X-Git-Url: http://www.chiark.greenend.org.uk/ucgi/~ianmdlvl/git?p=elogind.git;a=blobdiff_plain;f=TODO;h=eae357fa2eba567e4ecd80dcf85a57b35ceafd98;hp=4d6d55afee7c0d9ab32ac8e40287f7980676e732;hb=7cd28a8a2fdf01293b0c07681a5f36f557fe51b4;hpb=4ce849853c46b1e857df3c6c9e7752159a58ddf1 diff --git a/TODO b/TODO index 4d6d55afe..eae357fa2 100644 --- a/TODO +++ b/TODO @@ -1,12 +1,4 @@ Bugfixes: -* always open() tty6 to keep X from using it; move debug shell to tty6 - -* there is nothing to warn about here :) - $ systemctl stop systemd-udevd.service systemd-udevd-kernel.socket systemd-udevd-control.socket - Warning: Stopping systemd-udevd.service, but it can still be activated by: - systemd-udevd-control.socket - systemd-udevd-kernel.socket - * check systemd-tmpfiles for selinux context hookup for mknod(), symlink() and similar * swap units that are activated by one name but shown in the kernel under another are semi-broken @@ -21,76 +13,124 @@ Bugfixes: * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point. -* we pull src/core/manager.h into src/shared/src/shared/path-lookup.c which is the wrong direction - rename enum "ManagerRunningAs" to "SystemdRunningAs" and move it to shared/ +F18: -* crash happens when running a service as forking and then changing it to simple and reloading. +* Retest multi-seat - Jul 09 18:20:57 mop systemd[1]: usbmuxd.service operation timed out. Terminating. - Jul 09 18:20:57 mop systemd[1]: Unit usbmuxd.service entered failed state. - Jul 09 18:22:24 mop systemd[1]: PID 21814 read from file /var/run/usbmuxd.pid does not exist. - Jul 09 18:22:24 mop systemd[1]: Unit usbmuxd.service entered failed state. - Jul 09 18:22:33 mop systemd[1]: Reloading. - Jul 09 18:22:37 mop systemd[1]: Assertion 's->type == SERVICE_FORKING' failed at src/core/service.c:3007, function service_sigchld_eve...Aborting. - Jul 09 18:22:37 mop systemd[1]: Caught , dumped core as pid 21865. - Jul 09 18:22:37 mop systemd[1]: Freezing execution. - Jul 09 18:22:37 mop [21866]: Process 21865 (systemd) dumped core. +Features: -* support *static* (/run) hibernate inhibitors. All rpm -i actions should completely prevent any - sort of hibernate action until the next reboot. If the kernel or any other base tool is replaced - by rpm, the resume path might fail, the for resume needed kernel might even be uninstalled, and - the whole situation leads directly to data loss. +* journal: When interleaving journal files cache binary search results so that + we can cheaply continue where we left off. -F18: +* journal: When used on NFS, check payload hashes -* journald: warn if we drop messages we forward to the syslog socket +* journal: When used on NFS make sure wake up sd_journal_wait() every 2s, to handle missing inotify -* refuse automount triggers when automount is queued for stop, much like we refuse socket triggers when sockets are queued for stop +* "See journal for details" messages should include "journalctl -u" command line -* perfomance messages for selinux are gone from debug log? +* document that people can use file system ACLs to manage access to journal files, with example -* There's something wrong with escaping unit names: http://lists.freedesktop.org/archives/systemd-devel/2012-August/006292.html +* don't show cgroup in "systemctl status" if empty/non-existant, especially for foreign .mount units -* wiki: document new logind LockSessions() call +* timedated: export boolean that clarifies whether NTP is even available -* wiki: document new Finished signal params +* timedated: refuse time changes when NTP is on -* logind: different policy actions for idle, suspend, shutdown blockers: allow idle blockers by default, don't allow suspend blockers by default +* clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed -* selinux: merge systemd selinux access controls (dwalsh) +* introduce hybrid-sleep.target -* make logind reserve tty6 or so for text logins, so that gdm never picks it up +* _SOURCE_MONOTONIC_TIMESTAMP entries from the kernel seem to be off by 1000000 -Features: +* document unit_name_mangle() + +* add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible + +* introduce generic AUGMENT_PID=, AUGMENT_DEVICE= fields + +* deal with sendmail/postfix exclusivity + +* systemctl enable: improve the success messages (i.e. more human readable, less shell-like) + +* systemctl enable: fail if target to alias into doesn't exist? maybe show how many units are enabled afterwards? + +* on shutdown: move utmp, wall, audit logic all into PID 1 itself, get rid of systemd-update-utmp-runlevel + +* add "provisioning" instructions to setup an empty /etc + /var + - used to setup a new container from a shared /usr + - superset of tmpfiles model + - instructions shipped by packages and stored in /usr/lib/ + - compose /etc/passwd and /etc/group, copy files + - able to create uid + gid used by packages, for file ownership + +* make repeated alt-ctrl-del presses printing a dump, or even force a reboot without + waiting for the timeout + +* high level net_prio setting in execution context + +* Introduce journalctl -b to show journal messages of a previous boot + +* hostnamed: before returning information from /etc/machine-info.conf check the modification data and reread. Similar for localed, ... + +* currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab isn't + +* WorkingDirectory: support env var replacements like in ExecStart= so that people can use $HOME + +* refuse boot if /etc/machine-id is not useful + +* nspawn: consider changing users for -u with su, so that NSS resolving works correctly + +* nspawn: implement personality changes a la linux32(8) + +* nspawn: reset all aux groups + +* cryptsetup-generator: warn if the password files are world-readable + +* cryptsetup-generator: add RequiresMountsFor= to cryptseup service files referencing a file, similar for devices + +* cryptsetup-generator: allow specification of passwords in crypttab itself + +* document that deps in [Unit] sections ignore Alias= fileds in + [Install] units of other units, unless those units are disabled + +* need to update LGPL2.1 text to newest version (with updated FSF address) + +* allow users from "wheel" to start/stop services + +* systemctl: when powering down/suspending check for inhibitors, and warn. + +* instantiated [Install] for target units + https://bugs.freedesktop.org/show_bug.cgi?id=54377 + +* support *static* (/run) hibernate inhibitors. All rpm -i actions should completely prevent any + sort of hibernate action until the next reboot. If the kernel or any other base tool is replaced + by rpm, the resume path might fail, the for resume needed kernel might even be uninstalled, and + the whole situation leads directly to data loss. + +* move debug shell to tty6 and make sure this doesn't break the gettys on tty6 + +* move cryptsetup key caching into kernel keyctl? + https://bugs.freedesktop.org/show_bug.cgi?id=54982 * make nspawn work without terminal * hw watchdog: optionally try to use the preset watchdog timeout instead of always overriding it + https://bugs.freedesktop.org/show_bug.cgi?id=54712 * after deserializing sockets in socket.c we should reapply sockopts and things -* does vasprintf advance the struct vaargs? http://pastie.org/pastes/4712773/text - -* do shutdown audit/utmp msgs inside of PID 1, get rid of systemd-update-utmp-runlevel - * make timer units go away after they elapsed * http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html -* don't use writev() in tmpfiles for sake of compat with sysfs? - * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead * journald: add kernel cmdline option to disable ratelimiting for debug purposes -* Add a way to reference the machine/boot ID from ExecStart= and similar command lines - * move PID 1 segfaults to /var/lib/systemd/coredump? * Document word splitting syntax for ExecStart= and friends -* when writing journal entries order field items by their address to improve speed on rotating media - * create /sbin/init symlinks from the build system * Query Paul Moore about relabelling socket fds while they are open @@ -99,8 +139,6 @@ Features: * journald: check whether it is OK if the client can still modify delivered journal entries -* json: use jensson - * json: properly serialize multiple fields with the same name per entry * journal live copy, based on libneon (client) and libmicrohttpd @@ -109,10 +147,6 @@ Features: * system-wide seccomp filter -* securityfs: don't mount in container - -* slave/shared remount root fs in container might clash with CAP_SYS_MOUNTS - * ability to pass fds into systemd * system.conf should have controls for cgroups @@ -121,20 +155,14 @@ Features: * currently system services appear not to generate core dumps... -* wall messages for shutdown should move to logind - * allow writing multiple conditions in unit files on one line * cleanup ellipsation for log output in journalctl and systemctl status: have a sane way to disable ellipsation, and disable it by default when invoked in less/more -* enforce limits on fds openened by socket units - * explore multiple service instances per listening socket idea * testing tool for socket activation: some binary that listens on a socket and passes it on using the usual socket activation protocol to some server. -* maybe make systemd-detect-virt suid? or use fscaps? - * shutdown: don't read-only mount anything when running in container * nspawn: --read-only is not applied recursively to submounts @@ -168,10 +196,6 @@ Features: * initialize the hostname from the fs label of /, if /etc/hostname does not exist? -* install README to /etc/rc.d/init.d (if support for that is enabled) helping people who use "ls" there to figure out which services exist. - -* logind: ignore inactive login screens when checking whether power key should be handled - * rename "userspace" to "core-os" * systemctl: "Journal has been rotated since unit was started." message is misleading @@ -204,8 +228,6 @@ Features: * introduce Type=pid-file -* systemctl list-unit-files appears to be broken for symlinked units in /usr/lib - * maybe allow services with ExecStop= set, but no ExecStart=? * efi: implement /forcefsck as uefi variables thus not requiring file system altering to trigger a file system check @@ -233,10 +255,6 @@ Features: * support rd.luks.allow-discards= kernel cmdline params in cryptsetup generator -* systemctl: when stopping a service which has triggres and warning about it actually check the TriggeredBy= deps fields - -* journal: hook up with EFI firmware log - * nspawn: make use of device cgroup contrller by default * drop accountsservice's StandardOutput=syslog and Type=dbus fields @@ -333,8 +351,6 @@ Features: * default unix qlen is too small (10). bump sysctl? add sockopt? -* Possibly, detect whether SysV init scripts can do reloading by looking for "echo Usage:" lines - * figure out whether we should leave dbus around during shutdown * dbus: in fedora, make the machine a symlink to /etc/machine-id @@ -345,8 +361,6 @@ Features: * logind: add equivalent to sd_pid_get_owner_uid() to the D-Bus API -* journal: API for looking for retrieving "all values of this field" - * journal: deal nicely with byte-by-byte copied files, especially regards header * journal: local deserializer of export mode, http server @@ -355,10 +369,6 @@ Features: * document the exit codes when services fail before they are exec()ed -* systemctl journal command - -* journalctl: --cursor support - * save coredump in Windows/Mozilla minidump format * support crash reporting operation modes (https://live.gnome.org/GnomeOS/Design/Whiteboards/ProblemReporting) @@ -397,8 +407,6 @@ Features: * move PAM code into its own binary -* warn if the user stops a service but not its associated socket - * logind: spawn user@..service on login * logind: non-local X11 server handling @@ -441,13 +449,12 @@ Features: - how to pass throw-away units to systemd, or dynamically change properties of existing units - how to integrate cgconfig and suchlike with systemd - resource control in systemd - - inhibiting - testing with Harald's awesome test kit - - restart + - auto-restart * allow port=0 in .socket units -* move readahead files into /var, look for them with .path units +* move readahead files into /var (look for them with .path units?) * teach dbus to activate all services it finds in /etc/systemd/services/org-*.service @@ -469,20 +476,13 @@ Features: * timer events with system resume -* timer events on calendar time: - maybe use this time syntax? http://ohse.de/uwe/uschedule/uschedule.html - * dot output for --test showing the 'initial transaction' -* calendar time support in timer, iCalendar semantics for the timer stuff (RFC2445) - http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=99ee5315dac6211e972fa3f23bcc9a0343ff58c4 - -* implicitly import "defaults" settings file into all types +* calendar time support in timer: + https://docs.google.com/document/pub?id=1bAMyFAjWLpzR3GTDYdgj5FWRMxoZiWw5zmUHEtvdHKA * writable cgroups dbus properties for live changes -* read config fragments for all units from /lib/systemd/system/foobar.service.d/ to override/extend specific settings - * port over to LISTEN_FDS/LISTEN_PID: - rpcbind (/var/run/rpcbind.sock!) HAVEPATCH - cups HAVEPATCH @@ -542,8 +542,10 @@ Regularly: * set_put(), hashmap_put() return values check. i.e. == 0 doesn't free()! -* use __secure_getenv() instead of getenv() where appropriate +* use secure_getenv() instead of getenv() where appropriate + +Scheduled for removal or fixing: -Scheduled for removal (or fixing): +* xxxOverridable dependencies (probably: fix) -* xxxOverridable dependencies +* support for early-boot SysV services (definitely: remove)