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.
9 * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
11 * wiki: update journal format documentation for lz4 additions
15 * Rearrange tests so that the various test-xyz.c match a specific src/basic/xyz.c again
17 * rework mount.c and swap.c to follow proper state enumeration/deserialization
18 semantics, like we do for device.c now
22 * add --copy-from and --copy-to command to systemd-dissect which copies stuff
23 in and out of a disk image
25 * add systemd.random_seed= on the kernel cmdline, taking some hex or base64
26 encoded data. During earliest boot, credit it to entropy. This is not useful
27 for general purpose systems, but certainly for testing environments in VMs
28 and such, as it allows us to boot up instantly with fully initialized entropy
29 pool even if RNG pass-thru is not available.
31 * Support ProtectProc= or so, using: https://patchwork.kernel.org/cover/11310197/
33 * if /usr/bin/swapoff fails due to OOM, log a friendly explanatory message about it
35 * build short web pages out of each catalog entry, build them along with man
36 pages, and include hyperlinks to them in the journal output
38 * machined: add API to acquire UID range. add API to mount/dissect loopback
39 file. Both protected by PK. Then make nspawn use these APIs to run
40 unprivileged containers. i.e. push the truly privileged bits into machined,
41 so that the client side can remain entirely unprivileged, with SUID or
44 * add "throttling" to sd-event event sources: optionally, when we wake up too
45 often for one, let's turn it off entirely for a while. Use that for the
46 /proc/self/mountinfo logic.
48 * move our systemd-user PAM snippet to /usr/, which PAM appears to support
51 * nspawn: support time namespaces
53 * add ConditionSecurity=tpm2
55 * Remove any support for booting without /usr pre-mounted in the initrd entirely.
56 Update INITRD_INTERFACE.md accordingly.
58 * pid1: Move to tracking of main pid/control pid of units per pidfd
60 * pid1: support new clone3() fork-into-cgroup feature
62 * pid1: also remove PID files of a service when the service starts, not just
65 * make us use dynamically fewer deps for containers in general purpose distros:
66 o turn into dlopen() deps:
67 - pcre2 (always) — irrelevant on Fedora, since dep by
68 libselinux, but should benefit Debian
69 - libpwquality (always) - only relevant for homed, and maybe soon
72 - p11-kit-trust (always)
73 - kmod-libs (only when called from PID 1)
74 - cryptsetup-libs (only in RootImage= handling in PID 1, but not in systemd-cryptsetup)
76 - libpam (only when called from PID 1)
77 - bzip2, xz, lz4 (always — gzip and zstd should probably stay static deps the way they are,
78 since they are so basic and our defaults)
79 o move into separate libsystemd-shared-iptables.so .so
80 - iptables-libs (only used by nspawn + networkd)
82 * seccomp: when SystemCallArchitectures=native is set then don't install any
83 other seccomp filters for any of the other archs, in order to reduce the
84 number of seccomp filters we install needlessly.
86 * seccomp: maybe use seccomp_merge() to merge our filters per-arch if we can.
87 Apparently kernel performance is much better with fewer larger seccomp
88 filters than with more smaller seccomp filters.
90 * systemd-path: add ESP and XBOOTLDR path. Add "private" runtime/state/cache dir enum,
91 mapping to $RUNTIME_DIRECTORY, $STATE_DIRECTORY and such
93 * make "systemd-dissect" an official supported tool, i.e. move to /usr/bin/ and
94 provide man page. Given that we now have a tool that can generate images like
95 this, it's useful to have one that can dump contents of them, too.
97 * All tools that support --root= should also learn --image= so that they can
98 operate on disk images directly. Specifically: bootctl, firstboot, tmpfiles,
99 sysusers, systemctl, repart, journalctl, coredumpctl.
101 * seccomp: by default mask x32 ABI system wide on x86-64. it's on its way out
103 * seccomp: don't install filters for ABIs that are masked anyway for the
106 * seccomp: maybe merge all filters we install into one with that libseccomp API that allows merging.
108 * per-service credential system. Specifically: add LoadCredential= (for loading
109 cred from file), AcquireCredential= (for asking user for cred, via
110 ask-password), PassCredential= (for passing on credential systemd itself
111 got). Then, place credentials in a per-service, immutable ramfs instance (so
112 that it cannot be swapped out), destroy after use. Also pass via keyring
113 (with graceful fallback to cover for containers). Define CredentialPath= for
114 defining subdir of /run/credentials/ where to place it. Set $CREDENTIAL_PATH
115 env var for services to the result. Also pass via fd passing (optionally).
117 * homed: add native recovery key support. use 48 lowercase modhex characters
118 (192bit), show qr code of it, include pattern expression in user record.
120 * homed: introduce "degraded" state for home directories that weren't cleanly
121 unmounted (use xattr we add and remove on the loop back file)
123 * homed: during login resize fs automatically towards size goal. Specifically,
124 resize to diskSize if possible, but leave a certain amount (configured by a
125 new value diskLeaveFreeSize) of space free on the backing fs.
127 * homed: permit multiple user record signing keys to be used locally, and pick
128 the right one for signing records automatically depending on a pre-existing
131 * homed: add a way to "adopt" a home directory, i.e. strip foreign signatures
132 and insert a local signature instead.
134 * homed: as an extension to the directory+subvolume backend: if located on
135 especially marked fs, then sync down password into LUKS header of that fs,
136 and always verify passwords against it too. Bootstrapping is a problem
137 though: if no one is logged in (or no other user even exists yet), how do you
138 unlock the volume in order to create the first user and add the first pw.
140 * homed: support new FS_IOC_ADD_ENCRYPTION_KEY ioctl for setting up fscrypt
142 * busctl: maybe expose a verb "ping" for pinging a dbus service to see if it
145 * homed: maybe pre-create ~/.cache as subvol so that it can have separate quota
148 * when systemd-nspawn and suchlike dissect an OS image, and there are multiple
149 root partitions, do an strverscmp() on the partition label and boot
150 first. That is inspired how sd-boot figures out which kernel to boot, and
151 thus allows defining OS images which can be A/B updated and we default to the
152 newest version automatically, both in nspawn and in sd-boot
154 * cryptsetup/homed: also support FIDO2 HMAC password logic for unlocking
155 devices. (see: https://github.com/mjec/fido2-hmac-secret)
157 * systemd-gpt-auto should probably set x-systemd.growfs on the mounts it
160 * homed/userdb: distinguish passwords and recovery keys in the records, since
161 we probably want to use different PBKDF algorithms/settings for them:
162 passwords have low entropy but recovery keys should have good entropy key
163 hence we can make them quicker to work.
166 - teach it to prepare an ESP wholesale, i.e. with mkfs.vfat invocation
167 - teach it to copy in unified kernel images and maybe type #1 boot loader spec entries from host
168 - make it operate on loopback files, dissecting enough to find ESP to operate on
170 * by default, in systemd --user service bump the OOMAdjust to 100, as privs
171 allow so that systemd survives
173 * cryptsetup: allow encoding key directly in /etc/crypttab, maybe with a
174 "base64:" prefix. Useful in particular for pkcs11 mode.
176 * cryptsetup: reimplement the mkswap/mke2fs in cryptsetup-generator to use
177 systemd-makefs.service instead.
179 * socket units: allow creating a udev monitor socket with ListenDevices= or so,
180 with matches, then activate app through that passing socket over
182 * unify on openssl (as soon as OpenSSL 3.0 is out, and the Debian license
184 - port sd_id128_get_machine_app_specific() over from khash
185 - port resolved over from libgcrypt (DNSSEC code)
186 - port journald + fsprg over from libgcrypt
187 - port importd over from libgcrypt
188 - when that's done: kill khash.c
189 - when that's done: kill gnutls support in resolved
191 * kill zenata, all hail weblate?
193 * when we resize disks (homed?) always round up to 4K sectors, not 512K
195 * add growvol and makevol options for /etc/crypttab, similar to
196 x-systemd.growfs and x-systemd-makefs.
198 * hook up the TPM to /etc/crypttab, with a new option that is similar to the
199 new PKCS#11 option in crypttab, and allows unlocking a LUKS volume via a key
200 unsealed from the TPM. Optionally, if TPM is not available fall back to
201 TPM-less mode, and set up linear DM mapping instead (inspired by kpartx), so
202 that the device paths stay the same, regardless if crypto is used or not.
204 * systemd-repart: by default generate minimized partition tables (i.e. tables
205 that only cover the space actually used, excluding any free space at the
206 end), in order to maximize dd'ability. Requires libfdisk work, see
207 https://github.com/karelzak/util-linux/issues/907
209 * systemd-repart: optionally, allow specifying a path to initialize new
210 partitions from, i.e. an fs image file or a source device node. This would
211 then turn systemd-repart into a simple installer: with a few .repart files
212 you could replicate the host system on another device. a full installer would
213 then be: "systemd-repart /dev/sda && bootctl install /dev/sda &&
214 systemd-firstboot --image= …"
216 * systemd-repart: MBR partition table support. Care needs to be taken regarding
217 Type=, so that partition definitions can sanely apply to both the GPT and the
218 MBR case. Idea: accept syntax "Type=gpt:home mbr:0x83" for setting the types
219 for the two partition types explicitly. And provide an internal mapping so
220 that "Type=linux-generic" maps to the right types for both partition tables
223 * systemd-repart: allow sizing partitions as factor of available RAM, so that
224 we can reasonably size swap partitions for hibernation.
226 * systemd-repart: allow running mkfs before making partitions pop up +
227 encryption via LUKS to allow booting into an empty root with only /usr mounted in
229 * systemd-repart: allow managing the gpt read-only partition flag + auto-mount flag
231 * systemd-repart: allow disabling growing of specific partitions, or making
232 them (think ESP: we don't ever want to grow it, since we cannot resize vfat)
234 * systemd-repart: add specifier expansion, add especifier that refers to root
235 device node of current system, /usr device node, and matching verity, so that
236 an installer can be made a "copy" installer of the booted OS
238 * systemd-repart: make it a static checker during early boot for existence and
239 absence of other partitions for trusted boot environments
241 * systemd-repart: allow config of partition uuid
243 * userdb: allow username prefix searches in varlink API, allow realname and
244 realname substr searches in varlink API
246 * userdb: allow uid/gid range checks
248 * userdb: allow existence checks
250 * pid1: activation by journal search expression
252 * when switching root from initrd to host, set the machine_id env var so that
253 if the host has no machine ID set yet we continue to use the random one the
256 * sd-event: add native support for P_ALL waitid() watching, then move PID 1 to
257 it fo reaping assigned but unknown children. This needs to some special care
258 to operate somewhat sensibly in light of priorities: P_ALL will return
259 arbitrary processes, regardless of the priority we want to watch them with,
260 hence on each event loop iteration check all processes which we shall watch
261 with higher prio explicitly, and then watch the entire rest with P_ALL.
263 * tweak sd-event's child watching: keep a prioq of children to watch and use
264 waitid() only on the children with the highest priority until one is waitable
265 and ignore all lower-prio ones from that point on
267 * maybe introduce xattrs that can be set on the root dir of the root fs
268 partition that declare the volatility mode to use the image in. Previously I
269 thought marking this via GPT partition flags but that's not ideal since
270 that's outside of the LUKS encryption/verity verification, and we probably
271 shouldn't operate in a volatile mode unless we got told so from a trusted
274 * figure out automatic partition discovery when combining writable root dir
277 * coredump: maybe when coredumping read a new xattr from /proc/$PID/exe that
278 may be used to mark a whole binary as non-coredumpable. Would fix:
279 https://bugs.freedesktop.org/show_bug.cgi?id=69447
281 * teach parse_timestamp() timezones like the calendar spec already knows it
283 * beef up hibernation to optionally do swapon/swapoff immediately before/after
286 * beef up s2h to implement a battery watch loop: instead of entering
287 hibernation unconditionally after coming back from resume make a decision
288 based on the battery load level: if battery level is above a specific
289 threshold, go to suspend again, only hibernate if below it. This means we'd
290 stick to suspend usually, but fall back to hibernation only when battery runs
291 empty (well, subject to our sampling interval). Related to this, check if we
292 can make ACPI _BTP (i.e. /sys/class/power_supply/*/alarm) work for us too,
293 i.e. see if it can wake up machines from suspend, so that we could resume
294 automatically when the system is low on power and move automatically to
295 hibernation mode. (see
296 https://uefi.org/sites/default/files/resources/ACPI%206_2_A_Sept29.pdf
298 https://docs.microsoft.com/en-us/windows-hardware/design/device-experiences/modern-standby-wake-sources
301 * add an explicit "vertical" mode to format-table, so that "systemctl
302 status"-like outputs (i.e. with a series of field names left and values
303 right) become genuine first class citizens, and we gain automatic, sane JSON
306 * systemd-firstboot: teach it dissector magic, so that you can point it to some
307 disk image and it will just set everything in it all behind the scenes.
309 * We should probably replace /var/log/README, /etc/rc.d/README with symlinks
310 that are linked to these places instead of copied. After all they are
311 constant vendor data.
313 * maybe add kernel cmdline params: to force random seed crediting
315 * nspawn: on cgroupsv1 issue cgroup empty handler process based on host events,
316 so that we make cgroup agent logic safe
318 * nspawn/machined: add API to invoke binary in container, then use that as
319 fallback in "machinectl shell"
321 * logind: rework pam_logind to also do a bus call in case of invocation from
322 user@.service, which returns the XDG_RUNTIME_DIR value, and make this
323 behaviour selectable via pam module option.
326 - when user tries to log into record signed by unrecognized key, automatically add key to our chain after polkit auth
327 - hook up machined/nspawn users with a varlink user query interface
328 - rollback when resize fails mid-operation
329 - GNOME's side for forget key on suspend (requires rework so that lock screen runs outside of uid)
331 - shrink fs on logout?
332 - update LUKS password on login if we find there's a password that unlocks the JSON record but not the LUKS device.
333 - create on activate?
334 - properties: icon url?, preferred session type?, administrator bool (which translates to 'wheel' membership)?, address?, telephone?, vcard?, samba stuff?, parental controls?
335 - communicate clearly when usb stick is safe to remove. probably involves
336 beefing up logind to make pam session close hook synchronous and wait until
337 systemd --user is shut down.
338 - logind: maybe keep a "busy fd" as long as there's a non-released session around or the user@.service
339 - maybe make automatic, read-only, time-based reflink-copies of LUKS disk images (think: time machine)
340 - distinguish destroy / remove (i.e. currently we can unregister a user, unregister+remove their home directory, but not just remove their home directory)
341 - in systemd's PAMName= logic: query passwords with ssh-askpassword, so that we can make "loginctl set-linger" mode work
342 - fingerprint authentication, pattern authentication, …
343 - make sure "classic" user records can also be managed by homed
344 - description field for groups
345 - make size of $XDG_RUNTIME_DIR configurable in user record
346 - reuse pwquality magic in firstboot
347 - query password from kernel keyring first
348 - update even if record is "absent"
349 - add a "access mode" + "fstype" field to the "status" section of json identity records reflecting the actually used access mode and fstype, even on non-luks backends
350 - move acct mgmt stuff from pam_systemd_home to pam_systemd?
351 - when "homectl --pkcs11-token-uri=" is used, synthesize ssh-authorized-keys records for all keys we have private keys on the stick for
352 - make slice for users configurable (requires logind rework)
353 - logind: populate auto-login list bus property from PKCS#11 token
354 - when determining state of a LUKS home directory, check DM suspended sysfs file
356 * introduce a new per-process uuid, similar to the boot id, the machine id, the
357 invocation id, that is derived from process creds, specifically a hashed
358 combination of AT_RANDOM + getpid() + the starttime from
359 /proc/self/status. Then add these ids implicitly when logging. Deriving this
360 uuid from these three things has the benefit that it can be derived easily
361 from /proc/$PID/ in a stable, and unique way that changes on both fork() and
364 * let's not GC a unit while its ratelimits are still pending
366 * when killing due to service watchdog timeout maybe detect whether target
367 process is under ptracing and then log loudly and continue instead.
369 * introduce a new group to own TPM devices
371 * make rfkill uaccess controllable by default, i.e. steal rule from
372 gnome-bluetooth and friends
374 * tweak journald context caching. In addition to caching per-process attributes
375 keyed by PID, cache per-cgroup attributes (i.e. the various xattrs we read)
376 keyed by cgroup path, and guarded by ctime changes. This should provide us
377 with a nice speed-up on services that have many processes running in the same
380 * make MAINPID= message reception checks even stricter: if service uses User=,
381 then check sending UID and ignore message if it doesn't match the user or
384 * maybe trigger a uevent "change" on a device if "systemctl reload xyz.device"
387 * when importing an fs tree with machined, optionally apply userns-rec-chown
389 * when importing an fs tree with machined, complain if image is not an OS
391 * Maybe introduce a helper safe_exec() or so, which is to execve() which
392 safe_fork() is to fork(). And then make revert the RLIMIT_NOFILE soft limit
393 to 1K implicitly, unless explicitly opted-out.
395 * rework seccomp/nnp logic that even if User= is used in combination with
396 a seccomp option we don't have to set NNP. For that, change uid first whil
397 keeping CAP_SYS_ADMIN, then apply seccomp, the drop cap.
399 * add a concept for automatically loading per-unit secrets off disk and
400 inserting them into the kernel keyring. Maybe SecretsDirectory= similar to
401 ConfigurationDirectory=.
403 * when no locale is configured, default to UEFI's PlatformLang variable
405 * bootctl,sd-boot: actually honour the "architecture" key
407 * add a new syscall group "@esoteric" for more esoteric stuff such as bpf() and
408 usefaultd() and make systemd-analyze check for it.
410 * paranoia: whenever we process passwords, call mlock() on the memory
411 first. i.e. look for all places we use free_and_erasep() and
412 augment them with mlock(). Also use MADV_DONTDUMP.
414 * Move RestrictAddressFamily= to the new cgroup create socket
416 * support the bind/connect/sendmsg cgroup stuff for sandboxing, and possibly
419 * maybe implicitly attach monotonic+realtime timestamps to outgoing messages in
420 log.c and sd-journal-send
422 * optionally: turn on cgroup delegation for per-session scope units
424 * introduce per-unit (i.e. per-slice, per-service) journal log size limits.
426 * sd-boot: automatically load EFI modules from some drop-in dir, so that people
427 can add in file system drivers and such
429 * sd-boot: optionally, show boot menu when previous default boot item has
430 non-zero "tries done" count
432 * introduce an option (or replacement) for "systemctl show" that outputs all
433 properties as JSON, similar to busctl's new JSON output. In contrast to that
434 it should skip the variant type string though.
436 * augment CODE_FILE=, CODE_LINE= with something like CODE_BASE= or so which
437 contains some identifier for the project, which allows us to include
438 clickable links to source files generating these log messages. The identifier
439 could be some abberviated URL prefix or so (taking inspiration from Go
440 imports). For example, for systemd we could use
441 CODE_BASE=github.com/systemd/systemd/blob/98b0b1123cc or so which is
442 sufficient to build a link by prefixing "http://" and suffixing the
445 * Augment MESSAGE_ID with MESSAGE_BASE, in a similar fashion so that we can
446 make clickable links from log messages carrying a MESSAGE_ID, that lead to
447 some explanatory text online.
449 * maybe extend .path units to expose fanotify() per-mount change events
451 * Add a "systemctl list-units --by-slice" mode or so, which rearranges the
452 output of "systemctl list-units" slightly by showing the tree structure of
453 the slices, and the units attached to them.
455 * the a-posteriori stopping of units bound to units that disappeared logic
456 should be reworked: there should be a queue of units, and we should only
457 enqueue stop jobs from a defer event that processes queue instead of
458 right-away when we find a unit that is bound to one that doesn't exist
459 anymore. (similar to how the stop-unneeded queue has been reworked the same
462 * nspawn: make nspawn suitable for shell pipelines: instead of triggering a
463 hangup when input is finished, send ^D, which synthesizes an EOF. Then wait
464 for hangup or ^D before passing on the EOF.
466 * When reloading configuration PID 1 should reset all its properties to the
467 original defaults before calling parse_config()
469 * nspawn: greater control over selinux label?
471 * hibernate/s2h: make this robust and safe to enable in Fedora by default.
474 1. add resume_offset support to the resume code (i.e. support swap files
476 2. check if swap is on weird storage and refuse if so
477 3. add auto-detection of hibernation images
479 * cgroups: use inotify to get notified when somebody else modifies cgroups
480 owned by us, then log a friendly warning.
482 * beef up log.c with support for stripping ANSI sequences from strings, so that
483 it is OK to include them in log strings. This would be particularly useful so
484 that our log messages could contain clickable links for example for unit
485 files and suchlike we operate on.
487 * add support for "portablectl attach http://foobar.com/waaa.raw (i.e. importd integration)
489 * add attach --enable and attach --now (for attach+enable+start)
491 * sync dynamic uids/gids between host+portable srvice (i.e. if DynamicUser=1 is set for a service, make sure that the
492 selected user is resolvable in the service even if it ships its own /etc/passwd)
494 * Fix DECIMAL_STR_MAX or DECIMAL_STR_WIDTH. One includes a trailing NUL, the
495 other doesn't. What a disaster. Probably to exclude it. Also
496 DECIMAL_STR_WIDTH should probably add an extra "-" into account for negative
499 * Check that users of inotify's IN_DELETE_SELF flag are using it properly, as
500 usually IN_ATTRIB is the right way to watch deleted files, as the former only
501 fires when a file is actually removed from disk, i.e. the link count drops to
502 zero and is not open anymore, while the latter happens when a file is
503 unlinked from any dir.
505 * port systemctl, busctl, … over to format-table.[ch]'s table formatters
507 * pid1: lock image configured with RootDirectory=/RootImage= using the usual nspawn semantics while the unit is up
509 * add --vacuum-xyz options to coredumpctl, matching those journalctl already has.
511 * introduce Ephemeral= unit file switch, that creates an ephemeral copy of all
512 files and directories that are left writable for a unit, and which are
513 removed after the unit goes down again. A bit like --ephemeral for
514 systemd-nspawn but for system services. If used together with RootImage= this
515 should reflink the image file itself.
517 Related: add Ephemeral=<path1> <path2> … which would allow marking
518 specific paths only like this.
520 * add CopyFile= or so as unit file setting that may be used to copy files or
521 directory trees from the host to the services RootImage= and RootDirectory=
522 environment. Which we can use for /etc/machine-id and in particular
523 /etc/resolv.conf. Should be smart and do something useful on read-only
524 images, for example fallback to read-only bind mounting the file instead.
526 * show invocation ID in systemd-run output
528 * bypass SIGTERM state in unit files if KillSignal is SIGKILL
530 * add proper dbus APIs for the various sd_notify() commands, such as MAINPID=1
531 and so on, which would mean we could report errors and such.
533 * teach tmpfiles.d q/Q logic something sensible in the context of XFS/ext4
536 * introduce DefaultSlice= or so in system.conf that allows changing where we
537 place our units by default, i.e. change system.slice to something
538 else. Similar, ManagerSlice= should exist so that PID1's own scope unit could
539 be moved somewhere else too. Finally machined and logind should get similar
540 options so that it is possible to move user session scopes and machines to a
541 different slice too by default. Usecase: people who want to put resources on
542 the entire system, with the exception of one specific service. See:
543 https://lists.freedesktop.org/archives/systemd-devel/2018-February/040369.html
545 * maybe rework get_user_creds() to query the user database if $SHELL is used
546 for root, but only then.
548 * be stricter with fds we receive for the fdstore: close them asynchronously
550 * calenderspec: add support for week numbers and day numbers within a
551 year. This would allow us to define "bi-weekly" triggers safely.
553 * sd-bus: add vtable flag, that may be used to request client creds implicitly
554 and asynchronously before dispatching the operation
556 * make use of ethtool veth peer info in machined, for automatically finding out
557 host-side interface pointing to the container.
559 * add some special mode to LogsDirectory=/StateDirectory=… that allows
560 declaring these directories without necessarily pulling in deps for them, or
561 creating them when starting up. That way, we could declare that
562 systemd-journald writes to /var/log/journal, which could be useful when we
563 doing disk usage calculations and so on.
565 * taint systemd if there are fewer than 65536 users assigned (userns) to the system.
567 * deprecate RootDirectoryStartOnly= in favour of a new ExecStart= prefix char
569 * add a new RuntimeDirectoryPreserve= mode that defines a similar lifecycle for
570 the runtime dir as we maintain for the fdstore: i.e. keep it around as long
571 as the unit is running or has a job queued.
573 * support projid-based quota in machinectl for containers
575 * add a way to lock down cgroup migration: a boolean, which when set for a unit
576 makes sure the processes in it can never migrate out of it
578 * blog about fd store and restartable services
580 * document Environment=SYSTEMD_LOG_LEVEL=debug drop-in in debugging document
582 * rework ExecOutput and ExecInput enums so that EXEC_OUTPUT_NULL loses its
583 magic meaning and is no longer upgraded to something else if set explicitly.
585 * in the long run: permit a system with /etc/machine-id linked to /dev/null, to
586 make it lose its identity, i.e. be anonymous. For this we'd have to patch
587 through the whole tree to make all code deal with the case where no machine
590 * optionally, collect cgroup resource data, and store it in per-unit RRD files,
591 suitable for processing with rrdtool. Add bus API to access this data, and
592 possibly implement a CPULoad property based on it.
594 * beef up pam_systemd to take unit file settings such as cgroups properties as
597 * maybe hook of xfs/ext4 quotactl() with services? i.e. automatically manage
598 the quota of a the user indicated in User= via unit file settings, like the
599 other resource management concepts. Would mix nicely with DynamicUser=1. Or
600 alternatively, do this with projids, so that we can also cover services
601 running as root. Quota should probably cover all the special dirs such as
602 StateDirectory=, LogsDirectory=, CacheDirectory=, as well as RootDirectory= if it
603 is set, plus the whole disk space any image configured with RootImage=.
605 * In DynamicUser= mode: before selecting a UID, use disk quota APIs on relevant
606 disks to see if the UID is already in use.
608 * add "systemctl wait" or so, which does what "systemd-run --wait" does, but
609 for all units. It should be both a way to pin units into memory as well as a
610 wait to retrieve their exit data.
612 * expose IO accounting data on the bus, show it in systemd-run --wait and log
613 about it in the resource log message
615 * show whether a service has out-of-date configuration in "systemctl status" by
616 using mtime data of ConfigurationDirectory=.
618 * Add AddUser= setting to unit files, similar to DynamicUser=1 which however
619 creates a static, persistent user rather than a dynamic, transient user. We
620 can leverage code from sysusers.d for this.
622 * add some optional flag to ReadWritePaths= and friends, that has the effect
623 that we create the dir in question when the service is started. Example:
625 ReadWritePaths=:/var/lib/foobar
627 * maybe add call sd_journal_set_block_timeout() or so to set SO_SNDTIMEO for
628 the sd-journal logging socket, and, if the timeout is set to 0, sets
629 O_NONBLOCK on it. That way people can control if and when to block for
632 * hostnamed: populate form factor data from a new hwdb database, so that old
633 yogas can be recognized as "convertible" too, even if they predate the DMI
634 "convertible" form factor
636 * Maybe add PrivatePIDs= as new unit setting, and do minimal PID namespacing
637 after all. Be strict however, only support the equivalent of nspawn's
638 --as-pid2 switch, and sanely proxy sd_notify() messages dropping stuff such
641 * Add ExecMonitor= setting. May be used multiple times. Forks off a process in
642 the service cgroup, which is supposed to monitor the service, and when it
643 exits the service is considered failed by its monitor.
645 * track the per-service PAM process properly (i.e. as an additional control
646 process), so that it may be queried on the bus and everything.
648 * add a new "debug" job mode, that is propagated to unit_start() and for
649 services results in two things: we raise SIGSTOP right before invoking
650 execve() and turn off watchdog support. Then, use that to implement
651 "systemd-gdb" for attaching to the start-up of any system service in its
654 * gpt-auto logic: related to the above, maybe support a "secondary" root
655 partition, that is mounted to / and is writable, and where the actual root's
656 /usr is mounted into.
658 * 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
660 * drop nss-myhostname in favour of nss-resolve?
662 * add a percentage syntax for TimeoutStopSec=, e.g. TimeoutStopSec=150%, and
663 then use that for the setting used in user@.service. It should be understood
664 relative to the configured default value.
666 * in networkd, when matching device types, fix up DEVTYPE rubbish the kernel passes to us
668 * enable LockMLOCK to take a percentage value relative to physical memory
670 * Permit masking specific netlink APIs with RestrictAddressFamily=
672 * nspawn: support that /proc, /sys/, /dev are pre-mounted
674 * define gpt header bits to select volatility mode
676 * ProtectClock= (drops CAP_SYS_TIMES, adds seecomp filters for settimeofday, adjtimex), sets DeviceAllow o /dev/rtc
678 * ProtectTracing= (drops CAP_SYS_PTRACE, blocks ptrace syscall, makes /sys/kernel/tracing go away)
680 * ProtectMount= (drop mount/umount/pivot_root from seccomp, disallow fuse via DeviceAllow, imply Mountflags=slave)
682 * ProtectKeyRing= to take keyring calls away
684 * RemoveKeyRing= to remove all keyring entries of the specified user
686 * ProtectReboot= that masks reboot() and kexec_load() syscalls, prohibits kill
687 on PID 1 with the relevant signals, and makes relevant files in /sys and
688 /proc (such as the sysrq stuff) unavailable
690 * make sure the ratelimit object can deal with USEC_INFINITY as way to turn off things
692 * journalctl: make sure -f ends when the container indicated by -M terminates
694 * mount: automatically search for "main" partition of an image has multiple
697 * in nss-systemd, if we run inside of RootDirectory= with PrivateUsers= set,
698 find a way to map the User=/Group= of the service to the right name. This way
699 a user/group for a service only has to exist on the host for the right
702 * add bus API for creating unit files in /etc, reusing the code for transient units
704 * add bus API to remove unit files from /etc
706 * add bus API to retrieve current unit file contents (i.e. implement "systemctl cat" on the bus only)
708 * rework fopen_temporary() to make use of open_tmpfile_linkable() (problem: the
709 kernel doesn't support linkat() that replaces existing files, currently)
711 * transient units: don't bother with actually setting unit properties, we
712 reload the unit file anyway
714 * journald: sigbus API via a signal-handler safe function that people may call
715 from the SIGBUS handler
717 * optionally, also require WATCHDOG=1 notifications during service start-up and shutdown
719 * delay activation of logind until somebody logs in, or when /dev/tty0 pulls it
720 in or lingering is on (so that containers don't bother with it until PAM is used). also exit-on-idle
722 * cache sd_event_now() result from before the first iteration...
724 * PID1: find a way how we can reload unit file configuration for
725 specific units only, without reloading the whole of systemd
727 * add an explicit parser for LimitRTPRIO= that verifies
728 the specified range and generates sane error messages for incorrect
731 * when we detect that there are waiting jobs but no running jobs, do something
733 * push CPUAffinity= also into the "cpuset" cgroup controller
735 * PID 1 should send out sd_notify("WATCHDOG=1") messages (for usage in the --user mode, and when run via nspawn)
737 * there's probably something wrong with having user mounts below /sys,
738 as we have for debugfs. for example, src/core/mount.c handles mounts
739 prefixed with /sys generally special.
740 http://lists.freedesktop.org/archives/systemd-devel/2015-June/032962.html
742 * fstab-generator: default to tmpfs-as-root if only usr= is specified on the kernel cmdline
744 * docs: bring http://www.freedesktop.org/wiki/Software/systemd/MyServiceCantGetRealtime up to date
746 * add a job mode that will fail if a transaction would mean stopping
747 running units. Use this in timedated to manage the NTP service
749 http://lists.freedesktop.org/archives/systemd-devel/2015-April/030229.html
751 * The udev blkid built-in should expose a property that reflects
752 whether media was sensed in USB CF/SD card readers. This should then
753 be used to control SYSTEMD_READY=1/0 so that USB card readers aren't
754 picked up by systemd unless they contain a medium. This would mirror
755 the behaviour we already have for CD drives.
757 * networkd/udev: implement SR_IOV configuration in .link files:
758 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html
760 * hostnamectl: show root image uuid
762 * Find a solution for SMACK capabilities stuff:
763 http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html
765 * "systemctl preset-all" should probably order the unit files it
766 operates on lexicographically before starting to work, in order to
767 ensure deterministic behaviour if two unit files conflict (like DMs
770 * synchronize console access with BSD locks:
771 http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html
773 * as soon as we have sender timestamps, revisit coalescing multiple parallel daemon reloads:
774 http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
776 * figure out when we can use the coarse timers
778 * add "systemctl start -v foobar.service" that shows logs of a service
779 while the start command runs. This is non-trivial to do without
780 races though, since we should flush out all journal messages before
781 returning from the "systemctl stop".
783 * 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
785 * maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment=
787 * maybe allow timer units with an empty Units= setting, so that they
788 can be used for resuming the system but nothing else.
790 * what to do about udev db binary stability for apps? (raw access is not an option)
792 * systemctl: if some operation fails, show log output?
794 * systemctl edit: use equivalent of cat() to insert existing config as a comment, prepended with #.
795 Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc.
797 * exponential backoff in timesyncd when we cannot reach a server
799 * timesyncd: add ugly bus calls to set NTP servers per-interface, for usage by NM
801 * merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share....
803 * add systemd.abort_on_kill or some other such flag to send SIGABRT instead of SIGKILL
804 (throughout the codebase, not only PID1)
808 - service registration
809 - service/domain/types browsing
811 - DNS-SD service registration from socket units
812 - resolved should optionally register additional per-interface LLMNR
813 names, so that for the container case we can establish the same name
814 (maybe "host") for referencing the server, everywhere.
815 - allow clients to request DNSSEC for a single lookup even if DNSSEC is off (?)
816 - hook up resolved with machined-based address resolution
818 * refcounting in sd-resolve is borked
820 * Add a new verb "systemctl top"
822 * add new gpt type for btrfs volumes
824 * support empty /etc boots nicely:
825 - nspawn/gpt-generator: introduce new gpt partition type for /usr
827 * generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
829 * a way for container managers to turn off getty starting via $container_headless= or so...
831 * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
833 * For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
834 they run added to the initial transaction and thus confuse Type=idle.
836 * add bus api to query unit file's X fields.
838 * gpt-auto-generator:
839 - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap?
840 - Make /home automount rather than mount?
842 * add generator that pulls in systemd-network from containers when
843 CAP_NET_ADMIN is set, more than the loopback device is defined, even
844 when it is otherwise off
846 * MessageQueueMessageSize= (and suchlike) should use parse_iec_size().
848 * implement Distribute= in socket units to allow running multiple
849 service instances processing the listening socket, and open this up
852 * introduce bus call FreezeUnit(s, b), as well as "systemctl freeze
853 $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls
854 should SIGSTOP all unit processes in a loop until all processes of
855 it are fully stopped. This can later be used for app management by
856 desktop UIs such as gnome-shell to freeze apps that are not visible
857 on screen, not unlike how job control works on the shell
860 - implement per-slice CPUFairScheduling=1 switch
861 - introduce high-level settings for RT budget, swappiness
862 - how to reset dynamically changed unit cgroup attributes sanely?
863 - when reloading configuration, apply new cgroup configuration
864 - when recursively showing the cgroup hierarchy, optionally also show
865 the hierarchies of child processes
868 - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
870 * when we detect low battery and no AC on boot, show pretty splash and refuse boot
872 * libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
874 * be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
876 * 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
878 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
880 * If we try to find a unit via a dangling symlink, generate a clean
881 error. Currently, we just ignore it and read the unit from the search
884 * refuse boot if /usr/lib/os-release is missing or /etc/machine-id cannot be set up
886 * man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
888 * load .d/*.conf dropins for device units
892 - GetAllProperties() on a non-existing object does not result in a failure currently
893 - port to sd-resolve for connecting to TCP dbus servers
894 - 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
895 - see if we can drop more message validation on the sending side
896 - add API to clone sd_bus_message objects
897 - longer term: priority inheritance
899 - NameLost/NameAcquired obsolete
902 - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
905 - allow multiple signal handlers per signal?
906 - document chaining of signal handler for SIGCHLD and child handlers
907 - define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ...
909 * investigate endianness issues of UUID vs. GUID
911 * dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
912 should be able to safely try another attempt when the bus call LoadUnit() is invoked.
914 * add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
916 * add a pam module that on password changes updates any LUKS slot where the password matches
919 - add unit tests for config_parse_device_allow()
921 * seems that when we follow symlinks to units we prefer the symlink
922 destination path over /etc and /usr. We should not do that. Instead
923 /etc should always override /run+/usr and also any symlink
926 * when isolating, try to figure out a way how we implicitly can order
927 all units we stop before the isolating unit...
929 * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
931 * Add ConditionDirectoryNotEmpty= handle non-absoute paths as a search path or add
932 ConditionConfigSearchPathNotEmpty= or different syntax? See the discussion starting at
933 https://github.com/systemd/systemd/pull/15109#issuecomment-607740136.
935 * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify
936 that the boot loader should be installed to the ESP. Define a way
937 how an installer can figure out whether a BLS compliant boot loader
940 * think about requeuing jobs when daemon-reload is issued? usecase:
941 the initrd issues a reload after fstab from the host is accessible
942 and we might want to requeue the mounts local-fs acquired through
945 * systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
947 * remove any syslog support from log.c — we probably cannot do this before split-off udev is gone for good
949 * shutdown logging: store to EFI var, and store to USB stick?
951 * merge unit_kill_common() and unit_kill_context()
954 - honor language efi variables for default language selection (if there are any?)
955 - honor timezone efi variables for default timezone selection (if there are any?)
956 - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
958 - recognize the case when not booted on EFI
960 * maybe do not install getty@tty1.service symlink in /etc but in /usr?
962 * print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
964 * mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
967 - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
968 - logind: wakelock/opportunistic suspend support
969 - Add pretty name for seats in logind
970 - logind: allow showing logout dialog from system?
971 - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
972 - if pam_systemd is invoked by su from a process that is outside of a
973 any session we should probably just become a NOP, since that's
974 usually not a real user session but just some system code that just
976 - logind: make the Suspend()/Hibernate() bus calls wait for the for
977 the job to be completed. before returning, so that clients can wait
978 for "systemctl suspend" to finish to know when the suspending is
980 - logind: when the power button is pressed short, just popup a
981 logout dialog. If it is pressed for 1s, do the usual
982 shutdown. Inspiration are Macs here.
983 - expose "Locked" property on logind sesison objects
984 - maybe allow configuration of the StopTimeout for session scopes
985 - rename session scope so that it includes the UID. THat way
986 the session scope can be arranged freely in slices and we don't have
987 make assumptions about their slice anymore.
988 - follow PropertiesChanged state more closely, to deal with quick logouts and
990 - (optionally?) spawn seat-manager@$SEAT.service whenever a seat shows up that as CanGraphical set
993 - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields
994 - journald: also get thread ID from client, plus thread name
995 - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
996 - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
997 - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively?
998 - declare the local journal protocol stable in the wiki interface chart
999 - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
1000 - journald: when dropping msgs due to ratelimit make sure to write
1001 "dropped %u messages" not only when we are about to print the next
1002 message that works, but already after a short timeout
1003 - check if we can make journalctl by default use --follow mode inside of less if called without args?
1004 - maybe add API to send pairs of iovecs via sd_journal_send
1005 - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
1006 - journactl: support negative filtering, i.e. FOOBAR!="waldo",
1007 and !FOOBAR for events without FOOBAR.
1008 - journal: store timestamp of journal_file_set_offline() int he header,
1009 so it is possible to display when the file was last synced.
1010 - 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.
1011 - journal: find a way to allow dropping history early, based on priority, other rules
1012 - journal: When used on NFS, check payload hashes
1013 - journald: add kernel cmdline option to disable ratelimiting for debug purposes
1014 - refuse taking lower-case variable names in sd_journal_send() and friends.
1015 - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
1016 - journal: deal nicely with byte-by-byte copied files, especially regards header
1017 - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
1018 - Replace utmp, wtmp, btmp, and lastlog completely with journal
1019 - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
1020 - when a kernel driver logs in a tight loop, we should ratelimit that too.
1021 - journald: optionally, log debug messages to /run but everything else to /var
1022 - journald: when we drop syslog messages because the syslog socket is
1023 full, make sure to write how many messages are lost as first thing
1024 to syslog when it works again.
1025 - journald: allow per-priority and per-service retention times when rotating/vacuuming
1026 - journald: make use of uid-range.h to managed uid ranges to split
1028 - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
1029 - improve journalctl performance by loading journal files
1030 lazily. Encode just enough information in the file name, so that we
1031 do not have to open it to know that it is not interesting for us, for
1032 the most common operations.
1033 - man: document that corrupted journal files is nothing to act on
1034 - rework journald sigbus stuff to use mutex
1035 - Set RLIMIT_NPROC for systemd-journal-xyz, and all other of our
1036 services that run under their own user ids, and use User= (but only
1037 in a world where userns is ubiquitous since otherwise we cannot
1038 invoke those daemons on the host AND in a container anymore). Also,
1039 if LimitNPROC= is used without User= we should warn and refuse
1041 - journalctl --verify: don't show files that are currently being
1042 written to as FAIL, but instead show that their are being written to.
1043 - add journalctl -H that talks via ssh to a remote peer and passes through
1045 - add a version of --merge which also merges /var/log/journal/remote
1046 - journalctl: -m should access container journals directly by enumerating
1047 them via machined, and also watch containers coming and going.
1048 Benefit: nspawn --ephemeral would start working nicely with the journal.
1049 - assign MESSAGE_ID to log messages about failed services
1051 * add a test if all entries in the catalog are properly formatted.
1052 (Adding dashes in a catalog entry currently results in the catalog entry
1053 being silently skipped. journalctl --update-catalog must warn about this,
1054 and we should also have a unit test to check that all our message are OK.)
1057 - document that deps in [Unit] sections ignore Alias= fields in
1058 [Install] units of other units, unless those units are disabled
1059 - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
1060 - document that service reload may be implemented as service reexec
1061 - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
1062 - document systemd-journal-flush.service properly
1063 - documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
1064 - man: document the very specific env the shutdown drop-in tools live in
1065 - man: add more examples to man pages
1066 - man: maybe sort directives in man pages, and take sections from --help and apply them to man too
1067 - document root=gpt-auto properly
1070 - add systemctl switch to dump transaction without executing it
1071 - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
1072 - "systemctl disable" on a static unit prints no message and does
1073 nothing. "systemctl enable" does nothing, and gives a bad message
1074 about it. Should fix both to print nice actionable messages.
1075 - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
1076 - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
1077 - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
1078 - systemctl: "Journal has been rotated since unit was started." message is misleading
1079 - systemctl status output should include list of triggering units and their status
1082 - "systemctl mask" should find all names by which a unit is accessible
1083 (i.e. by scanning for symlinks to it) and link them all to /dev/null
1086 - timer units should get the ability to trigger when:
1088 - Modulate timer frequency based on battery state
1090 * add libsystemd-password or so to query passwords during boot using the password agent logic
1092 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
1094 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
1096 * make repeated alt-ctrl-del presses printing a dump
1098 * hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
1100 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
1103 - emulate /dev/kmsg using CUSE and turn off the syslog syscall
1104 with seccomp. That should provide us with a useful log buffer that
1105 systemd can log to during early boot, and disconnect container logs
1106 from the kernel's logs.
1107 - as soon as networkd has a bus interface, hook up --network-interface=,
1108 --network-bridge= with networkd, to trigger netdev creation should an
1109 interface be missing
1110 - a nice way to boot up without machine id set, so that it is set at boot
1111 automatically for supporting --ephemeral. Maybe hash the host machine id
1112 together with the machine name to generate the machine id for the container
1113 - fix logic always print a final newline on output.
1114 https://github.com/systemd/systemd/pull/272#issuecomment-113153176
1115 - should optionally support receiving WATCHDOG=1 messages from its payload
1117 - optionally automatically add FORWARD rules to iptables whenever nspawn is
1118 running, remove them when shut down.
1121 - refuse mounting over a mount point
1122 - automatically discover .roothash files in dissect, similarly to nspawn
1125 - add an API so that libvirt-lxc can inform us about network interfaces being
1126 removed or added to an existing machine
1127 - "machinectl migrate" or similar to copy a container from or to a
1128 difference host, via ssh
1129 - introduce systemd-nspawn-ephemeral@.service, and hook it into
1130 "machinectl start" with a new --ephemeral switch
1131 - "machinectl status" should also show internal logs of the container in
1133 - "machinectl history"
1135 - "machinectl commit" that takes a writable snapshot of a tree, invokes a
1136 shell in it, and marks it read-only after use
1139 - cryptsetup-generator: allow specification of passwords in crypttab itself
1140 - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
1142 * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
1143 https://bugs.freedesktop.org/show_bug.cgi?id=54712
1145 * add a dependency on standard-conf.xml and other included files to man pages
1147 * MountFlags=shared acts as MountFlags=slave right now.
1149 * properly handle loop back mounts via fstab, especially regards to fsck/passno
1151 * initialize the hostname from the fs label of /, if /etc/hostname does not exist?
1156 - add trigger --subsystem-match=usb/usb_device device
1157 - reimport udev db after MOVE events for devices without dev_t
1159 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
1161 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
1164 - save coredump in Windows/Mozilla minidump format
1165 - when truncating coredumps, also log the full size that the process had, and make a metadata field so we can report truncated coredumps
1167 * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
1169 * be able to specify a forced restart of service A where service B depends on, in case B
1170 needs to be auto-respawned?
1173 - apply "x" on "D" too (see patch from William Douglas)
1174 - instead of ignoring unknown fields, reject them.
1175 - creating new directories/subvolumes/fifos/device nodes
1176 should not follow symlinks. None of the other adjustment or creation
1177 calls follow symlinks.
1179 * make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
1181 * verify that the AF_UNIX sockets of a service in the fs still exist
1182 when we start a service in order to avoid confusion when a user
1183 assumes starting a service is enough to make it accessible
1185 * Make it possible to set the keymap independently from the font on
1186 the kernel cmdline. Right now setting one resets also the other.
1188 * and a dbus call to generate target from current state
1190 * write blog stories about:
1191 - hwdb: what belongs into it, lsusb
1192 - enabling dbus services
1193 - how to make changes to sysctl and sysfs attributes
1195 - how to pass throw-away units to systemd, or dynamically change properties of existing units
1196 - testing with Harald's awesome test kit
1198 - how to develop against journal browsing APIs
1199 - the journal HTTP iface
1200 - non-cgroup resource management
1201 - dynamic resource management with cgroups
1202 - refreshed, longer missions statement
1203 - calendar time events
1204 - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
1205 - how to create your own target
1206 - instantiated apache, dovecot and so on
1207 - hooking a script into various stages of shutdown/rearly booot
1209 * investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
1211 * dot output for --test showing the 'initial transaction'
1214 - When logging about multiple units (stopping BoundTo units, conflicts, etc.),
1215 log both units as UNIT=, so that journalctl -u triggers on both.
1216 - generate better errors when people try to set transient properties
1217 that are not supported...
1218 http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html
1219 - maybe introduce WantsMountsFor=? Usecase:
1220 http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html
1221 - recreate systemd's D-Bus private socket file on SIGUSR2
1222 - move PAM code into its own binary
1223 - when we automatically restart a service, ensure we restart its rdeps, too.
1224 - hide PAM options in fragment parser when compile time disabled
1225 - Support --test based on current system state
1226 - 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().
1227 - after deserializing sockets in socket.c we should reapply sockopts and things
1228 - drop PID 1 reloading, only do reexecing (difficult: Reload()
1229 currently is properly synchronous, Reexec() is weird, because we
1230 cannot delay the response properly until we are back, so instead of
1231 being properly synchronous we just keep open the fd and close it
1232 when done. That means clients do not get a successful method reply,
1233 but much rather a disconnect on success.
1234 - when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
1235 - when a bus name of a service disappears from the bus make sure to queue further activation requests
1238 - allow port=0 in .socket units
1239 - maybe introduce ExecRestartPre=
1240 - add ReloadSignal= for configuring a reload signal to use
1241 - implement Register= switch in .socket units to enable registration
1242 in Avahi, RPC and other socket registration services.
1243 - allow Type=simple with PIDFile=
1244 https://bugzilla.redhat.com/show_bug.cgi?id=723942
1245 - allow writing multiple conditions in unit files on one line
1246 - introduce Type=pid-file
1247 - introduce mix of BindTo and Requisite
1248 - add a concept of RemainAfterExit= to scope units
1249 - Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
1250 - add verification of [Install] section to systemd-analyze verify
1253 - Make sure ID_PATH is always exported and complete for
1254 network devices where possible, so we can safely rely
1258 - add support for more attribute types
1259 - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places
1262 - add more keys to [Route] and [Address] sections
1263 - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
1264 - add reduced [Link] support to .network files
1265 - properly handle routerless dhcp leases
1266 - work with non-Ethernet devices
1267 - dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from?
1268 - the DHCP lease data (such as NTP/DNS) is still made available when
1269 a carrier is lost on a link. It should be removed instantly.
1270 - expose in the API the following bits:
1271 - option 15, domain name
1272 - option 12, hostname and/or option 81, fqdn
1273 - option 123, 144, geolocation
1274 - option 252, configure http proxy (PAC/wpad)
1275 - provide a way to define a per-network interface default metric value
1276 for all routes to it. possibly a second default for DHCP routes.
1277 - allow Name= to be specified repeatedly in the [Match] section. Maybe also
1278 support Name=foo*|bar*|baz ?
1279 - duplicate address check for static IPs (like ARPCHECK in network-scripts)
1280 - whenever uplink info changes, make DHCP server send out FORCERENEW
1282 * Figure out how to do unittests of networkd's state serialization
1285 - figure out how much we can increase Maximum Message Size
1288 - add functions to set previously stored IPv6 addresses on startup and get
1289 them at shutdown; store them in client->ia_na
1290 - write more test cases
1291 - implement reconfigure support, see 5.3., 15.11. and 22.20.
1292 - implement support for temporary adressess (IA_TA)
1293 - implement dhcpv6 authentication
1294 - investigate the usefulness of Confirm messages; i.e. are there any
1295 situations where the link changes without any loss in carrier detection
1297 - some servers don't do rapid commit without a filled in IA_NA, verify
1304 - natively watch for dbus-*.service symlinks (PENDING)
1305 - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
1307 * make cryptsetup lower --iter-time
1309 * kernel: add device_type = "fb", "fbcon" to class "graphics"
1311 * /usr/bin/service should actually show the new command line
1313 * 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=
1315 * neither pkexec nor sudo initialize environ[] from the PAM environment?
1317 * fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
1319 * register catalog database signature as file magic
1321 * zsh shell completion:
1322 - <command> <verb> -<TAB> should complete options, but currently does not
1323 - systemctl add-wants,add-requires
1325 * systemctl status should know about 'systemd-analyze calendar ... --iterations='
1326 * If timer has just OnInactiveSec=..., it should fire after a specified time
1327 after being started.
1331 * look for close() vs. close_nointr() vs. close_nointr_nofail()
1333 * check for strerror(r) instead of strerror(-r)
1337 * set_put(), hashmap_put() return values check. i.e. == 0 does not free()!
1339 * use secure_getenv() instead of getenv() where appropriate
1341 * link up selected blog stories from man pages and unit files Documentation= fields
1342 String is not UTF-8 clean, ignoring assignment