X-Git-Url: https://www.chiark.greenend.org.uk/ucgi/~ianmdlvl/git?p=elogind.git;a=blobdiff_plain;f=TODO;h=255a4f2d0774f9d94d7df6eafdba990a56c6f7ae;hp=a09451d77719b52ad544465842477cf7450ad74e;hb=refs%2Ftags%2Fv219;hpb=409886c424cfaeeb5cf8928905560e5cac2d492d diff --git a/TODO b/TODO index a09451d77..255a4f2d0 100644 --- a/TODO +++ b/TODO @@ -1,11 +1,3 @@ -Preparations for 218: - -* port libmount hookup to use API's own inotify interface - -* cgroup delegation issues - -* should networkd's [BridgePort] section really be called like that? - Bugfixes: * Should systemctl status \* work on all unit types, not just .service? @@ -14,8 +6,6 @@ Bugfixes: automount points even when the original .automount file did not exist anymore. Only the .mount unit was still around. -* properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point. - * ExecStart with unicode characters fails in strv_split_quoted: [Service] @@ -26,12 +16,13 @@ Bugfixes: str_split_quoted() should return a real return code, so spawn_child can report the failure properly. +* When systemctl --host is used, underlying ssh connection can remain open. + bus_close does not kill children? + External: * Fedora: add an rpmlint check that verifies that all unit files in the RPM are listed in %systemd_post macros. -* Fedora: post FPC ticket to move add %tmpfiles_create to the packaging guidelines - * Fedora: move kernel image to /usr/lib/modules/, kernel-install will take care of populating /boot * Fedora: remove /etc/resolv.conf tmpfiles hack @@ -43,11 +34,140 @@ External: Features: -* bash completion for busctl, to make it truly useful +* journalctl --verify: don't show files that are currently being + written to as FAIL, but instead show that their are being written + to. + +* nspawn: allow configuring cgroup (and other) properties via + --property= when invoking from the command line. + +* add udev rule construct SYSCTL{} to write to sysctls + +* assign MESSAGE_ID to log messages about failed services + +* coredump: make the handler check /proc/$PID/rlimits for RLIMIT_CORE, + and supress coredump if turned off. Then change RLIMIT_CORE to + infinity by default for all services. This then allows per-service + control of coredumping. + +* introduce some call that iterates through cmsg and closes all fds + passed in, and use it everywhere... + +* generate better errors when people try to set transient properties + that are not supported... + http://lists.freedesktop.org/archives/systemd-devel/2015-February/028076.html + +* nspawn, if stdout/stderr/stdin are non-ttys, don't set up + /dev/console, but instead just pass the fds through directly. + +* When runlevel3.target is used to define dependencies on other units, + then we don't pick it up currently, since nothing ever references + runlevel3.target, and never figure out it actually is just an alias + for multi-user.target. A hackish fix could be to add a .wants link + from multi-user.target to runlevel3.target, if it is a symlink to + it. Best would be to create this .wants/ symlink from + sysv-generator. systemd would then load the referenced unit, figure + out it is just an alias and that the dependency would be on itself + and suppress it. Thus the alias and its deps would be loaded as + desired. + +* PID 1: when invoking systemctl preset-all on first boots, operate in + an exclusively additive way, i.e. never remove any pre-existing + symlinks, only add new ones. + +* Introduce $LISTEN_NAMES to complement $LISTEN_FDS, containing a + colon separated list of identifiers for the fds passed. + +* networkd: implement BindCarrier= logic to .network units that binds + application of the file to the carrier sense on another interface, + in order to implement uplink/downlink logic. + +* make networkd subscribe to PrepareForSleep(false) and refresh all + DHCP leases then. + +* when the fstab-generator runs in the initrd, it should create a /dev/null mask for systemd-fsck-root.service, to avoid that the the root fs is fsck'ed twice. + +* maybe introduce WantsMountsFor=? Usecase: + http://lists.freedesktop.org/archives/systemd-devel/2015-January/027729.html + +* rework kexec logic to use new kexec_file_load() syscall, so that we + don't have to call kexec tool anymore. + +* The udev blkid built-in should expose a property that reflects + whether media was sensed in USB CF/SD card readers. This should then + be used to control SYSTEMD_READY=1/0 so that USB card readers aren't + picked up by systemd unless they contain a medium. This would mirror + the behaviour we already have for CD drives. + +* We should remove really old cruft from cdrom_id + +* nspawn: emulate /dev/kmsg using CUSE and turn off the syslog syscall + with seccomp. That should provide us with a useful log buffer that + systemd can log to during early boot, and disconnect container logs + from the kernel's logs. + +* networkd/udev: implement SR_IOV configuration in .link files: + http://lists.freedesktop.org/archives/systemd-devel/2015-January/027451.html + +* When RLIMIT_NPROC is set from a unit file it currently always is set + for root, not for the user set in User=, which makes it + useless. After fixing this, set RLIMIT_NPROC for + systemd-journal-xyz, and all other of our services that run under + their own user ids, and use User= (but only in a world where userns + is ubiquitous since otherwise we cannot invoke those daemons on the + host AND in a container anymore). Also, if LimitNPROC= is used + without User= we should warn and refuse operation. + +* logind: maybe allow configuration of the StopTimeout for session scopes + +* Set NoNewPriviliges= on all of our own services, where that makes sense -* journald: broken file systems are real (btrfs), we need to handle - SIGBUS in some way if we cannot write or read from the disk. - https://bugzilla.redhat.com/show_bug.cgi?id=1151848 +* Rework systemctl's GetAll property parsing to use the generic bus_map_all_properties() API + +* rework journald sigbus stuff to use mutex + +* import-dkr: support tarsum checksum verification, if it becomes reality one day... + +* import-dkr: convert json bits to nspawn configuration + +* import: support import from local files, and export to local files + +* core/cgroup: support net_cls modules, and support automatically allocating class ids, then add support for making firewall changes depending on it, to implement a per-service firewall + +* introduce systemd-nspawn-ephemeral@.service, and hook it into "machinectl start" with a new --ephemeral switch + +* logind,machined: add generic catch-all polkit verbs for most privileged operations, similar to systemd itself + +* "machinectl status" should also show internal logs of the container in question + +* "machinectl list-images" should show os-release data, as well as machine-info data (including deployment level) + +* 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 + +* Port various tools to make use of verbs.[ch], where applicable + +* "machinectl history" + +* "machinectl diff" + +* "machinectl commit" that takes a writable snapshot of a tree, invokes a shell in it, and marks it read-only after use + +* "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 + +* 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 + +* systemd-nspawn -x should support ephemeral instances of gpt images + +* move machinectl's mount and copy commands into machined + +* hostnamectl: show root image uuid + +* sysfs set api in libudev is not const + +* Find a solution for SMACK capabilities stuff: + http://lists.freedesktop.org/archives/systemd-devel/2014-December/026188.html + +* port libmount hookup to use API's own inotify interface, as soon as that is table in libmount * "systemctl preset-all" should probably order the unit files it operates on lexicographically before starting to work, in order to @@ -60,7 +180,7 @@ Features: * systemd-journal-upload (or a new, related tool): allow pushing out journal messages onto the network in BSD syslog protocol, - continously. Default to some link-local IP mcast group, to make this + continuously. Default to some link-local IP mcast group, to make this useful as a one-stop debugging tool. * synchronize console access with BSD locks: @@ -80,10 +200,6 @@ Features: * timedated should compensate on SetTime for the time spent in polkit -* figure out what to do with libmount and utab? - -* We really should have a FOREACH macro to iterate through inotify events we read, similar to KDBUS_FOREACH_ITEM - * figure out when we can use the coarse timers * sd-resolve: drop res_query wrapping, people should call via the bus to resolved instead @@ -97,8 +213,6 @@ Features: * timesyncd + resolved: add ugly bus calls to set NTP and DNS servers per-interface, for usage by NM -* networkd-wait-online really should have a timeout by default - * 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 * machined: add an API so that libvirt-lxc can inform us about network interfaces being removed or added to an existing machine @@ -111,8 +225,6 @@ Features: * systemd --user should issue sd_notify() upon reaching basic.target, not on becoming idle -* 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 - * consider showing the unit names during boot up in the status output, not just the unit descriptions * dhcp: do we allow configuring dhcp routes on interfaces that are not the one we got the dhcp info from? @@ -134,7 +246,10 @@ Features: * systemctl: if some operation fails, show log output? -* systemctl edit: add commented help text to the end, like git commit +* systemctl edit: +- allow creation of units from scratch +- use equvalent of cat() to insert existing config as a comment, prepended with #. + Upon editor exit, lines with one # are removed, lines with two # are left with one #, etc. * refcounting in sd-resolve is borked @@ -170,6 +285,7 @@ Features: for all routes to it. possibly a second default for DHCP routes. - allow Name= to be specified repeatedly in the [Match] section. Maybe also support Name=foo*|bar*|baz ? + - duplicate address check for static IPs (like ARPCHECK in network-scripts) * resolved: - put networkd events and rtnl events at a higher priority, so that @@ -255,9 +371,6 @@ Features: desktop UIs such as gnome-shell to freeze apps that are not visible on screen, not unlike how job control works on the shell -* completions: - - manager property enumeration was broken when systemd moved to /usr/lib/ - * cgroups: - implement per-slice CPUFairScheduling=1 switch - handle jointly mounted controllers correctly @@ -306,20 +419,12 @@ Features: ReadOnlyDirectories=... for whitelisting files for a service. * sd-bus: - - make dsrt happy, and change userspace header for kdbus to yyyyuta{tv}v - - kdbus: remove NameOwnerChanged kernel messages for monitors - - kdbus: PID/TID goes missing for method calls from outside the PID namespace? - - kdbus: the kernel should not allow messages to be delivered that have a reply serial != 0, reply-expect unset, but no appropriate window - - kdbus: timestamps on kernel's NameOwnerChanged messages? - - kdbus' busnames.target should get pulled in by basic.target - - Ignore .busname units on classic D-Bus boots, systemd-resolved cannot be started on kdbus - without the active policy and should get a Wants=org.freedesktop.resolve1.busname to - pull-in the policy. + - GetAllProperties() on a non-existing object does not result in a failure currently + - kdbus: process fd=-1 for incoming msgs - port to sd-resolve for connecting to TCP dbus servers - - kdbus: maybe add controlling tty and ppid metadata fields + - kdbus: maybe add controlling tty metadata fields - 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 - when kdbus does not take our message without memfds, try again with memfds - - introduce sd_bus_emit_object_added()/sd_bus_emit_object_removed() that automatically includes the build-in interfaces in the list - see if we can drop more message validation on the sending side - add API to clone sd_bus_message objects - make AddMatch calls on dbus1 transports async? @@ -374,9 +479,6 @@ Features: * teach ConditionKernelCommandLine= globs or regexes (in order to match foobar={no,0,off}) -* after all byte-wise realloc() is slow, even on glibc, so i guess we - need manual exponential loops after all - * BootLoaderSpec: Clarify that the kernel has to be in $BOOT. Clarify that the boot loader should be installed to the ESP. Define a way how an installer can figure out whether a BLS compliant boot loader @@ -469,11 +571,11 @@ Features: - journal: add a setgid "systemd-journal" utility to invoke from libsystemd-journal, which passes fds via STDOUT and does PK access - journactl: support negative filtering, i.e. FOOBAR!="waldo", and !FOOBAR for events without FOOBAR. - - journal: send out marker messages every now and then, and immediately sync with fdatasync() afterwards, in order to have hourly guaranteed syncs. + - journal: store timestamp of journal_file_set_offline() int he header, + so it is possible to display when the file was last synced. - 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. - journal: find a way to allow dropping history early, based on priority, other rules - journal: When used on NFS, check payload hashes - - journal live copy, based on libneon (client) and libmicrohttpd (server) - journald: add kernel cmdline option to disable ratelimiting for debug purposes - refuse taking lower-case variable names in sd_journal_send() and friends. - journald: we currently rotate only after MaxUse+MaxFilesize has been reached. @@ -566,12 +668,7 @@ Features: * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not * nspawn: - - bind mount read-only the cgroup tree higher than nspawn - refuses to boot containers without /etc/machine-id (OK?), and with empty /etc/machine-id (not OK). - - support taking a btrfs snapshot at startup and dropping it afterwards - - introduce machines.target to order after all nspawn instances - - systemd-nspawn@.service should fail if some nspawn arg is invalid, with Type=notify - - PID 1 doesn't apply nspawns devices cgroup policy * cryptsetup: - cryptsetup-generator: allow specification of passwords in crypttab itself @@ -634,6 +731,7 @@ Features: - kill scsi_id - add trigger --subsystem-match=usb/usb_device device - reimport udev db after MOVE events for devices without dev_t + - don't keep stale db in case event processing fails (also notify userspace about the failure) * 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 @@ -658,6 +756,8 @@ Features: * tmpfiles: - apply "x" on "D" too (see patch from William Douglas) + - replace F with f+. + - instead of ignoring unknown fields, reject them. * for services: do not set $HOME in services unless requested @@ -768,7 +868,6 @@ Features: - write more test cases - implement and do duplicate address detection, see rfc 4862, 5.4. - implement reconfigure support, see 5.3., 15.11. and 22.20. - - implement information request, see 1.2. and 18.1.5. - implement support for temporary adressess (IA_TA) - implement dhcpv6 authentication - investigate the usefulness of Confirm messages; i.e. are there any