= | |
598 | ||
599 | * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze | |
600 | $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls | |
601 | should SIGSTOP all unit processes in a loop until all processes of | |
602 | it are fully stopped. This can later be used for app management by | |
603 | desktop UIs such as gnome-shell to freeze apps that are not visible | |
604 | on screen, not unlike how job control works on the shell | |
605 | ||
606 | * cgroups: | |
607 | - implement per-slice CPUFairScheduling=1 switch | |
608 | - introduce high-level settings for RT budget, swappiness | |
609 | - how to reset dynamically changed unit cgroup attributes sanely? | |
610 | - when reloading configuration, apply new cgroup configuration | |
611 | - when recursively showing the cgroup hierarchy, optionally also show | |
612 | the hierarchies of child processes | |
613 | ||
614 | * transient units: | |
615 | - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt | |
616 | ||
617 | * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk? | |
618 | ||
619 | * when we detect low battery and no AC on boot, show pretty splash and refuse boot | |
620 | ||
621 | * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops | |
622 | ||
623 | * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1 | |
624 | ||
625 | * 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 | |
626 | ||
627 | * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs | |
628 | ||
629 | * If we try to find a unit via a dangling symlink, generate a clean | |
630 | error. Currently, we just ignore it and read the unit from the search | |
631 | path anyway. | |
632 | ||
633 | * refuse boot if /usr/lib/os-release is missing or /etc/machine-id cannot be set up | |
634 | ||
635 | * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted. | |
636 | ||
637 | * load .d/*.conf dropins for device units | |
638 | ||
639 | * sd-bus: | |
640 | - EBADSLT handling | |
641 | - GetAllProperties() on a non-existing object does not result in a failure currently | |
642 | - port to sd-resolve for connecting to TCP dbus servers | |
643 | - see if we can introduce a new sd_bus_get_owner_machine_id() call to retrieve the machine ID of the machine of the bus itself | |
644 | - see if we can drop more message validation on the sending side | |
645 | - add API to clone sd_bus_message objects | |
646 | - longer term: priority inheritance | |
647 | - dbus spec updates: | |
648 | - NameLost/NameAcquired obsolete | |
649 | - GVariant | |
650 | - path escaping | |
651 | - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now | |
652 | ||
653 | * sd-event | |
654 | - allow multiple signal handlers per signal? | |
655 | - document chaining of signal handler for SIGCHLD and child handlers | |
656 | - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ... | |
657 | - generate a failure of a default event loop is executed out-of-thread | |
658 | ||
659 | * investigate endianness issues of UUID vs. GUID | |
660 | ||
661 | * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we | |
662 | should be able to safely try another attempt when the bus call LoadUnit() is invoked. | |
663 | ||
664 | * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login. | |
665 | ||
666 | * add a pam module that on password changes updates any LUKS slot where the password matches | |
667 | ||
668 | * test/: | |
669 | - add unit tests for config_parse_device_allow() | |
670 | ||
671 | * seems that when we follow symlinks to units we prefer the symlink | |
672 | destination path over /etc and /usr. We should not do that. Instead | |
673 | /etc should always override /run+/usr and also any symlink | |
674 | destination. | |
675 | ||
676 | * when isolating, try to figure out a way how we implicitly can order | |
677 | all units we stop before the isolating unit... | |
678 | ||
679 | * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off}) | |
680 | ||
681 | * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify | |
682 | that the boot loader should be installed to the ESP. Define a way | |
683 | how an installer can figure out whether a BLS compliant boot loader | |
684 | is installed. | |
685 | ||
686 | * think about requeuing jobs when daemon-reload is issued? usecase: | |
687 | the initrd issues a reload after fstab from the host is accessible | |
688 | and we might want to requeue the mounts local-fs acquired through | |
689 | that automatically. | |
690 | ||
691 | * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep() | |
692 | ||
693 | * remove any syslog support from log.c — we probably cannot do this before split-off udev is gone for good | |
694 | ||
695 | * shutdown logging: store to EFI var, and store to USB stick? | |
696 | ||
697 | * merge unit_kill_common() and unit_kill_context() | |
698 | ||
699 | * introduce ExecCondition= in services | |
700 | ||
701 | * EFI: | |
702 | - honor language efi variables for default language selection (if there are any?) | |
703 | - honor timezone efi variables for default timezone selection (if there are any?) | |
704 | - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables | |
705 | * bootctl | |
706 | - recognize the case when not booted on EFI | |
707 | ||
708 | * maybe do not install getty@tty1.service symlink in /etc but in /usr? | |
709 | ||
710 | * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word | |
711 | ||
712 | * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units. | |
713 | ||
714 | * logind: | |
715 | - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around | |
716 | - logind: wakelock/opportunistic suspend support | |
717 | - Add pretty name for seats in logind | |
718 | - logind: allow showing logout dialog from system? | |
719 | - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly. | |
720 | - if pam_systemd is invoked by su from a process that is outside of a | |
721 | any session we should probably just become a NOP, since that's | |
722 | usually not a real user session but just some system code that just | |
723 | needs setuid(). | |
724 | - logind: make the Suspend()/Hibernate() bus calls wait for the for | |
725 | the job to be completed. before returning, so that clients can wait | |
726 | for "systemctl suspend" to finish to know when the suspending is | |
727 | complete. | |
728 | - logind: when the power button is pressed short, just popup a | |
729 | logout dialog. If it is pressed for 1s, do the usual | |
730 | shutdown. Inspiration are Macs here. | |
731 | - expose "Locked" property on logind sesison objects | |
732 | - maybe allow configuration of the StopTimeout for session scopes | |
733 | - rename session scope so that it includes the UID. THat way | |
734 | the session scope can be arranged freely in slices and we don't have | |
735 | make assumptions about their slice anymore. | |
736 | - follow PropertiesChanged state more closely, to deal with quick logouts and | |
737 | relogins | |
738 | ||
739 | * journal: | |
740 | - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields | |
741 | - import and delete pstore filesystem content at startup | |
742 | - journald: also get thread ID from client, plus thread name | |
743 | - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups | |
744 | - add API to close/reopen/get fd for journal client fd in libsystemd-journal. | |
745 | - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively? | |
746 | - declare the local journal protocol stable in the wiki interface chart | |
747 | - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg | |
748 | - journald: when dropping msgs due to ratelimit make sure to write | |
749 | "dropped %u messages" not only when we are about to print the next | |
750 | message that works, but already after a short timeout | |
751 | - check if we can make journalctl by default use --follow mode inside of less if called without args? | |
752 | - maybe add API to send pairs of iovecs via sd_journal_send | |
753 | - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access | |
754 | - journactl: support negative filtering, i.e. FOOBAR!="waldo", | |
755 | and !FOOBAR for events without FOOBAR. | |
756 | - journal: store timestamp of journal_file_set_offline() int he header, | |
757 | so it is possible to display when the file was last synced. | |
758 | - 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. | |
759 | - journal: find a way to allow dropping history early, based on priority, other rules | |
760 | - journal: When used on NFS, check payload hashes | |
761 | - journald: add kernel cmdline option to disable ratelimiting for debug purposes | |
762 | - refuse taking lower-case variable names in sd_journal_send() and friends. | |
763 | - journald: we currently rotate only after MaxUse+MaxFilesize has been reached. | |
764 | - journal: deal nicely with byte-by-byte copied files, especially regards header | |
765 | - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit | |
766 | - Replace utmp, wtmp, btmp, and lastlog completely with journal | |
767 | - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax? | |
768 | - when a kernel driver logs in a tight loop, we should ratelimit that too. | |
769 | - journald: optionally, log debug messages to /run but everything else to /var | |
770 | - journald: when we drop syslog messages because the syslog socket is | |
771 | full, make sure to write how many messages are lost as first thing | |
772 | to syslog when it works again. | |
773 | - change systemd-journal-flush into a service that stays around during | |
774 | boot, and causes the journal to be moved back to /run on shutdown, | |
775 | so that we do not keep /var busy. This needs to happen synchronously, | |
776 | hence doing this via signals is not going to work. | |
777 | - optionally support running journald from the command line for testing purposes in external projects | |
778 | - journald: allow per-priority and per-service retention times when rotating/vacuuming | |
779 | - journald: make use of uid-range.h to managed uid ranges to split | |
780 | journals in. | |
781 | - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so... | |
782 | - improve journalctl performance by loading journal files | |
783 | lazily. Encode just enough information in the file name, so that we | |
784 | do not have to open it to know that it is not interesting for us, for | |
785 | the most common operations. | |
786 | - man: document that corrupted journal files is nothing to act on | |
787 | - rework journald sigbus stuff to use mutex | |
788 | - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our | |
789 | services that run under their own user ids, and use User= (but only | |
790 | in a world where userns is ubiquitous since otherwise we cannot | |
791 | invoke those daemons on the host AND in a container anymore). Also, | |
792 | if LimitNPROC= is used without User= we should warn and refuse | |
793 | operation. | |
794 | - journalctl --verify: don't show files that are currently being | |
795 | written to as FAIL, but instead show that their are being written to. | |
796 | - add journalctl -H that talks via ssh to a remote peer and passes through | |
797 | binary logs data | |
798 | - add a version of --merge which also merges /var/log/journal/remote | |
799 | - journalctl: -m should access container journals directly by enumerating | |
800 | them via machined, and also watch containers coming and going. | |
801 | Benefit: nspawn --ephemeral would start working nicely with the journal. | |
802 | - assign MESSAGE_ID to log messages about failed services | |
803 | ||
804 | * add a test if all entries in the catalog are properly formatted. | |
805 | (Adding dashes in a catalog entry currently results in the catalog entry | |
806 | being silently skipped. journalctl --update-catalog must warn about this, | |
807 | and we should also have a unit test to check that all our message are OK.) | |
808 | ||
809 | * document: | |
810 | - document that deps in [Unit] sections ignore Alias= fields in | |
811 | [Install] units of other units, unless those units are disabled | |
812 | - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets | |
813 | - document that service reload may be implemented as service reexec | |
814 | - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr. | |
815 | - document systemd-journal-flush.service properly | |
816 | - documentation: recommend to connect the timer units of a service to the service via Also= in [Install] | |
817 | - man: document the very specific env the shutdown drop-in tools live in | |
818 | - man: add more examples to man pages | |
819 | - man: maybe sort directives in man pages, and take sections from --help and apply them to man too | |
820 | - document root=gpt-auto properly | |
821 | ||
822 | * systemctl: | |
823 | - add systemctl switch to dump transaction without executing it | |
824 | - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done | |
825 | - "systemctl disable" on a static unit prints no message and does | |
826 | nothing. "systemctl enable" does nothing, and gives a bad message | |
827 | about it. Should fix both to print nice actionable messages. | |
828 | - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service | |
829 | - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible | |
830 | - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards? | |
831 | - systemctl: "Journal has been rotated since unit was started." message is misleading | |
832 | - systemctl status output should include list of triggering units and their status | |
833 | ||
834 | * unit install: | |
835 | - "systemctl mask" should find all names by which a unit is accessible | |
836 | (i.e. by scanning for symlinks to it) and link them all to /dev/null | |
837 | ||
838 | * timer units: | |
839 | - timer units should get the ability to trigger when: | |
840 | o DST changes | |
841 | - Modulate timer frequency based on battery state | |
842 | ||
843 | * add libsystemd-password or so to query passwords during boot using the password agent logic | |
844 | ||
845 | * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed | |
846 | ||
847 | * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel | |
848 | ||
849 | * make repeated alt-ctrl-del presses printing a dump | |
850 | ||
851 | * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ... | |
852 | ||
853 | * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not | |
854 | ||
855 | * nspawn: | |
856 | - emulate /dev/kmsg using CUSE and turn off the syslog syscall | |
857 | with seccomp. That should provide us with a useful log buffer that | |
858 | systemd can log to during early boot, and disconnect container logs | |
859 | from the kernel's logs. | |
860 | - as soon as networkd has a bus interface, hook up --network-interface=, | |
861 | --network-bridge= with networkd, to trigger netdev creation should an | |
862 | interface be missing | |
863 | - a nice way to boot up without machine id set, so that it is set at boot | |
864 | automatically for supporting --ephemeral. Maybe hash the host machine id | |
865 | together with the machine name to generate the machine id for the container | |
866 | - fix logic always print a final newline on output. | |
867 | https://github.com/systemd/systemd/pull/272#issuecomment-113153176 | |
868 | - should optionally support receiving WATCHDOG=1 messages from its payload | |
869 | PID 1... | |
870 | - optionally automatically add FORWARD rules to iptables whenever nspawn is | |
871 | running, remove them when shut down. | |
872 | ||
873 | * dissect | |
874 | - refuse mounting over a mount point | |
875 | - automatically discover .roothash files in dissect, similarly to nspawn | |
876 | ||
877 | * machined: | |
878 | - add an API so that libvirt-lxc can inform us about network interfaces being | |
879 | removed or added to an existing machine | |
880 | - "machinectl migrate" or similar to copy a container from or to a | |
881 | difference host, via ssh | |
882 | - introduce systemd-nspawn-ephemeral@.service, and hook it into | |
883 | "machinectl start" with a new --ephemeral switch | |
884 | - "machinectl status" should also show internal logs of the container in | |
885 | question | |
886 | - "machinectl history" | |
887 | - "machinectl diff" | |
888 | - "machinectl commit" that takes a writable snapshot of a tree, invokes a | |
889 | shell in it, and marks it read-only after use | |
890 | ||
891 | * cryptsetup: | |
892 | - cryptsetup-generator: allow specification of passwords in crypttab itself | |
893 | - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator | |
894 | ||
895 | * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it | |
896 | https://bugs.freedesktop.org/show_bug.cgi?id=54712 | |
897 | ||
898 | * add a dependency on standard-conf.xml and other included files to man pages | |
899 | ||
900 | * MountFlags=shared acts as MountFlags=slave right now. | |
901 | ||
902 | * properly handle loop back mounts via fstab, especially regards to fsck/passno | |
903 | ||
904 | * initialize the hostname from the fs label of /, if /etc/hostname does not exist? | |
905 | ||
906 | * udev: | |
907 | - move to LGPL | |
908 | - kill scsi_id | |
909 | - add trigger --subsystem-match=usb/usb_device device | |
910 | - reimport udev db after MOVE events for devices without dev_t | |
911 | ||
912 | * There's currently no way to cancel fsck (used to be possible via C-c or c on the console) | |
913 | ||
914 | * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/ | |
915 | ||
916 | * coredump: | |
917 | - save coredump in Windows/Mozilla minidump format | |
918 | - when truncating coredumps, also log the full size that the process had, and make a metadata field so we can report truncated coredumps | |
919 | ||
920 | * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting) | |
921 | ||
922 | * be able to specify a forced restart of service A where service B depends on, in case B | |
923 | needs to be auto-respawned? | |
924 | ||
925 | * tmpfiles: | |
926 | - apply "x" on "D" too (see patch from William Douglas) | |
927 | - replace F with f+. | |
928 | - instead of ignoring unknown fields, reject them. | |
929 | - creating new directories/subvolumes/fifos/device nodes | |
930 | should not follow symlinks. None of the other adjustment or creation | |
931 | calls follow symlinks. | |
932 | ||
933 | * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early | |
934 | ||
935 | * verify that the AF_UNIX sockets of a service in the fs still exist | |
936 | when we start a service in order to avoid confusion when a user | |
937 | assumes starting a service is enough to make it accessible | |
938 | ||
939 | * Make it possible to set the keymap independently from the font on | |
940 | the kernel cmdline. Right now setting one resets also the other. | |
941 | ||
942 | * and a dbus call to generate target from current state | |
943 | ||
944 | * write blog stories about: | |
945 | - hwdb: what belongs into it, lsusb | |
946 | - enabling dbus services | |
947 | - how to make changes to sysctl and sysfs attributes | |
948 | - remote access | |
949 | - how to pass throw-away units to systemd, or dynamically change properties of existing units | |
950 | - testing with Harald's awesome test kit | |
951 | - auto-restart | |
952 | - how to develop against journal browsing APIs | |
953 | - the journal HTTP iface | |
954 | - non-cgroup resource management | |
955 | - dynamic resource management with cgroups | |
956 | - refreshed, longer missions statement | |
957 | - calendar time events | |
958 | - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell | |
959 | - how to create your own target | |
960 | - instantiated apache, dovecot and so on | |
961 | - hooking a script into various stages of shutdown/rearly booot | |
962 | ||
963 | * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support. | |
964 | ||
965 | * dot output for --test showing the 'initial transaction' | |
966 | ||
967 | * pid1: | |
968 | - When logging about multiple units (stopping BoundTo units, conflicts, etc.), | |
969 | log both units as UNIT=, so that journalctl -u triggers on both. | |
970 | - generate better errors when people try to set transient properties | |
971 | that are not supported... | |
972 | http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html | |
973 | - maybe introduce WantsMountsFor=? Usecase: | |
974 | http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html | |
975 | - recreate systemd's D-Bus private socket file on SIGUSR2 | |
976 | - move PAM code into its own binary | |
977 | - when we automatically restart a service, ensure we restart its rdeps, too. | |
978 | - hide PAM options in fragment parser when compile time disabled | |
979 | - Support --test based on current system state | |
980 | - 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(). | |
981 | - after deserializing sockets in socket.c we should reapply sockopts and things | |
982 | - drop PID 1 reloading, only do reexecing (difficult: Reload() | |
983 | currently is properly synchronous, Reexec() is weird, because we | |
984 | cannot delay the response properly until we are back, so instead of | |
985 | being properly synchronous we just keep open the fd and close it | |
986 | when done. That means clients do not get a successful method reply, | |
987 | but much rather a disconnect on success. | |
988 | - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr | |
989 | - when a bus name of a service disappears from the bus make sure to queue further activation requests | |
990 | ||
991 | * unit files: | |
992 | - allow port=0 in .socket units | |
993 | - maybe introduce ExecRestartPre= | |
994 | - add ReloadSignal= for configuring a reload signal to use | |
995 | - implement Register= switch in .socket units to enable registration | |
996 | in Avahi, RPC and other socket registration services. | |
997 | - allow Type=simple with PIDFile= | |
998 | https://bugzilla.redhat.com/show_bug.cgi?id=723942 | |
999 | - allow writing multiple conditions in unit files on one line | |
1000 | - load-fragment: when loading a unit file via a chain of symlinks | |
1001 | verify that it is not masked via any of the names traversed. | |
1002 | - introduce Type=pid-file | |
1003 | - introduce mix of BindTo and Requisite | |
1004 | - add a concept of RemainAfterExit= to scope units | |
1005 | - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely | |
1006 | ||
1007 | * udev-link-config: | |
1008 | - Make sure ID_PATH is always exported and complete for | |
1009 | network devices where possible, so we can safely rely | |
1010 | on Path= matching | |
1011 | - check MTUBytes parsing (expecting size_t but we are using unsigned) | |
1012 | ||
1013 | * sd-rtnl: | |
1014 | - add support for more attribute types | |
1015 | - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places | |
1016 | ||
1017 | * networkd: | |
1018 | - add more keys to [Route] and [Address] sections | |
1019 | - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config) | |
1020 | - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline) | |
1021 | - add reduced [Link] support to .network files | |
1022 | - add Scope= parsing option for [Network] | |
1023 | - properly handle routerless dhcp leases | |
1024 | - work with non-Ethernet devices | |
1025 | - add support for more bond options | |
1026 | - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from? | |
1027 | - the DHCP lease data (such as NTP/DNS) is still made available when | |
1028 | a carrier is lost on a link. It should be removed instantly. | |
1029 | - expose in the API the following bits: | |
1030 | - option 15, domain name and/or option 119, search list | |
1031 | - option 12, host name and/or option 81, fqdn | |
1032 | - option 123, 144, geolocation | |
1033 | - option 252, configure http proxy (PAC/wpad) | |
1034 | - provide a way to define a per-network interface default metric value | |
1035 | for all routes to it. possibly a second default for DHCP routes. | |
1036 | - allow Name= to be specified repeatedly in the [Match] section. Maybe also | |
1037 | support Name=foo*|bar*|baz ? | |
1038 | - duplicate address check for static IPs (like ARPCHECK in network-scripts) | |
1039 | - allow DUID/IAID to be customized, see issue #394. | |
1040 | - whenever uplink info changes, make DHCP server send out FORCERENEW | |
1041 | ||
1042 | * networkd-wait-online: | |
1043 | - make operstates to wait for configurable? | |
1044 | ||
1045 | * dhcp: | |
1046 | - figure out how much we can increase Maximum Message Size | |
1047 | ||
1048 | * dhcp6: | |
1049 | - add functions to set previously stored IPv6 addresses on startup and get | |
1050 | them at shutdown; store them in client->ia_na | |
1051 | - write more test cases | |
1052 | - implement reconfigure support, see 5.3., 15.11. and 22.20. | |
1053 | - implement support for temporary adressess (IA_TA) | |
1054 | - implement dhcpv6 authentication | |
1055 | - investigate the usefulness of Confirm messages; i.e. are there any | |
1056 | situations where the link changes without any loss in carrier detection | |
1057 | or interface down | |
1058 | - some servers don't do rapid commit without a filled in IA_NA, verify | |
1059 | this behavior | |
1060 | - RouteTable= ? | |
1061 | ||
1062 | External: | |
1063 | ||
1064 | * dbus: | |
1065 | - natively watch for dbus-*.service symlinks (PENDING) | |
1066 | - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service | |
1067 | ||
1068 | * fix alsa mixer restore to not print error when no config is stored | |
1069 | ||
1070 | * make cryptsetup lower --iter-time | |
1071 | ||
1072 | * patch kernel for xattr support in /dev, /proc/, /sys? | |
1073 | ||
1074 | * kernel: add device_type = "fb", "fbcon" to class "graphics" | |
1075 | ||
1076 | * /usr/bin/service should actually show the new command line | |
1077 | ||
1078 | * 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= | |
1079 | ||
1080 | * fedora: F20: go timer units all the way, leave cron.daily for cron | |
1081 | ||
1082 | * neither pkexec nor sudo initialize environ[] from the PAM environment? | |
1083 | ||
1084 | * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it | |
1085 | ||
1086 | * register catalog database signature as file magic | |
1087 | ||
1088 | * zsh shell completion: | |
1089 | - <command> <verb> -<TAB> should complete options, but currently does not | |
1090 | - systemctl add-wants,add-requires | |
1091 | ||
1092 | ||
1093 | Regularly: | |
1094 | ||
1095 | * look for close() vs. close_nointr() vs. close_nointr_nofail() | |
1096 | ||
1097 | * check for strerror(r) instead of strerror(-r) | |
1098 | ||
1099 | * pahole | |
1100 | ||
1101 | * set_put(), hashmap_put() return values check. i.e. == 0 does not free()! | |
1102 | ||
1103 | * use secure_getenv() instead of getenv() where appropriate | |
1104 | ||
1105 | * link up selected blog stories from man pages and unit files Documentation= fields | |
1106 | String is not UTF-8 clean, ignoring assignment | |
1107 | timedatex.service: Consumed 26ms CPU time. |