]>
Commit | Line | Data |
---|---|---|
afe3ab58 | 1 | Bugfixes: |
c343be28 | 2 | |
b3ae710c ZJS |
3 | * Should systemctl status \* work on all unit types, not just .service? |
4 | ||
f957632b KS |
5 | * Dangling symlinks of .automount unit files in .wants/ directories, set up |
6 | automount points even when the original .automount file did not exist | |
7 | anymore. Only the .mount unit was still around. | |
8 | ||
8adaf7bd | 9 | * ExecStart with unicode characters fails in strv_split_extract: |
d19e85f0 ZJS |
10 | |
11 | [Service] | |
12 | Environment=ONE='one' "TWO='two two' too" THREE= | |
13 | ExecStart=/bin/python3 -c 'import sys;print(sys.argv)' $ONE $TWO $THREE | |
14 | ||
f38afcd0 | 15 | External: |
f85857df | 16 | |
f38afcd0 | 17 | * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros. |
bafb15ba | 18 | |
83a1ff25 ZJS |
19 | * wiki: update journal format documentation for lz4 additions |
20 | ||
5e524b40 LP |
21 | Janitorial Clean-ups: |
22 | ||
5e524b40 LP |
23 | * Rearrange tests so that the various test-xyz.c match a specific src/basic/xyz.c again |
24 | ||
fd6c2363 | 25 | Features: |
47ee3ee0 | 26 | |
fd63f36c LP |
27 | * add some optional flag to ReadWritePaths= and friends, that has the effect |
28 | that we create the dir in question when the service is started. Example: | |
29 | ||
30 | ReadWritePaths=:/var/lib/foobar | |
31 | ||
6fd1a291 LP |
32 | * sort generated hwdb files alphabetically when we import them, so that git |
33 | diffs remain minimal (in particular: the OUI databases we import are not | |
34 | sorted, and not stable) | |
35 | ||
b8076e3d LP |
36 | * set SystemCallArchitectures=native on all our services |
37 | ||
3576dddf LP |
38 | * maybe add call sd_journal_set_block_timeout() or so to set SO_SNDTIMEO for |
39 | the sd-journal logging socket, and, if the timeout is set to 0, sets | |
40 | O_NONBLOCK on it. That way people can control if and when to block for | |
41 | logging. | |
42 | ||
5e806f45 LP |
43 | * tighten sd_notify() MAINPID= checks a bit: don't accept foreign PIDs (i.e. |
44 | PIDs not managed by the service manager) | |
45 | ||
1d010699 LP |
46 | * journald: when we recv a log datagram via the native or syslog transports, |
47 | search for the PID in the active stream connections, and let's make sure to | |
48 | always process the datagrams before the streams. Then, cache client metadata | |
49 | per stream in the stream object. This way we can somewhat fix the race with | |
50 | quickly exiting processes which log as long as they had their own stream | |
51 | connection... | |
52 | ||
fe6a0235 LP |
53 | * hostnamed: populate form factor data from a new hwdb database, so that old |
54 | yogas can be recognized as "convertible" too, even if they predate the DMI | |
55 | "convertible" form factor | |
56 | ||
08422e63 LP |
57 | * Maybe add a small tool invoked early at boot, that adds in or resizes |
58 | partitions automatically, to be used when the media used is actually larger | |
59 | than the image written onto it is. | |
60 | ||
d543b9f5 LP |
61 | * Maybe add PrivatePIDs= as new unit setting, and do minimal PID namespacing |
62 | after all. Be strict however, only support the equivalent of nspawn's | |
63 | --as-pid2 switch, and sanely proxy sd_notify() messages dropping stuff such | |
64 | as MAINPID. | |
65 | ||
08422e63 LP |
66 | * change the dependency Set* objects in Unit structures to become Hashmap*, and |
67 | then store a bit mask who created a specific dependency: the source unit via | |
68 | fragment configuration, the destination unit via fragment configuration, or | |
69 | the source unit via udev rules (in case of .device units), or any combination | |
6b3d3783 | 70 | thereof. This information can then be used to flush out old udev-created |
08422e63 LP |
71 | dependencies when the udev properties change, and eventually to implement a |
72 | "systemctl refresh" operation for reloading the configuration of individual | |
73 | units without reloading the whole set. | |
74 | ||
f59d1da8 LP |
75 | * Add ExecMonitor= setting. May be used multiple times. Forks off a process in |
76 | the service cgroup, which is supposed to monitor the service, and when it | |
77 | exits the service is considered failed by its monitor. | |
78 | ||
33bac67b LP |
79 | * track the per-service PAM process properly (i.e. as an additional control |
80 | process), so that it may be queried on the bus and everything. | |
81 | ||
82 | * add a new "debug" job mode, that is propagated to unit_start() and for | |
83 | services results in two things: we raise SIGSTOP right before invoking | |
84 | execve() and turn off watchdog support. Then, use that to implement | |
85 | "systemd-gdb" for attaching to the start-up of any system service in its | |
86 | natural habitat. | |
87 | ||
0c426957 LP |
88 | * replace all canonicalize_file_name() invocations by chase_symlinks(), in |
89 | particulr those where a rootdir is relevant. | |
90 | ||
8eb7383b LP |
91 | * maybe introduce gpt auto discovery for /var/tmp? |
92 | ||
93 | * set ProtectSystem=strict for all our usual services. | |
94 | ||
6a1da642 LP |
95 | * fix PrivateNetwork= so that we fall back gracefully on kernels lacking |
96 | namespacing support (similar for the other namespacing options) | |
97 | ||
8eb7383b LP |
98 | * maybe add gpt-partition-based user management: each user gets his own |
99 | LUKS-encrypted GPT partition with a new GPT type. A small nss module | |
100 | enumerates users via udev partition enumeration. UIDs are assigned in a fixed | |
101 | way: the partition index is added as offset to some fixed base uid. User name | |
102 | is stored in GPT partition name. A PAM module authenticates the user via the | |
103 | LUKS partition password. Benefits: strong per-user security, compatibility | |
104 | with stateless/read-only/verity-enabled root. (other idea: do this based on | |
105 | loopback files in /home, without GPT involvement) | |
106 | ||
107 | * gpt-auto logic: introduce support for discovering /var matching an image. For | |
108 | that, use a partition type UUID that is hashed from the OS name (as encoded | |
109 | in /etc/os-release), the architecture, and 4 new bits from the gpt flags | |
110 | field of the root partition. This way can easily support multiple OS | |
111 | installations on the same GPT partition table, without problems with | |
112 | unmatched /var partitions. | |
113 | ||
114 | * gpt-auto logic: related to the above, maybe support a "secondary" root | |
115 | partition, that is mounted to / and is writable, and where the actual root's | |
116 | /usr is mounted into. | |
117 | ||
8eb7383b LP |
118 | * machined: add apis to query /etc/machine-info data of a container |
119 | ||
120 | * .mount and .swap units: add Format=yes|no option that formats the partition before mounting/enabling it, implicitly | |
121 | ||
122 | * 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 | |
123 | ||
67234d21 LP |
124 | * drop nss-myhostname in favour of nss-resolve? |
125 | ||
126 | * drop internal dlopen() based nss-dns fallback in nss-resolve, and rely on the | |
127 | external nsswitch.conf based one | |
128 | ||
b5bdbcd5 LP |
129 | * add a percentage syntax for TimeoutStopSec=, e.g. TimeoutStopSec=150%, and |
130 | then use that for the setting used in user@.service. It should be understood | |
131 | relative to the configured default value. | |
132 | ||
642d6665 LP |
133 | * on cgroupsv2 add DelegateControllers=, to pick the precise cgroup controllers to delegate |
134 | ||
d21494ea LP |
135 | * in networkd, when matching device types, fix up DEVTYPE rubbish the kernel passes to us |
136 | ||
137 | * enable LockMLOCK to take a percentage value relative to physical memory | |
138 | ||
04397464 LP |
139 | * switch to ProtectSystem=strict for all our long-running services where that's possible |
140 | ||
04397464 | 141 | * Permit masking specific netlink APIs with RestrictAddressFamily= |
d82047be LP |
142 | |
143 | * nspawn: start UID allocation loop from hash of container name | |
144 | ||
d82047be LP |
145 | * nspawn: support that /proc, /sys/, /dev are pre-mounted |
146 | ||
d82047be LP |
147 | * define gpt header bits to select volatility mode |
148 | ||
42d61ded LP |
149 | * ProtectKernelLogs= (drops CAP_SYSLOG, add seccomp for syslog() syscall, and DeviceAllow to /dev/kmsg) in service files |
150 | ||
151 | * ProtectClock= (drops CAP_SYS_TIMES, adds seecomp filters for settimeofday, adjtimex), sets DeviceAllow o /dev/rtc | |
152 | ||
04397464 | 153 | * ProtectTracing= (drops CAP_SYS_PTRACE, blocks ptrace syscall, makes /sys/kernel/tracing go away) |
42d61ded | 154 | |
04397464 | 155 | * ProtectMount= (drop mount/umount/pivot_root from seccomp, disallow fuse via DeviceAllow, imply Mountflags=slave) |
563a69f4 | 156 | |
04397464 | 157 | * ProtectKeyRing= to take keyring calls away |
2c5f2958 | 158 | |
8ce9b83a | 159 | * RemoveKeyRing= to remove all keyring entries of the specified user |
a46eac1b LP |
160 | |
161 | * ProtectReboot= that masks reboot() and kexec_load() syscalls, prohibits kill | |
162 | on PID 1 with the relevant signals, and makes relevant files in /sys and | |
163 | /proc (such as the sysrq stuff) unavailable | |
164 | ||
165 | * DeviceAllow= should also generate seccomp filters for mknod() | |
8ce9b83a | 166 | |
563a69f4 LP |
167 | * Add DataDirectory=, CacheDirectory= and LogDirectory= to match |
168 | RuntimeDirectory=, and create it as necessary when starting a service, owned by the right user. | |
169 | ||
e40a326c | 170 | * make sure the ratelimit object can deal with USEC_INFINITY as way to turn off things |
89f193fa | 171 | |
6809de5b LP |
172 | * journalctl: make sure -f ends when the container indicated by -M terminates |
173 | ||
2c5f2958 LP |
174 | * mount: automatically search for "main" partition of an image has multiple |
175 | partitions | |
176 | ||
f8afc2a9 LP |
177 | * expose the "privileged" flag of ExecCommand on the bus, and open it up to |
178 | transient units | |
179 | ||
8ce9b83a LP |
180 | * in nss-systemd, if we run inside of RootDirectory= with PrivateUsers= set, |
181 | find a way to map the User=/Group= of the service to the right name. This way | |
182 | a user/group for a service only has to exist on the host for the right | |
183 | mapping to work. | |
184 | ||
36376e0b LP |
185 | * allow attaching additional journald log fields to cgroups |
186 | ||
2c5f2958 LP |
187 | * add bus API for creating unit files in /etc, reusing the code for transient units |
188 | ||
189 | * add bus API to remove unit files from /etc | |
190 | ||
191 | * add bus API to retrieve current unit file contents (i.e. implement "systemctl cat" on the bus only) | |
192 | ||
b8c7afdf LP |
193 | * rework fopen_temporary() to make use of open_tmpfile_linkable() (problem: the |
194 | kernel doesn't support linkat() that replaces existing files, currently) | |
195 | ||
b8c7afdf LP |
196 | * check if DeviceAllow= should split first, resolve specifiers later |
197 | ||
1e555cb5 LP |
198 | * transient units: don't bother with actually setting unit properties, we |
199 | reload the unit file anyway | |
200 | ||
1e555cb5 LP |
201 | * journald: sigbus API via a signal-handler safe function that people may call |
202 | from the SIGBUS handler | |
203 | ||
c8048350 LP |
204 | * optionally, also require WATCHDOG=1 notifications during service start-up and shutdown |
205 | ||
7a79d1ec LP |
206 | * resolved: when routing queries, make sure only look for the *longest* suffix... |
207 | ||
648b1220 LP |
208 | * delay activation of logind until somebody logs in, or when /dev/tty0 pulls it |
209 | in or lingering is on (so that containers don't bother with it until PAM is used). also exit-on-idle | |
210 | ||
3d39e6e5 LP |
211 | * cache sd_event_now() result from before the first iteration... |
212 | ||
f506d09f LP |
213 | * add systemctl stop --job-mode=triggering that follows TRIGGERED_BY deps and adds them to the same transaction |
214 | ||
bd098bce LP |
215 | * PID1: find a way how we can reload unit file configuration for |
216 | specific units only, without reloading the whole of systemd | |
217 | ||
f9bf1b8f | 218 | * add an explicit parser for LimitRTPRIO= that verifies |
de7399eb | 219 | the specified range and generates sane error messages for incorrect |
f9bf1b8f | 220 | specifications. |
de7399eb | 221 | |
0d4c4b71 LP |
222 | * do something about "/control" subcgroups in the unified cgroup hierarchy |
223 | ||
3efc8c72 LP |
224 | * when we detect that there are waiting jobs but no running jobs, do something |
225 | ||
06ce859b | 226 | * push CPUAffinity= also into the "cpuset" cgroup controller (only after the cpuset controller got ported to the unified hierarchy) |
03364e47 | 227 | |
e6a26d8c LP |
228 | * PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn) |
229 | ||
a2088fd0 LP |
230 | * there's probably something wrong with having user mounts below /sys, |
231 | as we have for debugfs. for exmaple, src/core/mount.c handles mounts | |
232 | prefixed with /sys generally special. | |
233 | http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html | |
234 | ||
eedb4ac8 LP |
235 | * man: document that unless you use StandardError=null the shell >/dev/stderr won't work in shell scripts in services |
236 | ||
8aa20381 LP |
237 | * fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline |
238 | ||
b18d23d7 LP |
239 | * docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date |
240 | ||
449a22a5 | 241 | * mounting and unmounting mount points manually with different source |
61233823 | 242 | devices will result in collected on all devices used. |
449a22a5 LP |
243 | http://lists.freedesktop.org/archives/systemd-devel/2015-April/030225.html |
244 | ||
60d17b74 LP |
245 | * add a job mode that will fail if a transaction would mean stopping |
246 | running units. Use this in timedated to manage the NTP service | |
247 | state. | |
248 | http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html | |
249 | ||
2f653bde LP |
250 | * Maybe add support for the equivalent of "ethtool advertise" to .link files? |
251 | http://lists.freedesktop.org/archives/systemd-devel/2015-April/030112.html | |
252 | ||
477e75ef LP |
253 | * The udev blkid built-in should expose a property that reflects |
254 | whether media was sensed in USB CF/SD card readers. This should then | |
255 | be used to control SYSTEMD_READY=1/0 so that USB card readers aren't | |
256 | picked up by systemd unless they contain a medium. This would mirror | |
257 | the behaviour we already have for CD drives. | |
258 | ||
2bc39683 LP |
259 | * networkd/udev: implement SR_IOV configuration in .link files: |
260 | http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html | |
261 | ||
17fe5614 LP |
262 | * Rework systemctl's GetAll property parsing to use the generic bus_map_all_properties() API |
263 | ||
39609489 | 264 | * implement a per-service firewall based on net_cls |
c14db9b3 | 265 | |
54ff1d69 | 266 | * Port various tools to make use of verbs.[ch], where applicable: busctl, |
04397464 | 267 | coredumpctl, hostnamectl, localectl, systemd-analyze, timedatectl |
c3a0d00d | 268 | |
c3a0d00d LP |
269 | * hostnamectl: show root image uuid |
270 | ||
82e6c50c LP |
271 | * sysfs set api in libudev is not const |
272 | ||
d2f81fb0 LP |
273 | * Find a solution for SMACK capabilities stuff: |
274 | http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html | |
98cd2651 | 275 | |
2a441c8a LP |
276 | * "systemctl preset-all" should probably order the unit files it |
277 | operates on lexicographically before starting to work, in order to | |
278 | ensure deterministic behaviour if two unit files conflict (like DMs | |
279 | do, for example) | |
280 | ||
0a86c1a9 LP |
281 | * synchronize console access with BSD locks: |
282 | http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html | |
283 | ||
e031c227 | 284 | * as soon as we have sender timestamps, revisit coalescing multiple parallel daemon reloads: |
0a86c1a9 LP |
285 | http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html |
286 | ||
e3286870 LP |
287 | * in systemctl list-unit-files: show the install value the presets would suggest for a service in a third column |
288 | ||
8514b677 LP |
289 | * figure out when we can use the coarse timers |
290 | ||
7b0fce61 LP |
291 | * add "systemctl start -v foobar.service" that shows logs of a service |
292 | while the start command runs. This is non-trivial to do without | |
293 | races though, since we should flush out all journal messages before | |
294 | returning from the "systemctl stop". | |
295 | ||
b568ef14 | 296 | * 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 |
c2d5b3c9 | 297 | |
d48b7bd2 LP |
298 | * maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment= |
299 | ||
758c4d7a LP |
300 | * introduce systemd-timesync-wait.service or so to sync on an NTP fix? |
301 | ||
821cc13d LP |
302 | * systemd --user should issue sd_notify() upon reaching basic.target, not on becoming idle |
303 | ||
8483d73f LP |
304 | * consider showing the unit names during boot up in the status output, not just the unit descriptions |
305 | ||
8483d73f LP |
306 | * maybe allow timer units with an empty Units= setting, so that they |
307 | can be used for resuming the system but nothing else. | |
308 | ||
25e773ee | 309 | * what to do about udev db binary stability for apps? (raw access is not an option) |
b857e042 | 310 | |
6dab5bb1 LP |
311 | * maybe provide an API to allow migration of foreign PIDs into existing scopes. |
312 | ||
3bcde97e LP |
313 | * man: maybe use the word "inspect" rather than "introspect"? |
314 | ||
279f0366 | 315 | * systemctl: if some operation fails, show log output? |
533bb267 | 316 | |
39c38ce1 | 317 | * systemctl edit: use equvalent of cat() to insert existing config as a comment, prepended with #. |
b42de08a | 318 | Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc. |
202aea45 | 319 | |
720652b3 | 320 | * exponential backoff in timesyncd when we cannot reach a server |
42aeb14a | 321 | |
720652b3 | 322 | * timesyncd: add ugly bus calls to set NTP servers per-interface, for usage by NM |
e25b5a8d | 323 | |
563b1bdc LP |
324 | * merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share.... |
325 | ||
5dcf9838 LP |
326 | * systemd.show_status= should probably have a mode where only failed |
327 | units are shown. | |
328 | ||
6bd7941e TG |
329 | * add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL |
330 | (throughout the codebase, not only PID1) | |
331 | ||
9d6db739 | 332 | * resolved: |
9d6db739 | 333 | - mDNS/DNS-SD |
ccc3e8a1 LP |
334 | - service registration |
335 | - service/domain/types browsing | |
0f47ed0a | 336 | - avahi compat |
9d6db739 | 337 | - DNS-SD service registration from socket units |
e25b5a8d DH |
338 | - resolved should optionally register additional per-interface LLMNR |
339 | names, so that for the container case we can establish the same name | |
340 | (maybe "host") for referencing the server, everywhere. | |
720652b3 | 341 | - allow clients to request DNSSEC for a single lookup even if DNSSEC is off (?) |
3efb871a | 342 | - hook up resolved with machined-based address resolution |
3f77a1b1 | 343 | |
e25b5a8d | 344 | * refcounting in sd-resolve is borked |
e2a69298 | 345 | |
2834ffe7 LP |
346 | * Add a new verb "systemctl top" |
347 | ||
a940778f LP |
348 | * add new gpt type for btrfs volumes |
349 | ||
a01647e5 | 350 | * support empty /etc boots nicely: |
3c779fa5 | 351 | - nspawn/gpt-generator: introduce new gpt partition type for /usr |
a01647e5 | 352 | |
3de03738 LP |
353 | * generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them. |
354 | ||
37efac5d LP |
355 | * a way for container managers to turn off getty starting via $container_headless= or so... |
356 | ||
7348b3ad LP |
357 | * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit |
358 | ||
e107ed18 LP |
359 | * mount_cgroup_controllers(): symlinks need to get the label applied |
360 | ||
81429136 KS |
361 | * For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services |
362 | they run added to the initial transaction and thus confuse Type=idle. | |
e107ed18 | 363 | |
edb2935c LP |
364 | * add bus api to query unit file's X fields. |
365 | ||
6a3f892a | 366 | * gpt-auto-generator: |
2a781fc9 | 367 | - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap? |
6a3f892a | 368 | - Make /home automount rather than mount? |
6a3f892a | 369 | |
65026403 LP |
370 | * add generator that pulls in systemd-network from containers when |
371 | CAP_NET_ADMIN is set, more than the loopback device is defined, even | |
372 | when it is otherwise off | |
f8901862 | 373 | |
f9bf1b8f | 374 | * MessageQueueMessageSize= (and suchlike) should use parse_iec_size(). |
eda8f067 | 375 | |
af1082b0 LP |
376 | * implement Distribute= in socket units to allow running multiple |
377 | service instances processing the listening socket, and open this up | |
378 | for ReusePort= | |
379 | ||
488ad3b3 LP |
380 | * socket units: support creating sockets in different namespace, |
381 | opening it up for JoinsNamespaceOf=. This would require to fork off | |
382 | a tiny process that joins the namespace and creates/binds the socket | |
383 | and passes this back to PID1 via SCM_RIGHTS. This also could be used | |
384 | to allow Chown/chgrp on sockets without requiring NSS in PID 1. | |
385 | ||
09ecd746 LP |
386 | * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze |
387 | $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls | |
388 | should SIGSTOP all unit processes in a loop until all processes of | |
389 | it are fully stopped. This can later be used for app management by | |
390 | desktop UIs such as gnome-shell to freeze apps that are not visible | |
391 | on screen, not unlike how job control works on the shell | |
392 | ||
f38afcd0 | 393 | * cgroups: |
f38afcd0 LP |
394 | - implement per-slice CPUFairScheduling=1 switch |
395 | - handle jointly mounted controllers correctly | |
396 | - introduce high-level settings for RT budget, swappiness | |
397 | - how to reset dynamically changed unit cgroup attributes sanely? | |
398 | - when reloading configuration, apply new cgroup configuration | |
399 | - when recursively showing the cgroup hierarchy, optionally also show | |
400 | the hierarchies of child processes | |
0bee65f0 | 401 | |
f38afcd0 | 402 | * transient units: |
f38afcd0 LP |
403 | - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt |
404 | - ensure scope units may be started only a single time | |
ebcf1f97 | 405 | |
47c94a96 LP |
406 | * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk? |
407 | ||
718db961 LP |
408 | * when we detect low battery and no AC on boot, show pretty splash and refuse boot |
409 | ||
718db961 LP |
410 | * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops |
411 | ||
966204e0 LP |
412 | * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1 |
413 | ||
41644622 LP |
414 | * 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 |
415 | ||
7f79cd71 | 416 | * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs |
0aafd43d | 417 | |
19aadacf JE |
418 | * If we try to find a unit via a dangling symlink, generate a clean |
419 | error. Currently, we just ignore it and read the unit from the search | |
df5f6971 LP |
420 | path anyway. |
421 | ||
04397464 | 422 | * refuse boot if /usr/lib/os-release is missing or /etc/machine-id cannot be set up |
fcba531e | 423 | |
bdeeb6b5 LP |
424 | * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted. |
425 | ||
426 | * load .d/*.conf dropins for device units | |
427 | ||
4e09014d LP |
428 | * allow implementation of InaccessibleDirectories=/ plus |
429 | ReadOnlyDirectories=... for whitelisting files for a service. | |
430 | ||
6bb648a1 | 431 | * sd-bus: |
0d43ffef | 432 | - EBADSLT handling |
fa607802 | 433 | - GetAllProperties() on a non-existing object does not result in a failure currently |
d0928791 | 434 | - kdbus: process fd=-1 for incoming msgs |
7212c608 | 435 | - port to sd-resolve for connecting to TCP dbus servers |
1f140dd8 | 436 | - kdbus: maybe add controlling tty metadata fields |
57f2a947 | 437 | - 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 |
d28315e4 | 438 | - when kdbus does not take our message without memfds, try again with memfds |
f38afcd0 | 439 | - see if we can drop more message validation on the sending side |
f38afcd0 | 440 | - add API to clone sd_bus_message objects |
dc17bcef | 441 | - make AddMatch calls on dbus1 transports async? |
80caea6c | 442 | - kdbus: matches against source or destination pids for an "strace -p"-like feel. Problem: The PID info needs to be available in userspace too... |
ab9716c2 | 443 | - longer term: priority inheritance |
df98a87b LP |
444 | - dbus spec updates: |
445 | - kdbus mapping | |
446 | - NameLost/NameAcquired obsolete | |
447 | - GVariant | |
2ecfc64e | 448 | - path escaping |
6a8b5fa4 | 449 | - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now |
69727e6d | 450 | - test bloom filter generation indexes |
b568ef14 | 451 | - kdbus: introduce a concept of "send-only" connections |
0ad68f87 | 452 | - kdbus: add counter for refused unicast messages that is passed out via the RECV ioctl. SImilar to the counter for dropped multicast messages we already have. |
f38afcd0 LP |
453 | |
454 | * sd-event | |
12179984 | 455 | - allow multiple signal handlers per signal? |
0536ce5d | 456 | - document chaining of signal handler for SIGCHLD and child handlers |
e25b5a8d DH |
457 | - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ... |
458 | - generate a failure of a default event loop is executed out-of-thread | |
459 | - maybe add support for inotify events | |
eb01ba5d | 460 | |
69af4503 | 461 | * investigate endianness issues of UUID vs. GUID |
e8a7a315 | 462 | |
461bd8e4 LP |
463 | * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we |
464 | should be able to safely try another attempt when the bus call LoadUnit() is invoked. | |
465 | ||
ab8e074c LP |
466 | * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login. |
467 | ||
468 | * add a pam module that on password changes updates any LUKS slot where the password matches | |
469 | ||
b454b112 LP |
470 | * maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases... |
471 | ||
4ff49cb6 LP |
472 | * cgtop: make cgtop useful in a container |
473 | ||
fff87a35 ZJS |
474 | * test/: |
475 | - add 'set -e' to scripts in test/ | |
476 | - make stuff in test/ work with separate output dir | |
b8b4d3dd | 477 | |
b5c03638 | 478 | * seems that when we follow symlinks to units we prefer the symlink |
d28315e4 | 479 | destination path over /etc and /usr. We should not do that. Instead |
b5c03638 LP |
480 | /etc should always override /run+/usr and also any symlink |
481 | destination. | |
482 | ||
eece8c6f LP |
483 | * when isolating, try to figure out a way how we implicitly can order |
484 | all units we stop before the isolating unit... | |
485 | ||
356ce991 LP |
486 | * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off}) |
487 | ||
c1446921 LP |
488 | * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify |
489 | that the boot loader should be installed to the ESP. Define a way | |
490 | how an installer can figure out whether a BLS compliant boot loader | |
491 | is installed. | |
795607b2 | 492 | |
0be8342c LP |
493 | * think about requeuing jobs when daemon-reload is issued? usecase: |
494 | the initrd issues a reload after fstab from the host is accessible | |
495 | and we might want to requeue the mounts local-fs acquired through | |
496 | that automatically. | |
497 | ||
e5ec62c5 | 498 | * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep() |
54c31a79 | 499 | |
ccddd104 | 500 | * remove any syslog support from log.c — we probably cannot do this before split-off udev is gone for good |
826872b6 | 501 | |
3679d112 LP |
502 | * shutdown logging: store to EFI var, and store to USB stick? |
503 | ||
3679d112 LP |
504 | * think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service |
505 | ||
356ce991 | 506 | * merge unit_kill_common() and unit_kill_context() |
490b7e47 | 507 | |
81d112a8 | 508 | * introduce ExecCondition= in services |
490b7e47 | 509 | |
b44be3ec | 510 | * EFI: |
b44be3ec LP |
511 | - honor language efi variables for default language selection (if there are any?) |
512 | - honor timezone efi variables for default timezone selection (if there are any?) | |
466784c8 | 513 | - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables |
e4181484 | 514 | |
c0fe5db5 | 515 | * maybe do not install getty@tty1.service symlink in /etc but in /usr? |
e9fd44b7 | 516 | |
b44be3ec | 517 | * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word |
8ed20651 | 518 | |
b44be3ec | 519 | * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units. |
e6c6e7af | 520 | |
b44be3ec LP |
521 | * logind: |
522 | - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around | |
523 | - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot | |
524 | - logind: wakelock/opportunistic suspend support | |
525 | - Add pretty name for seats in logind | |
526 | - logind: allow showing logout dialog from system? | |
e5ec62c5 | 527 | - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case |
f38afcd0 LP |
528 | - session scopes/user unit: add RequiresMountsFor for the home directory of the user |
529 | - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly. | |
530 | - if pam_systemd is invoked by su from a process that is outside of a | |
531 | any session we should probably just become a NOP, since that's | |
532 | usually not a real user session but just some system code that just | |
533 | needs setuid(). | |
279f0366 LP |
534 | - logind: make the Suspend()/Hibernate() bus calls wait for the for |
535 | the job to be completed. before returning, so that clients can wait | |
536 | for "systemctl suspend" to finish to know when the suspending is | |
537 | complete. | |
538 | - logind: when the power button is pressed short, just popup a | |
539 | logout dialog. If it is pressed for 1s, do the usual | |
540 | shutdown. Inspiration are Macs here. | |
279f0366 LP |
541 | - expose "Locked" property on logind sesison objects |
542 | - given that logind now lets PID 1 do all nasty work, we can | |
543 | probably reduce the capability set it retains substantially. | |
544 | (we need CAP_SYS_ADMIN for drmSetMaster(), so maybe not worth it) | |
b568ef14 | 545 | - expose orientation sensors and tablet mode through logind |
e25b5a8d DH |
546 | - maybe allow configuration of the StopTimeout for session scopes |
547 | - rename session scope so that it includes the UID. THat way | |
548 | the session scope can be arranged freely in slices and we don't have | |
549 | make assumptions about their slice anymore. | |
550 | - follow PropertiesChanged state more closely, to deal with quick logouts and | |
551 | relogins | |
e673ad04 | 552 | |
b44be3ec | 553 | * exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty. |
e673ad04 | 554 | |
b44be3ec | 555 | * service: watchdog logic: for testing purposes allow ping, but do not require pong |
e673ad04 | 556 | |
b44be3ec | 557 | * journal: |
57f2a947 | 558 | - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields |
c06bf414 | 559 | - import and delete pstore filesystem content at startup |
b44be3ec LP |
560 | - journald: also get thread ID from client, plus thread name |
561 | - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups | |
562 | - add API to close/reopen/get fd for journal client fd in libsystemd-journal. | |
d28315e4 | 563 | - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively? |
b44be3ec | 564 | - declare the local journal protocol stable in the wiki interface chart |
b44be3ec LP |
565 | - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg |
566 | - journald: when dropping msgs due to ratelimit make sure to write | |
567 | "dropped %u messages" not only when we are about to print the next | |
568 | message that works, but alraedy after a short tiemout | |
569 | - check if we can make journalctl by default use --follow mode inside of less if called without args? | |
570 | - maybe add API to send pairs of iovecs via sd_journal_send | |
f47ec8eb | 571 | - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access |
b44be3ec LP |
572 | - journactl: support negative filtering, i.e. FOOBAR!="waldo", |
573 | and !FOOBAR for events without FOOBAR. | |
038cf334 ZJS |
574 | - journal: store timestamp of journal_file_set_offline() int he header, |
575 | so it is possible to display when the file was last synced. | |
b44be3ec LP |
576 | - 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. |
577 | - journal: find a way to allow dropping history early, based on priority, other rules | |
578 | - journal: When used on NFS, check payload hashes | |
b44be3ec LP |
579 | - journald: add kernel cmdline option to disable ratelimiting for debug purposes |
580 | - refuse taking lower-case variable names in sd_journal_send() and friends. | |
581 | - journald: we currently rotate only after MaxUse+MaxFilesize has been reached. | |
582 | - journal: deal nicely with byte-by-byte copied files, especially regards header | |
b44be3ec | 583 | - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit |
601d9d6f | 584 | - Replace utmp, wtmp, btmp, and lastlog completely with journal |
f38afcd0 | 585 | - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax? |
f38afcd0 LP |
586 | - when a kernel driver logs in a tight loop, we should ratelimit that too. |
587 | - journald: optionally, log debug messages to /run but everything else to /var | |
588 | - journald: when we drop syslog messages because the syslog socket is | |
589 | full, make sure to write how many messages are lost as first thing | |
590 | to syslog when it works again. | |
591 | - journald: make sure ratelimit is actually really per-service with the new cgroup changes | |
592 | - change systemd-journal-flush into a service that stays around during | |
593 | boot, and causes the journal to be moved back to /run on shutdown, | |
d28315e4 | 594 | so that we do not keep /var busy. This needs to happen synchronously, |
f38afcd0 | 595 | hence doing this via signals is not going to work. |
279f0366 LP |
596 | - optionally support running journald from the command line for testing purposes in external projects |
597 | - journald: allow per-priority and per-service retention times when rotating/vacuuming | |
598 | - journald: make use of uid-range.h to managed uid ranges to split | |
599 | journals in. | |
600 | - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so... | |
601 | - improve journalctl performance by loading journal files | |
602 | lazily. Encode just enough information in the file name, so that we | |
603 | do not have to open it to know that it is not interesting for us, for | |
604 | the most common operations. | |
e25b5a8d DH |
605 | - journal-or-kmsg is currently broken? See reverted |
606 | commit 4a01181e460686d8b4a543b1dfa7f77c9e3c5ab8. | |
607 | - man: document that corrupted journal files is nothing to act on | |
e25b5a8d DH |
608 | - rework journald sigbus stuff to use mutex |
609 | - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our | |
610 | services that run under their own user ids, and use User= (but only | |
611 | in a world where userns is ubiquitous since otherwise we cannot | |
612 | invoke those daemons on the host AND in a container anymore). Also, | |
613 | if LimitNPROC= is used without User= we should warn and refuse | |
614 | operation. | |
615 | - journalctl --verify: don't show files that are currently being | |
616 | written to as FAIL, but instead show that their are being written to. | |
617 | - add journalctl -H that talks via ssh to a remote peer and passes through | |
618 | binary logs data | |
e25b5a8d DH |
619 | - add a version of --merge which also merges /var/log/journal/remote |
620 | - log accumulated resource usage after each service invocation | |
621 | - journalctl: -m should access container journals directly by enumerating | |
622 | them via machined, and also watch containers coming and going. | |
623 | Benefit: nspawn --ephemeral would start working nicely with the journal. | |
624 | - assign MESSAGE_ID to log messages about failed services | |
b44be3ec LP |
625 | |
626 | * document: | |
8b8f2591 | 627 | - document that deps in [Unit] sections ignore Alias= fields in |
b44be3ec LP |
628 | [Install] units of other units, unless those units are disabled |
629 | - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets | |
b44be3ec | 630 | - document the exit codes when services fail before they are exec()ed |
b44be3ec | 631 | - document that service reload may be implemented as service reexec |
eb124a97 | 632 | - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications |
f38afcd0 LP |
633 | - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr. |
634 | - document systemd-journal-flush.service properly | |
f38afcd0 LP |
635 | - documentation: recommend to connect the timer units of a service to the service via Also= in [Install] |
636 | - man: document the very specific env the shutdown drop-in tools live in | |
f38afcd0 LP |
637 | - man: add more examples to man pages |
638 | - man: maybe sort directives in man pages, and take sections from --help and apply them to man too | |
b44be3ec LP |
639 | |
640 | * systemctl: | |
b44be3ec LP |
641 | - add systemctl switch to dump transaction without executing it |
642 | - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done | |
643 | - "systemctl disable" on a static unit prints no message and does | |
644 | nothing. "systemctl enable" does nothing, and gives a bad message | |
645 | about it. Should fix both to print nice actionable messages. | |
646 | - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service | |
647 | - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible | |
d28315e4 | 648 | - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards? |
b44be3ec | 649 | - systemctl: "Journal has been rotated since unit was started." message is misleading |
f38afcd0 | 650 | - better error message if you run systemctl without systemd running |
61233823 | 651 | - systemctl status output should include list of triggering units and their status |
f38afcd0 LP |
652 | |
653 | * unit install: | |
654 | - "systemctl mask" should find all names by which a unit is accessible | |
655 | (i.e. by scanning for symlinks to it) and link them all to /dev/null | |
8b04b925 | 656 | |
b44be3ec | 657 | * timer units: |
b44be3ec LP |
658 | - timer units should get the ability to trigger when: |
659 | o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET) | |
660 | o DST changes | |
90e071d1 | 661 | - Modulate timer frequency based on battery state |
d2e83c23 | 662 | |
a19554ed LP |
663 | * add libsystemd-password or so to query passwords during boot using the password agent logic |
664 | ||
a7a3f28b LP |
665 | * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed |
666 | ||
b44be3ec | 667 | * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel |
08f95888 | 668 | |
f59d1da8 | 669 | * make repeated alt-ctrl-del presses printing a dump |
dcfc4b2e | 670 | |
71ef24d0 LP |
671 | * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ... |
672 | ||
d28315e4 | 673 | * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not |
1b89884b | 674 | |
b44be3ec | 675 | * nspawn: |
e25b5a8d DH |
676 | - emulate /dev/kmsg using CUSE and turn off the syslog syscall |
677 | with seccomp. That should provide us with a useful log buffer that | |
678 | systemd can log to during early boot, and disconnect container logs | |
679 | from the kernel's logs. | |
680 | - as soon as networkd has a bus interface, hook up --network-interface=, | |
681 | --network-bridge= with networkd, to trigger netdev creation should an | |
682 | interface be missing | |
e25b5a8d DH |
683 | - a nice way to boot up without machine id set, so that it is set at boot |
684 | automatically for supporting --ephemeral. Maybe hash the host machine id | |
685 | together with the machine name to generate the machine id for the container | |
686 | - fix logic always print a final newline on output. | |
687 | https://github.com/systemd/systemd/pull/272#issuecomment-113153176 | |
688 | - should optionally support receiving WATCHDOG=1 messages from its payload | |
689 | PID 1... | |
690 | - should send out sd_notify("WATCHDOG=1") messages | |
691 | - optionally automatically add FORWARD rules to iptables whenever nspawn is | |
692 | running, remove them when shut down. | |
e25b5a8d DH |
693 | - Improve error message when --bind= is used on a non-existing source |
694 | directory | |
695 | - maybe make copying of /etc/resolv.conf optional, and skip it if --read-only | |
696 | is used | |
697 | ||
347a17d5 ZJS |
698 | * dissect |
699 | - refuse mounting over a mount point | |
700 | - automatically discover .roothash files in dissect, similarly to nspawn | |
701 | ||
e25b5a8d | 702 | * machined: |
e25b5a8d DH |
703 | - add an API so that libvirt-lxc can inform us about network interfaces being |
704 | removed or added to an existing machine | |
705 | - "machinectl migrate" or similar to copy a container from or to a | |
706 | difference host, via ssh | |
e25b5a8d DH |
707 | - introduce systemd-nspawn-ephemeral@.service, and hook it into |
708 | "machinectl start" with a new --ephemeral switch | |
709 | - "machinectl status" should also show internal logs of the container in | |
710 | question | |
711 | - "machinectl list-images" should show os-release data, as well as | |
712 | machine-info data (including deployment level) | |
713 | - "machinectl history" | |
714 | - "machinectl diff" | |
715 | - "machinectl commit" that takes a writable snapshot of a tree, invokes a | |
716 | shell in it, and marks it read-only after use | |
717 | ||
718 | * importd: | |
e25b5a8d | 719 | - generate a nice warning if mkfs.btrfs is missing |
b44be3ec LP |
720 | |
721 | * cryptsetup: | |
b44be3ec LP |
722 | - cryptsetup-generator: allow specification of passwords in crypttab itself |
723 | - move cryptsetup key caching into kernel keyctl? | |
724 | https://bugs.freedesktop.org/show_bug.cgi?id=54982 | |
5f1dac6b | 725 | - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator |
f8019684 | 726 | |
424a19f8 | 727 | * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it |
41f9172f | 728 | https://bugs.freedesktop.org/show_bug.cgi?id=54712 |
424a19f8 | 729 | |
5e8b2883 LP |
730 | * create /sbin/init symlinks from the build system |
731 | ||
68f16003 | 732 | * MountFlags=shared acts as MountFlags=slave right now. |
0790b9fe | 733 | |
fd4d89b2 LP |
734 | * properly handle loop back mounts via fstab, especially regards to fsck/passno |
735 | ||
47ae7201 LP |
736 | * initialize the hostname from the fs label of /, if /etc/hostname does not exist? |
737 | ||
abd55b16 | 738 | * udev: |
abd55b16 | 739 | - move to LGPL |
abd55b16 KS |
740 | - kill scsi_id |
741 | - add trigger --subsystem-match=usb/usb_device device | |
e8d842a0 | 742 | - reimport udev db after MOVE events for devices without dev_t |
b8217b7b | 743 | |
169c4f65 LP |
744 | * 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 |
745 | ||
7f110ff9 LP |
746 | * There's currently no way to cancel fsck (used to be possible via C-c or c on the console) |
747 | ||
d0e5a333 LP |
748 | * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/ |
749 | ||
e25b5a8d DH |
750 | * coredump: |
751 | - save coredump in Windows/Mozilla minidump format | |
73a99163 | 752 | - when truncating coredumps, also log the full size that the process had, and make a metadata field so we can report truncated coredumps |
87a8baa3 LP |
753 | |
754 | * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting) | |
755 | ||
70a44afe | 756 | * default to actual 32-bit PIDs, via /proc/sys/kernel/pid_max |
65c0cf71 | 757 | |
f957632b KS |
758 | * be able to specify a forced restart of service A where service B depends on, in case B |
759 | needs to be auto-respawned? | |
760 | ||
f38afcd0 | 761 | * tmpfiles: |
f38afcd0 | 762 | - apply "x" on "D" too (see patch from William Douglas) |
614cc34f ZJS |
763 | - replace F with f+. |
764 | - instead of ignoring unknown fields, reject them. | |
e25b5a8d DH |
765 | - creating new directories/subvolumes/fifos/device nodes |
766 | should not follow symlinks. None of the other adjustment or creation | |
767 | calls follow symlinks. | |
1258097c | 768 | |
a4c279f8 LP |
769 | * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early |
770 | ||
f28f1daf LP |
771 | * verify that the AF_UNIX sockets of a service in the fs still exist |
772 | when we start a service in order to avoid confusion when a user | |
773 | assumes starting a service is enough to make it accessible | |
774 | ||
88a07670 LP |
775 | * Make it possible to set the keymap independently from the font on |
776 | the kernel cmdline. Right now setting one resets also the other. | |
777 | ||
71092d70 | 778 | * and a dbus call to generate target from current state |
916abb21 | 779 | |
68c7d001 | 780 | * write blog stories about: |
796b06c2 | 781 | - hwdb: what belongs into it, lsusb |
68c7d001 LP |
782 | - enabling dbus services |
783 | - status update | |
7a2a0b90 | 784 | - how to make changes to sysctl and sysfs attributes |
253ee27a | 785 | - remote access |
5d0fcd7c | 786 | - how to pass throw-away units to systemd, or dynamically change properties of existing units |
8bbabc44 | 787 | - testing with Harald's awesome test kit |
f530371f | 788 | - auto-restart |
e707c494 | 789 | - how to develop against journal browsing APIs |
a19554ed | 790 | - the journal HTTP iface |
a73d88fa | 791 | - non-cgroup resource management |
a74a8793 | 792 | - dynamic resource management with cgroups |
3040728b | 793 | - refreshed, longer missions statement |
3679d112 LP |
794 | - calendar time events |
795 | - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell | |
796 | - how to create your own target | |
797 | - instantiated apache, dovecot and so on | |
798 | - hooking a script into various stages of shutdown/rearly booot | |
73090dc8 | 799 | |
21bdae12 LP |
800 | * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support. |
801 | ||
59fee421 LP |
802 | * dot output for --test showing the 'initial transaction' |
803 | ||
71092d70 | 804 | * fingerprint.target, wireless.target, gps.target, netdevice.target |
246756ca | 805 | |
e25b5a8d DH |
806 | * pid1: |
807 | - .timer units should optionally support CLOCK_BOOTTIME in addition to CLOCK_MONOTONIC | |
808 | - When logging about multiple units (stopping BoundTo units, conflicts, etc.), | |
809 | log both units as UNIT=, so that journalctl -u triggers on both. | |
810 | - generate better errors when people try to set transient properties | |
811 | that are not supported... | |
812 | http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html | |
813 | - maybe introduce WantsMountsFor=? Usecase: | |
814 | http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html | |
815 | - recreate systemd's D-Bus private socket file on SIGUSR2 | |
e25b5a8d DH |
816 | - move PAM code into its own binary |
817 | - when we automatically restart a service, ensure we restart its rdeps, too. | |
e25b5a8d DH |
818 | - hide PAM options in fragment parser when compile time disabled |
819 | - Support --test based on current system state | |
820 | - 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(). | |
821 | - after deserializing sockets in socket.c we should reapply sockopts and things | |
e25b5a8d DH |
822 | - drop PID 1 reloading, only do reexecing (difficult: Reload() |
823 | currently is properly synchronous, Reexec() is weird, because we | |
824 | cannot delay the response properly until we are back, so instead of | |
825 | being properly synchronous we just keep open the fd and close it | |
826 | when done. That means clients do not get a successful method reply, | |
827 | but much rather a disconnect on success. | |
828 | - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr | |
829 | - when a bus name of a service disappears from the bus make sure to queue further activation requests | |
830 | ||
831 | * unit files: | |
832 | - allow port=0 in .socket units | |
833 | - maybe introduce ExecRestartPre= | |
834 | - add ReloadSignal= for configuring a reload signal to use | |
835 | - implement Register= switch in .socket units to enable registration | |
836 | in Avahi, RPC and other socket registration services. | |
837 | - allow Type=simple with PIDFile= | |
838 | https://bugzilla.redhat.com/show_bug.cgi?id=723942 | |
839 | - allow writing multiple conditions in unit files on one line | |
840 | - load-fragment: when loading a unit file via a chain of symlinks | |
841 | verify that it is not masked via any of the names traversed. | |
842 | - introduce Type=pid-file | |
843 | - ExecOnFailure=/usr/bin/foo | |
844 | - introduce mix of BindTo and Requisite | |
845 | - add a concept of RemainAfterExit= to scope units | |
846 | - Set NoNewPrivileges= on all of our own services, where that makes sense | |
847 | - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely | |
848 | - consider adding RuntimeDirectoryUser= + RuntimeDirectoryGroup= | |
849 | ||
af6f0d42 TG |
850 | * udev-link-config: |
851 | - Make sure ID_PATH is always exported and complete for | |
852 | network devices where possible, so we can safely rely | |
853 | on Path= matching | |
8faf88e5 | 854 | - check MTUBytes parsing (expecting size_t but we are using unsigned) |
af6f0d42 | 855 | |
88e4d1d7 | 856 | * sd-rtnl: |
88e4d1d7 | 857 | - add support for more attribute types |
c589a0e6 | 858 | - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places |
88e4d1d7 | 859 | |
0a4b9a07 | 860 | * networkd: |
c74ecd71 TG |
861 | - add more keys to [Route] and [Address] sections |
862 | - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config) | |
0a4b9a07 | 863 | - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline) |
e8d842a0 | 864 | - add reduced [Link] support to .network files |
5c1d3fc9 | 865 | - add Scope= parsing option for [Network] |
798e174a | 866 | - properly handle routerless dhcp leases |
a8eaaee7 | 867 | - work with non-Ethernet devices |
c9fa3670 | 868 | - add support for more bond options |
e25b5a8d | 869 | - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from? |
e25b5a8d DH |
870 | - the DHCP lease data (such as NTP/DNS) is still made available when |
871 | a carrier is lost on a link. It should be removed instantly. | |
872 | - expose in the API the following bits: | |
873 | - option 15, domain name and/or option 119, search list | |
874 | - option 12, host name and/or option 81, fqdn | |
875 | - option 123, 144, geolocation | |
876 | - option 252, configure http proxy (PAC/wpad) | |
877 | - provide a way to define a per-network interface default metric value | |
878 | for all routes to it. possibly a second default for DHCP routes. | |
879 | - allow Name= to be specified repeatedly in the [Match] section. Maybe also | |
880 | support Name=foo*|bar*|baz ? | |
881 | - duplicate address check for static IPs (like ARPCHECK in network-scripts) | |
882 | - allow DUID/IAID to be customized, see issue #394. | |
e25b5a8d | 883 | - whenever uplink info changes, make DHCP server send out FORCERENEW |
155e8b9a | 884 | |
473dfd7b TG |
885 | * networkd-wait-online: |
886 | - make operstates to wait for configurable? | |
8f5675cc | 887 | |
ac976532 | 888 | * dhcp: |
424a8732 | 889 | - figure out how much we can increase Maximum Message Size |
99f9dd8d | 890 | - support RFC4702 (pass FQDN) |
ac976532 | 891 | |
37d8b536 PF |
892 | * dhcp6: |
893 | - add functions to set previously stored IPv6 addresses on startup and get | |
894 | them at shutdown; store them in client->ia_na | |
895 | - write more test cases | |
37d8b536 | 896 | - implement reconfigure support, see 5.3., 15.11. and 22.20. |
37d8b536 | 897 | - implement support for temporary adressess (IA_TA) |
37d8b536 PF |
898 | - implement dhcpv6 authentication |
899 | - investigate the usefulness of Confirm messages; i.e. are there any | |
900 | situations where the link changes without any loss in carrier detection | |
901 | or interface down | |
902 | - some servers don't do rapid commit without a filled in IA_NA, verify | |
903 | this behavior | |
4a77c53d | 904 | - RouteTable= ? |
37d8b536 | 905 | |
2f8cd170 | 906 | External: |
74fe1fe3 | 907 | |
cc9784c6 | 908 | * dbus: |
3377af3e | 909 | - natively watch for dbus-*.service symlinks (PENDING) |
b44be3ec | 910 | - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service |
cc9784c6 | 911 | |
71092d70 LP |
912 | * fix alsa mixer restore to not print error when no config is stored |
913 | ||
74fe1fe3 LP |
914 | * make cryptsetup lower --iter-time |
915 | ||
9096ad36 | 916 | * patch kernel for xattr support in /dev, /proc/, /sys? |
06ae4bfe | 917 | |
71092d70 | 918 | * kernel: add device_type = "fb", "fbcon" to class "graphics" |
14bd37fe | 919 | |
8973790e LP |
920 | * drop accountsservice's StandardOutput=syslog and Type=dbus fields |
921 | ||
8973790e LP |
922 | * dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id |
923 | ||
8973790e LP |
924 | * /usr/bin/service should actually show the new command line |
925 | ||
d2bbca68 | 926 | * 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= |
8973790e LP |
927 | |
928 | * fedora: F20: go timer units all the way, leave cron.daily for cron | |
929 | ||
930 | * neither pkexec nor sudo initialize environ[] from the PAM environment? | |
b1ce67bf | 931 | |
4641a16b | 932 | * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it |
19aa7c4f | 933 | |
baf167ee ZJS |
934 | * register catalog database signature as file magic |
935 | ||
1cf3c30c ZJS |
936 | * zsh shell completion: |
937 | - <command> <verb> -<TAB> should complete options, but currently does not | |
409886c4 | 938 | - systemctl add-wants,add-requires |
1cf3c30c | 939 | |
2c12a402 | 940 | |
129126f3 | 941 | Regularly: |
874aa2cc | 942 | |
129126f3 | 943 | * look for close() vs. close_nointr() vs. close_nointr_nofail() |
5021be21 | 944 | |
129126f3 | 945 | * check for strerror(r) instead of strerror(-r) |
5b6319dc LP |
946 | |
947 | * Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel | |
8c6db833 | 948 | |
444a79d3 LP |
949 | * pahole |
950 | ||
d28315e4 | 951 | * set_put(), hashmap_put() return values check. i.e. == 0 does not free()! |
a3a3e5b6 | 952 | |
4db17f29 | 953 | * use secure_getenv() instead of getenv() where appropriate |
88fae6e0 | 954 | |
d05c556b | 955 | * link up selected blog stories from man pages and unit files Documentation= fields |