TODO revision 35c594e044bea03baf5b9b6c429b9ac9bb5aa0b5
1516N/ABugfixes:
1N/A
1N/A* Should systemctl status \* work on all unit types, not just .service?
1N/A
1N/A* Dangling symlinks of .automount unit files in .wants/ directories, set up
1N/A automount points even when the original .automount file did not exist
1N/A anymore. Only the .mount unit was still around.
1N/A
1N/A* ExecStart with unicode characters fails in strv_split_quoted:
1N/A
1N/A [Service]
1N/A Environment=ONE='one' "TWO='two two' too" THREE=
1N/A ExecStart=/bin/python3 -c 'import sys;print(sys.argv)' $ONE $TWO $THREE
1N/A
1N/A* MEMORY return code is overloaded for syntax errors in the command line.
1N/A str_split_quoted() should return a real return code, so spawn_child can
1N/A report the failure properly.
1N/A
1N/AExternal:
1N/A
1N/A* Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros.
1889N/A
1889N/A* Fedora: move kernel image to /usr/lib/modules/, kernel-install will take care of populating /boot
2659N/A
1889N/A* Fedora: remove /etc/resolv.conf tmpfiles hack
1889N/A
1N/A* wiki: update journal format documentation for lz4 additions
22N/A
22N/A* When lz4 gets an API for lz4 command output, make use of it to
539N/A compress coredumps in a way compatible with /usr/bin/lz4.
539N/A
221N/AFeatures:
14N/A
14N/A* teach fd_get_crtime_at() the btrfs crtime
221N/A
14N/A* import pull-gpt: create writable snapshot of downloaded image, by the right name
14N/A
22N/A* nspawn: when start a container "foobar" look for its configuration in a file "foobar.nspawn" in /etc/systemd/nspawn/ as well as next to the actualy directory or image to boot
22N/A
22N/A* Port various tools to make use of verbs.[ch], where applicable
22N/A
22N/A* Check all invocations of access() and consider turning them into laccess()
22N/A
14N/A* "machinectl start/enable/disable foo" as aliases for "systemctl start/enable/disable systemd-nspawn@foo.service"
2160N/A
2160N/A* "machinectl snapshot" to make a snapshot of a tree or container into /var/lib/containers
2446N/A
2160N/A* "machinectl rm" to remove a container tree from /var/lib/containers
2160N/A
2160N/A* "machinectl history"
2160N/A
2160N/A* "machinectl diff"
2160N/A
2500N/A* show btrfs quota in machinectl
2160N/A
2160N/A* "machinectl commit" that takes a writable snapshot of a tree, invokes a shell in it, and marks it read-only after use
2160N/A
2160N/A* "machinectl status" should show 10 most recent log lines of both the host logs of the unit of the machine, plus the logs generated in the machine
2160N/A
2160N/A* add transparent btrfs pool in a loopback file in /var if btrfs operations (such as systemd-import pull-dkr) are used and /var is not a btrfs file system
2160N/A
2160N/A* systemd-nspawn -M should look for gpt images if neither -D nor -i are specified.
2160N/A
2160N/A* systemd-nspawn -i should run things read-only if image file has no writable bits
2160N/A
2160N/A* systemd-nspawn -x should support ephemeral instances of gpt images
2160N/A
2160N/A* systemd-import pull-gpt should make a read-only download, plus a writable reflinked copy
2160N/A
2339N/A* machinectl list should show crtimes of snapshots and GPT images
2160N/A
2160N/A* hostnamectl: show root image uuid
2160N/A
2160N/A* sysfs set api in libudev is not const
2160N/A
2160N/A* Find a solution for SMACK capabilities stuff:
2524N/A http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html
2524N/A
2524N/A* port libmount hookup to use API's own inotify interface, as soon as that is table in libmount
2524N/A
2524N/A* journald: broken file systems are real (btrfs), we need to handle
2524N/A SIGBUS in some way if we cannot write or read from the disk.
2160N/A https://bugzilla.redhat.com/show_bug.cgi?id=1151848
2339N/A
2160N/A* "systemctl preset-all" should probably order the unit files it
2160N/A operates on lexicographically before starting to work, in order to
2160N/A ensure deterministic behaviour if two unit files conflict (like DMs
2160N/A do, for example)
565N/A
2612N/A* resolved should optionally register additional per-interface LLMNR
565N/A names, so that for the container case we can establish the same name
565N/A (maybe "host") for referencing the server, everywhere.
1351N/A
1351N/A* systemd-journal-upload (or a new, related tool): allow pushing out
1351N/A journal messages onto the network in BSD syslog protocol,
1351N/A continously. Default to some link-local IP mcast group, to make this
1431N/A useful as a one-stop debugging tool.
1431N/A
1563N/A* synchronize console access with BSD locks:
1351N/A http://lists.freedesktop.org/archives/systemd-devel/2014-October/024582.html
1351N/A
1351N/A* as soon as we have kdbus, and sender timestamps, revisit coalescing multiple parallel daemon reloads:
2144N/A http://lists.freedesktop.org/archives/systemd-devel/2014-December/025862.html
2144N/A
2144N/A* set $REMOTE_IP (or $REMOTE_ADDR/$REMOTE_PORT) environment variable when doing per-connection socket activation. use format introduced by xinetd or CGI for this
2144N/A
2144N/A* the install state probably shouldn't get confused by generated units, think dbus1/kdbus compat!
2144N/A
2144N/A* in systemctl list-unit-files: show the install value the presets would suggest for a service in a third column
2144N/A
2144N/A* we should try harder to collapse start jobs for swaps that end up being the same:
2144N/A http://lists.freedesktop.org/archives/systemd-devel/2014-November/025359.html
1086N/A
429N/A* timedated should compensate on SetTime for the time spent in polkit
384N/A
2126N/A* figure out when we can use the coarse timers
2126N/A
1086N/A* sd-resolve: drop res_query wrapping, people should call via the bus to resolved instead
1086N/A
497N/A* add "systemctl start -v foobar.service" that shows logs of a service
497N/A while the start command runs. This is non-trivial to do without
497N/A races though, since we should flush out all journal messages before
497N/A returning from the "systemctl stop".
497N/A
2126N/A* 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
2126N/A
2126N/A* timesyncd + resolved: add ugly bus calls to set NTP and DNS servers per-interface, for usage by NM
2126N/A
2126N/A* networkd-wait-online really should have a timeout by default
2126N/A
384N/A* add infrastructure to allocate dynamic/transient users and UID ranges, for use in user-namespaced containers, per-seat gdm login screens and gdm guest sessions
384N/A
1431N/A* machined: add an API so that libvirt-lxc can inform us about network interfaces being removed or added to an existing machine
384N/A
2339N/A* maybe add support for specifier expansion in user.conf, specifically DefaultEnvironment=
429N/A
301N/A* code cleanup: retire FOREACH_WORD_QUOTED, port to unquote_first_word() loops instead
301N/A
301N/A* introduce systemd-timesync-wait.service or so to sync on an NTP fix?
1086N/A
301N/A* systemd --user should issue sd_notify() upon reaching basic.target, not on becoming idle
2047N/A
2047N/A* configure.ac pretends dbus was optional but actually hardcodes use of dbus' pkg-config file to determine various dbus dirs such as policy and activation dirs
2521N/A
2047N/A* consider showing the unit names during boot up in the status output, not just the unit descriptions
2219N/A
2219N/A* dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from?
2047N/A
2047N/A* maybe allow timer units with an empty Units= setting, so that they
2517N/A can be used for resuming the system but nothing else.
2517N/A
2517N/A* what to do about udev db binary stability for apps? (raw access is not an option)
2517N/A
2517N/A* maybe provide an API to allow migration of foreign PIDs into existing scopes.
2047N/A
2517N/A* maybe support a new very "soft" reboot mode, that simply kills all processes, disassembles everything, flushes /run and sysvipc, and then reexecs systemd again
2517N/A
2517N/A* man: document that corrupted journal files is nothing to act on
2517N/A
2089N/A* man: maybe use the word "inspect" rather than "introspect"?
2517N/A
2517N/A* "machinectl list" should probably show columns for OS version and IP addresses
2517N/A
2517N/A* systemctl: if some operation fails, show log output?
2517N/A
2047N/A* systemctl edit:
2219N/A- allow creation of units from scratch
2219N/A- use equvalent of cat() to insert existing config as a comment, prepended with #.
2047N/A Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc.
2047N/A
2047N/A* refcounting in sd-resolve is borked
2339N/A
2339N/A* exponential backoff in timesyncd and resolved when we cannot reach a server
2339N/A
2339N/A* tmpfiles: port to unquote_many_words(), similar to sysusers
2339N/A
2339N/A* unquote_many_words() should probably be used by a lot of code that
2339N/A currently uses FOREACH_WORD and friends. For example, most conf
2339N/A parsing callbacks should use it.
2339N/A
2453N/A* merge ~/.local/share and ~/.local/lib into one similar /usr/lib and /usr/share....
2453N/A
2453N/A* systemd.show_status= should probably have a mode where only failed
2453N/A units are shown.
2339N/A
2339N/A* networkd:
2339N/A - add LLDP client side support
2339N/A - the DHCP lease data (such as NTP/DNS) is still made available when
2339N/A a carrier is lost on a link. It should be removed instantly.
2339N/A - .network setting that allows overriding of the hostname to send to the dhcp server
2339N/A http://lists.freedesktop.org/archives/systemd-devel/2014-July/021550.html
2339N/A - expose in the API the following bits:
2339N/A - option 15, domain name and/or option 119, search list
2452N/A - option 12, host name and/or option 81, fqdn
2452N/A - option 100, 101, timezone
2452N/A - option 123, 144, geolocation
2339N/A - option 252, configure http proxy (PAC/wpad)
2339N/A - networkd's dhcp server should transparently pass on the DNS and
2339N/A NTP server list it got from user configuration and its dhcp client
2339N/A to clients. It should also pass on its own timezone information.
2339N/A - provide a way to define a per-network interface default metric value
2339N/A for all routes to it. possibly a second default for DHCP routes.
2339N/A - allow Name= to be specified repeatedly in the [Match] section. Maybe also
2339N/A support Name=foo*|bar*|baz ?
2339N/A
2339N/A* resolved:
2339N/A - put networkd events and rtnl events at a higher priority, so that
2339N/A we always process them before we process client requests
2339N/A - DNSSEC
2339N/A - use base64 for key presentation?
2339N/A - add display of private key types (http://tools.ietf.org/html/rfc4034#appendix-A.1.1)?
2339N/A - add nice formatting of DNS timestamps
2339N/A - DNS
2219N/A - search paths
2219N/A - mDNS/DNS-SD
2219N/A - avahi compat
2219N/A - DNS-SD service registration from socket units
2219N/A - edns0
2219N/A - dname
2219N/A - cname on PTR (?)
2047N/A - maybe randomize DNS UDP source ports
2047N/A - maybe compare query section of DNS replies
2047N/A
2047N/A* Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
2047N/A
2219N/A* Add a new verb "systemctl top"
2517N/A
2517N/A* add new gpt type for btrfs volumes
2517N/A
2517N/A* support empty /etc boots nicely:
2047N/A - nspawn/gpt-generator: introduce new gpt partition type for /usr
2047N/A - fstab-generator: support systemd.volatile=yes|no|state on the kernel cmdline, too, similar to nspawn's --volatile=
2047N/A
2047N/A* generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
2047N/A
2047N/A* timer units: actually add extra delays to timer units with high AccuracySec values, don't start them already when we are awake...
2258N/A
2431N/A* a way for container managers to turn off getty starting via $container_headless= or so...
2047N/A
2517N/A* figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
2517N/A
2517N/A* mount_cgroup_controllers(): symlinks need to get the label applied
2517N/A
2517N/A* For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
2047N/A they run added to the initial transaction and thus confuse Type=idle.
2047N/A
2517N/A* Run most system services with cgroupfs read-only and procfs with a more secure mode (doesn't work, since the hidepid= option is per-pid-namespace, not per-mount)
2517N/A
2517N/A* sd-event: generate a failure of a default event loop is executed out-of-thread
2517N/A
2517N/A* add bus api to query unit file's X fields.
2047N/A
2453N/A* consider adding RuntimeDirectoryUser= + RuntimeDirectoryGroup=
2453N/A
2517N/A* sd-event: define more intervals where we will shift wakeup intervals around in, 1h, 6h, 24h, ...
2517N/A
2517N/A* gpt-auto-generator:
2517N/A - Support LUKS for root devices
2517N/A - Define new partition type for encrypted swap? Support probed LUKS for encrypted swap?
2453N/A - Make /home automount rather than mount?
2517N/A
2453N/A* add generator that pulls in systemd-network from containers when
2453N/A CAP_NET_ADMIN is set, more than the loopback device is defined, even
2453N/A when it is otherwise off
2047N/A
2047N/A* MessageQueueMessageSize= and RLimitFSIZE= (and suchlike) should use parse_iec_size().
2228N/A
2228N/A* "busctl status" works only as root on dbus1, since we cannot read
2452N/A /proc/$PID/exe
2452N/A
2047N/A* implement Distribute= in socket units to allow running multiple
2047N/A service instances processing the listening socket, and open this up
2047N/A for ReusePort=
2047N/A
2047N/A* socket units: support creating sockets in different namespace,
2047N/A opening it up for JoinsNamespaceOf=. This would require to fork off
2047N/A a tiny process that joins the namespace and creates/binds the socket
2047N/A and passes this back to PID1 via SCM_RIGHTS. This also could be used
2047N/A to allow Chown/chgrp on sockets without requiring NSS in PID 1.
2047N/A
2047N/A* New service property: maximum CPU and wallclock runtime for a service
2047N/A
2047N/A* introduce bus call FreezeUnit(s, b), as well as "systemctl freeze
2047N/A $UNIT" and "systemctl thaw $UNIT" as wrappers around this. The calls
2047N/A should SIGSTOP all unit processes in a loop until all processes of
2431N/A it are fully stopped. This can later be used for app management by
2047N/A desktop UIs such as gnome-shell to freeze apps that are not visible
2047N/A on screen, not unlike how job control works on the shell
2047N/A
2047N/A* completions:
2047N/A - manager property enumeration was broken when systemd moved to /usr/lib/
2047N/A
2047N/A* cgroups:
2047N/A - implement per-slice CPUFairScheduling=1 switch
2047N/A - handle jointly mounted controllers correctly
2047N/A - introduce high-level settings for RT budget, swappiness
2047N/A - how to reset dynamically changed unit cgroup attributes sanely?
2135N/A - when reloading configuration, apply new cgroup configuration
2047N/A - when recursively showing the cgroup hierarchy, optionally also show
2047N/A the hierarchies of child processes
2144N/A
2047N/A* transient units:
2521N/A - allow creating auxiliary units with the same call
2521N/A - add field to transient units that indicate whether systemd or somebody else saves/restores its settings, for integration with libvirt
2503N/A - ensure scope units may be started only a single time
2503N/A
2339N/A* Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk?
2339N/A
2339N/A* when we detect low battery and no AC on boot, show pretty splash and refuse boot
2521N/A
2339N/A* machined, localed: when we try to kill an empty cgroup, generate an ESRCH error over the bus
2521N/A
2521N/A* libsystemd-journal, libsystemd-login, libudev: add calls to easily attach these objects to sd-event event loops
2521N/A
2503N/A* be more careful what we export on the bus as (usec_t) 0 and (usec_t) -1
2503N/A
2339N/A* unify dispatch table in systemctl_main() and friends
2521N/A
2339N/A* 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
2339N/A
2339N/A* After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
2339N/A
2047N/A* If we try to find a unit via a dangling symlink, generate a clean
2047N/A error. Currently, we just ignore it and read the unit from the search
2047N/A path anyway.
2047N/A
2047N/A* refuse boot if /etc/os-release is missing or /etc/machine-id cannot be set up
2047N/A
2047N/A* btrfs raid assembly: some .device jobs stay stuck in the queue
2047N/A
2047N/A* make sure gdm does not use multi-user-x but the new default X configuration file, and then remove multi-user-x from systemd
2047N/A
2047N/A* man: the documentation of Restart= currently is very misleading and suggests the tools from ExecStartPre= might get restarted.
2047N/A
2047N/A* load .d/*.conf dropins for device units
2047N/A
2047N/A* allow implementation of InaccessibleDirectories=/ plus
2047N/A ReadOnlyDirectories=... for whitelisting files for a service.
2047N/A
2047N/A* sd-bus:
2047N/A - kdbus: process fd=-1 for incoming msgs
2047N/A - make dsrt happy, and change userspace header for kdbus to yyyyuta{tv}v
2521N/A - kdbus: PID/TID goes missing for method calls from outside the PID namespace?
2521N/A - kdbus: the kernel should not allow messages to be delivered that have a reply serial != 0, reply-expect unset, but no appropriate window
2521N/A - kdbus: timestamps on kernel's NameOwnerChanged messages?
2521N/A - kdbus' busnames.target should get pulled in by basic.target
2521N/A - Ignore .busname units on classic D-Bus boots, systemd-resolved cannot be started on kdbus
2047N/A without the active policy and should get a Wants=org.freedesktop.resolve1.busname to
2047N/A pull-in the policy.
2047N/A - port to sd-resolve for connecting to TCP dbus servers
2521N/A - kdbus: maybe add controlling tty metadata fields
2047N/A - 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
2047N/A - when kdbus does not take our message without memfds, try again with memfds
2047N/A - introduce sd_bus_emit_object_added()/sd_bus_emit_object_removed() that automatically includes the build-in interfaces in the list
2047N/A - see if we can drop more message validation on the sending side
1266N/A - add API to clone sd_bus_message objects
1431N/A - make AddMatch calls on dbus1 transports async?
1266N/A - kdbus: matches against source or destination pids for an "strace -p"-like feel. Problem: The PID info needs to be available in userspace too...
1266N/A - longer term: priority inheritance
1431N/A - dbus spec updates:
0N/A - kdbus mapping
0N/A - NameLost/NameAcquired obsolete
2047N/A - GVariant
2047N/A - path escaping
2047N/A - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
2047N/A - test bloom filter generation indexes
2047N/A - bus-proxy: when passing messages from kdbus, make sure we properly
2047N/A handle the case where a large number of fds is appended that we
2047N/A cannot pass into sendmsg() of the AF_UNIX sokcet (which only accepts
2047N/A 253 messages)
0N/A - kdbus: introduce a concept of "send-only" connections
362N/A - 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.
934N/A
0N/A* sd-event
0N/A - allow multiple signal handlers per signal?
384N/A - document chaining of signal handler for SIGCHLD and child handlers
1086N/A
0N/A* in the final killing spree, detect processes from the root directory, and
2339N/A complain loudly if they have argv[0][0] == '@' set.
526N/A https://bugzilla.redhat.com/show_bug.cgi?id=961044
1835N/A
1835N/A* investigate endianness issues of UUID vs. GUID
1774N/A
2339N/A* dbus: when a unit failed to load (i.e. is in UNIT_ERROR state), we
1835N/A should be able to safely try another attempt when the bus call LoadUnit() is invoked.
1835N/A
1835N/A* add a pam module that passes the hdd passphrase into the PAM stack and then expires it, for usage by gdm auto-login.
1835N/A
1835N/A* add a pam module that on password changes updates any LUKS slot where the password matches
1835N/A
1835N/A* maybe add a generator that looks for "systemd.run=" on the kernel cmdline for container usercases...
1835N/A
1835N/A* cgtop: make cgtop useful in a container
2183N/A
1835N/A* test/:
1835N/A - add 'set -e' to scripts in test/
1774N/A - make stuff in test/ work with separate output dir
1774N/A
1835N/A* seems that when we follow symlinks to units we prefer the symlink
526N/A destination path over /etc and /usr. We should not do that. Instead
2339N/A /etc should always override /run+/usr and also any symlink
2339N/A destination.
2339N/A
2339N/A* when isolating, try to figure out a way how we implicitly can order
996N/A all units we stop before the isolating unit...
996N/A
2339N/A* teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off})
2339N/A
1537N/A* after all byte-wise realloc() is slow, even on glibc, so i guess we
2339N/A need manual exponential loops after all
2339N/A
2339N/A* BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify
1639N/A that the boot loader should be installed to the ESP. Define a way
2339N/A how an installer can figure out whether a BLS compliant boot loader
1639N/A is installed.
2339N/A
1639N/A* think about requeuing jobs when daemon-reload is issued? usecase:
2339N/A the initrd issues a reload after fstab from the host is accessible
1639N/A and we might want to requeue the mounts local-fs acquired through
1639N/A that automatically.
219N/A
2336N/A* systemd-inhibit: make taking delay locks useful: support sending SIGINT or SIGTERM on PrepareForSleep()
301N/A
2336N/A* journal-or-kmsg is currently broken? See reverted commit 4a01181e460686d8b4a543b1dfa7f77c9e3c5ab8.
219N/A
2544N/A* remove any syslog support from log.c -- we probably cannot do this before split-off udev is gone for good
219N/A
1835N/A* shutdown logging: store to EFI var, and store to USB stick?
1835N/A
1835N/A* write UI tool that pops up emergency messages from the journal as notification
1835N/A
2339N/A* think about window-manager-run-as-user-service problem: exit 0 → activate shutdown.target; exit != 0 → restart service
1835N/A
1835N/A* merge unit_kill_common() and unit_kill_context()
1835N/A
2339N/A* introduce ExecCondition= in services
1537N/A
1537N/A* EFI:
1537N/A - honor language efi variables for default language selection (if there are any?)
1537N/A - honor timezone efi variables for default timezone selection (if there are any?)
1537N/A - change bootctl to be backed by systemd-bootd to control temporary and persistent default boot goal plus efi variables
1537N/A
1537N/A* maybe do not install getty@tty1.service symlink in /etc but in /usr?
1537N/A
1538N/A* re-enable "make check" for gtk-doc (broken for unknown reason)
1538N/A
1538N/A* fstab: add new mount option x-systemd-after=/foobar/waldo to allow manual dependencies to other mount points
1538N/A https://bugzilla.redhat.com/show_bug.cgi?id=812826
1538N/A
1538N/A* print a nicer explanation if people use variable/specifier expansion in ExecStart= for the first word
1538N/A
1736N/A* mount: turn dependency information from /proc/self/mountinfo into dependency information between systemd units.
1538N/A
1538N/A* logind:
1538N/A - logind: optionally, ignore idle-hint logic for autosuspend, block suspend as long as a session is around
1538N/A - When we update the kernel all kind of hibernation should be prohibited until shutdown/reboot
1538N/A - logind: wakelock/opportunistic suspend support
1538N/A - Add pretty name for seats in logind
1538N/A - logind: allow showing logout dialog from system?
1537N/A - we should probably handle SIGTERM/SIGINT to not leave dot files around, just in case
1537N/A - session scopes/user unit: add RequiresMountsFor for the home directory of the user
1537N/A - add Suspend() bus calls which take timestamps to fix double suspend issues when somebody hits suspend and closes laptop quickly.
1537N/A - if pam_systemd is invoked by su from a process that is outside of a
1537N/A any session we should probably just become a NOP, since that's
1537N/A usually not a real user session but just some system code that just
2336N/A needs setuid().
1537N/A - logind: make the Suspend()/Hibernate() bus calls wait for the for
2336N/A the job to be completed. before returning, so that clients can wait
2336N/A for "systemctl suspend" to finish to know when the suspending is
2336N/A complete.
2336N/A - logind: when the power button is pressed short, just popup a
1537N/A logout dialog. If it is pressed for 1s, do the usual
1537N/A shutdown. Inspiration are Macs here.
1774N/A - logind: allow users to kill or lock their own sessions
526N/A - expose "Locked" property on logind sesison objects
2310N/A - given that logind now lets PID 1 do all nasty work, we can
2310N/A probably reduce the capability set it retains substantially.
2310N/A (we need CAP_SYS_ADMIN for drmSetMaster(), so maybe not worth it)
219N/A - expose orientation sensors and tablet mode through logind
1970N/A
2219N/A* exec: when deinitializating a tty device fix the perms and group, too, not only when initializing. Set access mode/gid to 0620/tty.
2493N/A
1537N/A* service: watchdog logic: for testing purposes allow ping, but do not require pong
2339N/A
1537N/A* journal:
2336N/A - consider introducing implicit _TTY= + _PPID= + _EUID= + _EGID= + _FSUID= + _FSGID= fields
2336N/A - import and delete pstore filesystem content at startup
1537N/A - journald: also get thread ID from client, plus thread name
2336N/A - journal: when waiting for journal additions in the client always sleep at least 1s or so, in order to minimize wakeups
1537N/A - add API to close/reopen/get fd for journal client fd in libsystemd-journal.
1537N/A - fallback to /dev/log based logging in libsystemd-journal, if we cannot log natively?
1537N/A - declare the local journal protocol stable in the wiki interface chart
2336N/A - journal: reuse XZ context
1537N/A - sd-journal: speed up sd_journal_get_data() with transparent hash table in bg
2336N/A - journald: when dropping msgs due to ratelimit make sure to write
2339N/A "dropped %u messages" not only when we are about to print the next
1537N/A message that works, but alraedy after a short tiemout
2336N/A - check if we can make journalctl by default use --follow mode inside of less if called without args?
1537N/A - maybe add API to send pairs of iovecs via sd_journal_send
1537N/A - journal: when writing journal auto-rotate if time jumps backwards
1537N/A - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access
1537N/A - journactl: support negative filtering, i.e. FOOBAR!="waldo",
1537N/A and !FOOBAR for events without FOOBAR.
301N/A - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs.
1537N/A - 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.
2336N/A - journal: find a way to allow dropping history early, based on priority, other rules
1537N/A - journal: When used on NFS, check payload hashes
1537N/A - journal live copy, based on libneon (client) and libmicrohttpd (server)
1537N/A - journald: add kernel cmdline option to disable ratelimiting for debug purposes
1537N/A - refuse taking lower-case variable names in sd_journal_send() and friends.
307N/A - journald: we currently rotate only after MaxUse+MaxFilesize has been reached.
1537N/A - journal: deal nicely with byte-by-byte copied files, especially regards header
1537N/A - journal: sanely deal with entries which are larger than the individual file size, but where the components would fit
1774N/A - Replace utmp, wtmp, btmp, and lastlog completely with journal
301N/A - journalctl: instead --after-cursor= maybe have a --cursor=XYZ+1 syntax?
1537N/A - when a kernel driver logs in a tight loop, we should ratelimit that too.
2336N/A - journald: optionally, log debug messages to /run but everything else to /var
1537N/A - journald: when we drop syslog messages because the syslog socket is
1537N/A full, make sure to write how many messages are lost as first thing
1774N/A to syslog when it works again.
1537N/A - journald: make sure ratelimit is actually really per-service with the new cgroup changes
1537N/A - change systemd-journal-flush into a service that stays around during
1970N/A boot, and causes the journal to be moved back to /run on shutdown,
1970N/A so that we do not keep /var busy. This needs to happen synchronously,
1537N/A hence doing this via signals is not going to work.
1537N/A - optionally support running journald from the command line for testing purposes in external projects
1537N/A - journald: allow per-priority and per-service retention times when rotating/vacuuming
1774N/A - journald: make use of uid-range.h to managed uid ranges to split
1537N/A journals in.
2336N/A - journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
301N/A - improve journalctl performance by loading journal files
1970N/A lazily. Encode just enough information in the file name, so that we
1537N/A do not have to open it to know that it is not interesting for us, for
1970N/A the most common operations.
1970N/A
2524N/A* document:
1970N/A - document that deps in [Unit] sections ignore Alias= fields in
1970N/A [Install] units of other units, unless those units are disabled
1970N/A - man: clarify that time-sync.target is not only sysv compat but also useful otherwise. Same for similar targets
1970N/A - document the exit codes when services fail before they are exec()ed
1537N/A - document that service reload may be implemented as service reexec
2524N/A - document in wiki how to map ical recurrence events to systemd timer unit calendar specifications
1505N/A - add a man page containing packaging guidelines and recommending usage of things like Documentation=, PrivateTmp=, PrivateNetwork= and ReadOnlyDirectories=/etc /usr.
1970N/A - document systemd-journal-flush.service properly
1970N/A - documentation: recommend to connect the timer units of a service to the service via Also= in [Install]
1970N/A - man: document the very specific env the shutdown drop-in tools live in
1970N/A - man: extend runlevel(8) to mention that runlevels suck, and are dead. Maybe add runlevel(7) with a note about that too
1537N/A - man: add more examples to man pages
2524N/A - man: maybe sort directives in man pages, and take sections from --help and apply them to man too
1505N/A
1970N/A* systemctl:
1970N/A - systemctl list-jobs - show dependencies
2524N/A - add systemctl switch to dump transaction without executing it
1970N/A - Add a verbose mode to "systemctl start" and friends that explains what is being done or not done
556N/A - "systemctl disable" on a static unit prints no message and does
1537N/A nothing. "systemctl enable" does nothing, and gives a bad message
1505N/A about it. Should fix both to print nice actionable messages.
1755N/A - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
1859N/A - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
1755N/A - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
2339N/A - systemctl: "Journal has been rotated since unit was started." message is misleading
1672N/A - support "systemctl stop foobar@.service" to stop all units matching a certain template
565N/A - Something is wrong with symlink handling of "autovt@.service" in "systemctl list-unit-files"
526N/A - better error message if you run systemctl without systemd running
526N/A - systemctl status output should should include list of triggering units and their status
526N/A
1537N/A* unit install:
2524N/A - "systemctl mask" should find all names by which a unit is accessible
1505N/A (i.e. by scanning for symlinks to it) and link them all to /dev/null
219N/A - systemctl list-unit-files should list generated files (and probably with a new state "generated" for them, or so)
1970N/A
1970N/A* timer units:
1970N/A - timer units should get the ability to trigger when:
1970N/A o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
1970N/A o DST changes
1970N/A - Support 2012-02~4 as syntax for specifying the fourth to last day of the month.
1970N/A - calendarspec: support value ranges with ".." notation. Example: 2013-4..8-1
1853N/A - when parsing calendar timestamps support the UTC timezone (even if we will not support arbitrary timezone specs, support UTC itself certainly makes sense), also support syntaxes such as +0200
2339N/A - Modulate timer frequency based on battery state
1537N/A
1853N/A* add libsystemd-password or so to query passwords during boot using the password agent logic
2452N/A
1853N/A* 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().
2339N/A
1853N/A* fedup: add --unit to systemctl switch-root somehow
1970N/A* fedup: do not delete initrd on switch-root
1970N/A* fedup: generator
1970N/A
2339N/A* clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
526N/A
1853N/A* on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
2339N/A
1774N/A* make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
1774N/A waiting for the timeout
1774N/A
1774N/A* hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ...
1774N/A
2524N/A* currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
1505N/A
301N/A* nspawn:
2500N/A - bind mount read-only the cgroup tree higher than nspawn
235N/A - refuses to boot containers without /etc/machine-id (OK?), and with empty /etc/machine-id (not OK).
2500N/A - introduce machines.target to order after all nspawn instances
2500N/A - systemd-nspawn@.service should fail if some nspawn arg is invalid, with Type=notify
235N/A - PID 1 doesn't apply nspawns devices cgroup policy
235N/A
236N/A* cryptsetup:
2339N/A - cryptsetup-generator: allow specification of passwords in crypttab itself
2500N/A - move cryptsetup key caching into kernel keyctl?
2500N/A https://bugs.freedesktop.org/show_bug.cgi?id=54982
236N/A - support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator
2339N/A
2500N/A* hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it
2500N/A https://bugs.freedesktop.org/show_bug.cgi?id=54712
235N/A
222N/A* after deserializing sockets in socket.c we should reapply sockopts and things
2158N/A
2500N/A* make timer units go away after they elapsed
1618N/A
1618N/A* move PID 1 segfaults to /var/lib/systemd/coredump?
1618N/A
1618N/A* create /sbin/init symlinks from the build system
1618N/A
1618N/A* allow writing multiple conditions in unit files on one line
1618N/A
1618N/A* MountFlags=shared acts as MountFlags=slave right now.
1618N/A
1618N/A* drop PID 1 reloading, only do reexecing (difficult: Reload()
2158N/A currently is properly synchronous, Reexec() is weird, because we
1970N/A cannot delay the response properly until we are back, so instead of
1970N/A being properly synchronous we just keep open the fd and close it
1970N/A when done. That means clients do not get a successful method reply,
1970N/A but much rather a disconnect on success.
1970N/A
1970N/A* properly handle loop back mounts via fstab, especially regards to fsck/passno
1970N/A
1970N/A* initialize the hostname from the fs label of /, if /etc/hostname does not exist?
1970N/A
1970N/A* rename "userspace" to "core-os"
1970N/A
1970N/A* load-fragment: when loading a unit file via a chain of symlinks
1970N/A verify that it is not masked via any of the names traversed.
1970N/A
1970N/A* introduce Type=pid-file
1970N/A
1970N/A* change Requires=basic.target to RequisiteOverride=basic.target
1970N/A
1970N/A* when breaking cycles drop sysv services first, then services from /run, then from /etc, then from /usr
1970N/A
1970N/A* automount: implement expire:
1970N/A - set superblock timeout AUTOFS_DEV_IOCTL_TIMEOUT_CMD
1970N/A - periodically run AUTOFS_DEV_IOCTL_EXPIRE_CMD
1970N/A - every timeout/4 (original autofs logic)
1970N/A - blocking, needs a thread
1970N/A - run until -EAGAIN
2026N/A - receive expire packet on pipe if kernel tells the timeout is over
1970N/A - call umount
1970N/A - answer expire packet on pipe with AUTOFS_DEV_IOCTL_{READY,FAIL}_CMD
1970N/A - AUTOFS_DEV_IOCTL_EXPIRE_CMD returns
1685N/A
1685N/A* ExecOnFailure=/usr/bin/foo
1970N/A
1970N/A* udev:
1970N/A - move to LGPL
1970N/A - kill scsi_id
1970N/A - add trigger --subsystem-match=usb/usb_device device
1970N/A - reimport udev db after MOVE events for devices without dev_t
2026N/A
2026N/A* 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
2026N/A
1970N/A* introduce mix of BindTo and Requisite
1970N/A
1970N/A* There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
1970N/A
1970N/A* add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
1970N/A
1970N/A* default unix qlen is too small (10). bump sysctl? add sockopt?
1970N/A
1970N/A* save coredump in Windows/Mozilla minidump format
1970N/A
1970N/A* support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting)
1970N/A
1970N/A* default to actual 32-bit PIDs, via /proc/sys/kernel/pid_max
1970N/A
1970N/A* be able to specify a forced restart of service A where service B depends on, in case B
1970N/A needs to be auto-respawned?
1970N/A
1970N/A* when a bus name of a service disappears from the bus make sure to queue further activation requests
1970N/A
1970N/A* tmpfiles:
1970N/A - apply "x" on "D" too (see patch from William Douglas)
1970N/A
1970N/A* for services: do not set $HOME in services unless requested
583N/A
583N/A* hide PAM options in fragment parser when compile time disabled
583N/A
2154N/A* when we automatically restart a service, ensure we restart its rdeps, too.
2154N/A
2154N/A* allow Type=simple with PIDFile=
2154N/A https://bugzilla.redhat.com/show_bug.cgi?id=723942
583N/A
1775N/A* move PAM code into its own binary
583N/A
2105N/A* implement Register= switch in .socket units to enable registration
834N/A in Avahi, RPC and other socket registration services.
1945N/A
1945N/A* make sure systemd-ask-password-wall does not shutdown systemd-ask-password-console too early
605N/A
1814N/A* add ReloadSignal= for configuring a reload signal to use
2105N/A
1814N/A* verify that the AF_UNIX sockets of a service in the fs still exist
1328N/A when we start a service in order to avoid confusion when a user
1618N/A assumes starting a service is enough to make it accessible
2154N/A
2154N/A* Make it possible to set the keymap independently from the font on
1755N/A the kernel cmdline. Right now setting one resets also the other.
1775N/A
1859N/A* and a dbus call to generate target from current state
2026N/A
2026N/A* GC unreferenced jobs (such as .device jobs)
2205N/A
2205N/A* write blog stories about:
1710N/A - hwdb: what belongs into it, lsusb
2144N/A - enabling dbus services
2144N/A - status update
2154N/A - how to make changes to sysctl and sysfs attributes
2524N/A - remote access
2144N/A - how to pass throw-away units to systemd, or dynamically change properties of existing units
1618N/A - testing with Harald's awesome test kit
1618N/A - auto-restart
1618N/A - how to develop against journal browsing APIs
1618N/A - the journal HTTP iface
1618N/A - non-cgroup resource management
1618N/A - dynamic resource management with cgroups
1618N/A - refreshed, longer missions statement
1618N/A - calendar time events
1618N/A - init=/bin/sh vs. "emergency" mode, vs. "rescue" mode, vs. "multi-user" mode, vs. "graphical" mode, and the debug shell
1618N/A - how to create your own target
2154N/A - instantiated apache, dovecot and so on
2524N/A - hooking a script into various stages of shutdown/rearly booot
1618N/A
1328N/A* allow port=0 in .socket units
1431N/A
1431N/A* recreate systemd's D-Bus private socket file on SIGUSR2
1431N/A
1431N/A* Support --test based on current system state
2154N/A
2524N/A* investigate whether the gnome pty helper should be moved into systemd, to provide cgroup support.
1505N/A
2161N/A* maybe introduce ExecRestartPre=
2154N/A
2154N/A* dot output for --test showing the 'initial transaction'
2154N/A
583N/A* fingerprint.target, wireless.target, gps.target, netdevice.target
2154N/A
2524N/A* systemd-python:
1505N/A - figure out a simple way to wait for journal events in a way that
2524N/A works with ^C
1505N/A - add documentation to systemd.daemon
789N/A
2158N/A* bootchart:
235N/A - plot per-process IO utilization
235N/A - group processes based on service association (cgroups)
2183N/A - document initcall_debug
2034N/A - kernel cmdline "bootchart" option for simplicity?
443N/A
443N/A* udev-link-config:
222N/A - Make sure ID_PATH is always exported and complete for
227N/A network devices where possible, so we can safely rely
235N/A on Path= matching
235N/A - check MTUBytes parsing (expecting size_t but we are using unsigned)
227N/A
2034N/A* sd-rtnl:
227N/A - add support for more attribute types
227N/A - inbuilt piping support (essentially degenerate async)? see loopback-setup.c and other places
235N/A
2183N/A* networkd:
235N/A - add more keys to [Route] and [Address] sections
235N/A - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
235N/A - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
1266N/A - add reduced [Link] support to .network files
2207N/A - add Scope= parsing option for [Network]
2207N/A - properly handle routerless dhcp leases
2047N/A - add more attribute support for SIT tunnel
2158N/A - work with non-ethernet devices
227N/A - add support for more bond options
1970N/A
1970N/A* networkd-wait-online:
2500N/A - make operstates to wait for configurable?
1970N/A
1970N/A* dhcp:
1970N/A - figure out how much we can increase Maximum Message Size
1970N/A - export timezone information
1970N/A - support RFC4702 (pass FQDN)
1970N/A
1970N/A* dhcp6:
1970N/A - add functions to set previously stored IPv6 addresses on startup and get
1970N/A them at shutdown; store them in client->ia_na
1970N/A - write more test cases
1970N/A - implement and do duplicate address detection, see rfc 4862, 5.4.
1970N/A - implement reconfigure support, see 5.3., 15.11. and 22.20.
1970N/A - implement information request, see 1.2. and 18.1.5.
1970N/A - implement support for temporary adressess (IA_TA)
1970N/A - implement dhcpv6 authentication
1970N/A - investigate the usefulness of Confirm messages; i.e. are there any
2158N/A situations where the link changes without any loss in carrier detection
1970N/A or interface down
1970N/A - some servers don't do rapid commit without a filled in IA_NA, verify
1970N/A this behavior
1970N/A
1970N/AExternal:
1970N/A
1970N/A* dbus:
2158N/A - natively watch for dbus-*.service symlinks (PENDING)
1970N/A - teach dbus to activate all services it finds in /etc/systemd/services/org-*.service
1970N/A
2158N/A* fix alsa mixer restore to not print error when no config is stored
1970N/A
1970N/A* make cryptsetup lower --iter-time
1970N/A
1970N/A* patch kernel for xattr support in /dev, /proc/, /sys?
1970N/A
1970N/A* kernel: add device_type = "fb", "fbcon" to class "graphics"
1970N/A
1970N/A* drop accountsservice's StandardOutput=syslog and Type=dbus fields
2463N/A
2463N/A* dbus upstream still refers to dbus.target and should not
1970N/A
1970N/A* dbus: in fedora, make /var/lib/dbus/machine-id a symlink to /etc/machine-id
2463N/A
1970N/A* add "# export SYSTEMD_PAGER=" to bash login
1970N/A
1970N/A* /usr/bin/service should actually show the new command line
1970N/A
2026N/A* 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=
2026N/A
1970N/A* fedora: F20: go timer units all the way, leave cron.daily for cron
384N/A
1970N/A* neither pkexec nor sudo initialize environ[] from the PAM environment?
1970N/A
1970N/A* fedora: update policy to declare access mode and ownership of unit files to root:root 0644, and add an rpmlint check for it
1970N/A
1970N/A* register catalog database signature as file magic
1970N/A
1970N/A* zsh shell completion:
1970N/A - <command> <verb> -<TAB> should complete options, but currently does not
1970N/A - systemctl add-wants,add-requires
1970N/A
1970N/A
1970N/ARegularly:
1970N/A
2144N/A* look for close() vs. close_nointr() vs. close_nointr_nofail()
2144N/A
2144N/A* check for strerror(r) instead of strerror(-r)
1970N/A
419N/A* Use PR_SET_PROCTITLE_AREA if it becomes available in the kernel
1970N/A
1970N/A* pahole
1970N/A
1970N/A* set_put(), hashmap_put() return values check. i.e. == 0 does not free()!
1970N/A
2260N/A* use secure_getenv() instead of getenv() where appropriate
1970N/A
1970N/A* link up selected blog stories from man pages and unit files Documentation= fields
419N/A
2463N/AScheduled for removal or fixing:
419N/A
2463N/A* xxxOverridable dependencies (probably: fix)
1431N/A