networkd: update manager_save() to use fflush_and_check() to simplify things a bit

[elogind.git] / TODO

diff --git a/TODO b/TODO
index 12d144af2fce23577261ca467687bd572f17a9b1..067691f4d53e20f5cf711c05c8d999c04e63fd55 100644 (file)
--- a/TODO
+++ b/TODO
@@ -1,10 +1,8 @@
 Bugfixes:
 Bugfixes:

-* Should systemctl status \* work on all unit types, not just .service?
+* Re-enable "fsck -l" when it is ready:
+   https://bugs.freedesktop.org/show_bug.cgi?id=79576#c5

 
 

-* enabling an instance unit creates a pointless link, and
-  the unit will be started with getty@getty.service:
-    $ systemctl enable getty@.service
-    ln -s '/usr/lib/systemd/system/getty@.service' '/etc/systemd/system/getty.target.wants/getty@.service'
+* Should systemctl status \* work on all unit types, not just .service?

 
 * Dangling symlinks of .automount unit files in .wants/ directories, set up
   automount points even when the original .automount file did not exist
 
 * Dangling symlinks of .automount unit files in .wants/ directories, set up
   automount points even when the original .automount file did not exist


@@ -12,15 +10,7 @@ Bugfixes:
 
 * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point.
 
 
 * properly handle .mount unit state tracking when two mount points are stacked one on top of another on the exact same mount point.
 

-* shorten the message to sane length:
-
-  Cannot add dependency job for unit display-manager.service, ignoring: Unit display-manager.service failed to load: No such file or directory. See system logs and 'systemctl status display-manager.service' for details.
-
-* sd_bus_unref() is broken regarding self-references and "pseudo thread-safety".
-  See the comment in sd_bus_unref() for more..
-

 External:
 External:

-

 * Fedora: when installing fedora with yum --installroot /var/run is a directory, not a symlink
    https://bugzilla.redhat.com/show_bug.cgi?id=975864
 
 * Fedora: when installing fedora with yum --installroot /var/run is a directory, not a symlink
    https://bugzilla.redhat.com/show_bug.cgi?id=975864
 


@@ -28,52 +18,87 @@ External:
 
 * Fedora: post FPC ticket to move add %tmpfiles_create to the packaging guidelines
 
 
 * 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
+

 Features:
 
 Features:
 

-* tmpfiles: figure out relation between Z and m?
+* libsystemd-network:
+  - I don't think there should be a client side object for dhcp
+    releases (probably not even a directory in /run for it)
+
+* resolved:
+  - DNSSEC
+        - use base64 for key presentation?
+        - add display of private key types (http://tools.ietf.org/html/rfc4034#appendix-A.1.1)?
+        - add nice formatting of DNS timestamps
+  - DNS
+        - search paths
+  - mDNS/DNS-SD
+        - avahi compat
+  - DNS-SD service registration from socket units
+  - edns0
+  - dname
+  - cname on PTR (?)
+
+* 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.
+
+* Allow multiple ExecStart= for all Type= settings, so that we can cover rescue.service nicely
+
+* the resolv.conf tmpfiles line should be covered by ENABLE_NETWORKD...

 
 

-* systemd-notify: fake ucred of msg to PID of parent, if that works, with fallback to our own if it doesn't.
+* sysusers:
+  - also create entries in /etc/shadow, /etc/gshadow
+  - allow setting the home directory of system users
+
+* Add a new verb "systemctl top"
+
+* order OnCalendar timer units after timer-sync.target if DefaultDependencies=no so that we don't trigger them prematurely
+
+* refuse mounting on symlinks
+
+* logind: allow users to kill or lock their own sessions
+
+* add new gpt type for btrfs volumes
+
+* support empty /etc boots nicely:
+  - nspawn/gpt-generator: introduce new gpt partition type for /usr
+  - fstab-generator: support systemd.volatile=yes|no|state on the kernel cmdline, too, similar to nspawn's --volatile=
+  - fstab-generator: add support for usr= in addition to root= on the kernel cmdline
+
+* generator that automatically discovers btrfs subvolumes, identifies their purpose based on some xattr on them.
+
+* support setting empty environment variables with Environment= and EnvironmentFile=

 
 * machined/machinectl: sort IP addresses we return by scope and protocol
 * machined: write NSS module for looking up IP addresses for machines
 
 * timer units: actually add extra delays to timer units with high AccuracySec values, don't start them already when we are awake...
 
 
 * machined/machinectl: sort IP addresses we return by scope and protocol
 * machined: write NSS module for looking up IP addresses for machines
 
 * timer units: actually add extra delays to timer units with high AccuracySec values, don't start them already when we are awake...
 

-* timesyncd:
-  - hookup with networkd: NTP servers from dhcp
+* timesyncd - hookup with networkd: NTP servers from dhcp

 
 * a way for container managers to turn off getty starting via $container_headless= or so...
 
 * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
 
 
 * a way for container managers to turn off getty starting via $container_headless= or so...
 
 * figure out a nice way how we can let the admin know what child/sibling unit causes cgroup membership for a specific unit
 

-* add a kernel command line option to enable the debug shell
-

 * journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
 
 * set NOTIFY_SOCKET also for control processes
 
 * journalctl: add the ability to look for the most recent process of a binary. journalctl /usr/bin/X11 --pid=-1 or so...
 
 * set NOTIFY_SOCKET also for control processes
 

-* drop parsing of chkconfig header lines from service.c
-

 * mount_cgroup_controllers(): symlinks need to get the label applied
 
 * mount_cgroup_controllers(): symlinks need to get the label applied
 

-* For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services they run added to the initial transaction and thus confuse Type=idle.
-
-* Add RPM macros for registering/unregistering binfmt drop-ins
+* For timer units: add some mechanisms so that timer units that trigger immediately on boot do not have the services
+  they run added to the initial transaction and thus confuse Type=idle.

 
 * Add timeout to early-boot, and shut down the system if it is hit. Solves the laptop-in-bag problem and is useful for embedded cases
 
 
 * Add timeout to early-boot, and shut down the system if it is hit. Solves the laptop-in-bag problem and is useful for embedded cases
 

-* ImmutableSystem=yes/no or so to mount /usr, /boot read-only/invisible, and leave /var and /etc writable
-
-* InaccessibleHome=yes/no or so to hide /home and /run/user from a service
-
-* Run most system services with cgroupfs read-only and procfs with a more secure mode
+* 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)

 
 * sd-event: generate a failure of a default event loop is executed out-of-thread
 
 
 * sd-event: generate a failure of a default event loop is executed out-of-thread
 

-* add "M" as recursive version of "m" to tmpfiles, then use it for
-  chowning /run/log/journal (but not /var/log/journal), so that we
-  adjust the perms of journal files created before tmpfiles ran.
-

 * expose "Locked" property on logind sesison objects
 
 * add bus api to query unit file's X fields.
 * expose "Locked" property on logind sesison objects
 
 * add bus api to query unit file's X fields.


@@ -92,10 +117,6 @@ Features:
   do not have to open it to know that it is not interesting for us, for
   the most common operations.
 
   do not have to open it to know that it is not interesting for us, for
   the most common operations.
 

-* support transient mount units
-
-* add an "input" group to udev logic and add all input devices to it
-

 * add generator that pulls in systemd-network from containers when
   CAP_NET_ADMIN is set, more than the loopback device is defined, even
   when it is otherwise off
 * add generator that pulls in systemd-network from containers when
   CAP_NET_ADMIN is set, more than the loopback device is defined, even
   when it is otherwise off


@@ -155,8 +176,6 @@ Features:
 
 * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk?
 
 
 * Automatically configure swap partition to use for hibernation by looking for largest swap partition on the root disk?
 

-* socket-proxyd: Use sd-resolve to resolve the server address
-

 * 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
 
 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs
 * 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
 
 * After coming back from hibernation reset hibernation swap partition using the /dev/snapshot ioctl APIs


@@ -183,17 +202,15 @@ Features:
   ReadOnlyDirectories=... for whitelisting files for a service.
 
 * sd-bus:
   ReadOnlyDirectories=... for whitelisting files for a service.
 
 * sd-bus:

-  - make AddMatch calls on dbus1 transports async

   - when kdbus does not take our message without memfds, try again with memfds
   - systemd-bus-proxyd needs to enforce good old XML policy
   - when kdbus does not take our message without memfds, try again with memfds
   - systemd-bus-proxyd needs to enforce good old XML policy

-  - port exit-on-idle logic to byebye ioctl

   - allow updating attach flags during runtime
   - pid1: peek into activating message when activating a service
   - introduce sd_bus_emit_object_added()/sd_bus_emit_object_removed() that automatically includes the build-in interfaces in the list
   - allow updating attach flags during runtime
   - pid1: peek into activating message when activating a service
   - introduce sd_bus_emit_object_added()/sd_bus_emit_object_removed() that automatically includes the build-in interfaces in the list

-  - constructors for bus messages should probably not be OK with a NULL bus pointer
-  - .busname units should not use get_user_creds()/get_cgroup_creds() but instead do NSS only in temporarily forked off child
+  - port to sd-resolve for connecting to TCP dbus servers

   - see if we can drop more message validation on the sending side
   - add API to clone sd_bus_message objects
   - 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?

   - kdbus: matches against source or destination pids for an "strace -p"-like feel. Problem: The PID info needs to be available in userspace too...
   - longer term: priority inheritance
   - dbus spec updates:
   - kdbus: matches against source or destination pids for an "strace -p"-like feel. Problem: The PID info needs to be available in userspace too...
   - longer term: priority inheritance
   - dbus spec updates:


@@ -204,7 +221,6 @@ Features:
        - path escaping
   - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
   - test bloom filter generation indexes
        - path escaping
   - update systemd.special(7) to mention that dbus.socket is only about the compatibility socket now
   - test bloom filter generation indexes

-  - port to sd-resolve for connecting to TCP dbus servers

 
 * sd-event
   - allow multiple signal handlers per signal?
 
 * sd-event
   - allow multiple signal handlers per signal?


@@ -234,12 +250,6 @@ Features:
   - add 'set -e' to scripts in test/
   - make stuff in test/ work with separate output dir
 
   - add 'set -e' to scripts in test/
   - make stuff in test/ work with separate output dir
 

-* systemctl delete x.snapshot leaves no trace in logs (at least at default level).
-
-* make the coredump collector tool move itself into the user's cgroup
-  so that the coredump is properly written to the user's own journal
-  file.
-

 * seems that when we follow symlinks to units we prefer the symlink
   destination path over /etc and /usr. We should not do that. Instead
   /etc should always override /run+/usr and also any symlink
 * seems that when we follow symlinks to units we prefer the symlink
   destination path over /etc and /usr. We should not do that. Instead
   /etc should always override /run+/usr and also any symlink


@@ -277,8 +287,6 @@ Features:
 
 * use "log level" rather than "log priority" everywhere
 
 
 * use "log level" rather than "log priority" everywhere
 

-* timedate: have global on/off switches for auto-time (NTP), and auto-timezone that connman can subscribe to.
-

 * merge unit_kill_common() and unit_kill_context()
 
 * introduce ExecCondition= in services
 * merge unit_kill_common() and unit_kill_context()
 
 * introduce ExecCondition= in services


@@ -385,7 +393,6 @@ Features:
     about it. Should fix both to print nice actionable messages.
   - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
   - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible
     about it. Should fix both to print nice actionable messages.
   - print nice message from systemctl --failed if there are no entries shown, and hook that into ExecStartPre of rescue.service/emergency.service
   - add new command to systemctl: "systemctl system-reexec" which reexecs as many daemons as virtually possible

-  - systemctl enable: improve the success messages (i.e. more human readable, less shell-like)

   - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
   - systemctl: "Journal has been rotated since unit was started." message is misleading
   - support "systemctl stop foobar@.service" to stop all units matching a certain template
   - systemctl enable: fail if target to alias into does not exist? maybe show how many units are enabled afterwards?
   - systemctl: "Journal has been rotated since unit was started." message is misleading
   - support "systemctl stop foobar@.service" to stop all units matching a certain template


@@ -396,14 +403,9 @@ Features:
 * unit install:
   - "systemctl mask" should find all names by which a unit is accessible
     (i.e. by scanning for symlinks to it) and link them all to /dev/null
 * unit install:
   - "systemctl mask" should find all names by which a unit is accessible
     (i.e. by scanning for symlinks to it) and link them all to /dev/null

-  - "systemctl disable" of a unit instance removes all symlinks, but should
-    only remove the instance symlink (systemctl disable of a template
-    unit however should remove them all).

   - systemctl list-unit-files should list generated files (and probably with a new state "generated" for them, or so)
   - systemctl: maybe add "systemctl add-wants" or so...
 
   - systemctl list-unit-files should list generated files (and probably with a new state "generated" for them, or so)
   - systemctl: maybe add "systemctl add-wants" or so...
 

-* deal with sendmail/postfix exclusivity
-

 * timer units:
   - timer units should get the ability to trigger when:
     o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)
 * timer units:
   - timer units should get the ability to trigger when:
     o CLOCK_REALTIME makes jumps (TFD_TIMER_CANCEL_ON_SET)


@@ -422,13 +424,6 @@ Features:
   mode, it will never touch the RTC if the no reliable time source is active or the
   user did not request anything like it.
 
   mode, it will never touch the RTC if the no reliable time source is active or the
   user did not request anything like it.
 

-* libunwind support for coredump pattern hook, and includes this in
-  the message for coredumps. After all, libunwind is now capable to
-  unwind coredumps since a few weeks ago. This probably requires that
-  we have nice support for multi-line messages on display in
-  logs-show.c. Alternatively: use libelfutil, which seems to be the
-  better supported alternative.
-

 * add libsystemd-password or so to query passwords during boot using the password agent logic
 
 * 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().
 * add libsystemd-password or so to query passwords during boot using the password agent logic
 
 * 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().


@@ -437,19 +432,10 @@ Features:
 * fedup: do not delete initrd on switch-root
 * fedup: generator
 
 * fedup: do not delete initrd on switch-root
 * fedup: generator
 

-* timedated: refuse time changes when NTP is on
-

 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
 
 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
 
 * clean up date formatting and parsing so that all absolute/relative timestamps we format can also be parsed
 
 * on shutdown: move utmp, wall, audit logic all into PID 1 (or logind?), get rid of systemd-update-utmp-runlevel
 

-* add "factory" 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
 
 * make repeated alt-ctrl-del presses printing a dump, or even force a reboot without
   waiting for the timeout
 


@@ -458,9 +444,7 @@ Features:
 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
 
 * nspawn:
 * currently x-systemd.timeout is lost in the initrd, since crypttab is copied into dracut, but fstab is not
 
 * nspawn:

-  - nspawn: --read-only is not applied recursively to submounts

   - bind mount read-only the cgroup tree higher than nspawn
   - bind mount read-only the cgroup tree higher than nspawn

-  - nspawn: make it work for dwalsh and shared /usr containers -- tmpfs mounts as command line parameters

   - 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
 
   - 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
 


@@ -477,25 +461,16 @@ Features:
 
 * make timer units go away after they elapsed
 
 
 * make timer units go away after they elapsed
 

-* http://lists.freedesktop.org/archives/systemd-devel/2012-September/006502.html
-  (network and remote-fs on shutdown)
-

 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
 
 * move PID 1 segfaults to /var/lib/systemd/coredump?
 
 * create /sbin/init symlinks from the build system
 
 * come up with a nice way to write queue/read_ahead_kb for a block device without interfering with readahead
 
 * move PID 1 segfaults to /var/lib/systemd/coredump?
 
 * create /sbin/init symlinks from the build system
 

-* Query Paul Moore about relabelling socket fds while they are open
-

 * allow writing multiple conditions in unit files on one line
 
 * allow writing multiple conditions in unit files on one line
 

-* explore multiple service instances per listening socket idea
-

 * MountFlags=shared acts as MountFlags=slave right now.
 
 * MountFlags=shared acts as MountFlags=slave right now.
 

-* ReadOnlyDirectories= is not applied recursively to submounts
-

 * drop PID 1 reloading, only do reexecing (difficult: Reload()
   currently is properly synchronous, Reexec() is weird, because we
   cannot delay the response properly until we are back, so instead of
 * drop PID 1 reloading, only do reexecing (difficult: Reload()
   currently is properly synchronous, Reexec() is weird, because we
   cannot delay the response properly until we are back, so instead of


@@ -546,8 +521,6 @@ Features:
 
 * introduce mix of BindTo and Requisite
 
 
 * introduce mix of BindTo and Requisite
 

-* add DeleteSocketsOnStop=yes|no option to socket units
-

 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
 
 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/
 * There's currently no way to cancel fsck (used to be possible via C-c or c on the console)
 
 * add option to sockets to avoid activation. Instead just drop packets/connections, see http://cyberelk.net/tim/2012/02/15/portreserve-systemd-solution/


@@ -566,7 +539,6 @@ Features:
 * when a bus name of a service disappears from the bus make sure to queue further activation requests
 
 * tmpfiles:
 * when a bus name of a service disappears from the bus make sure to queue further activation requests
 
 * tmpfiles:

-  - check systemd-tmpfiles for selinux context hookup for mknod(), symlink() and similar

   - apply "x" on "D" too (see patch from William Douglas)
 
 * for services: do not set $HOME in services unless requested
   - apply "x" on "D" too (see patch from William Douglas)
 
 * for services: do not set $HOME in services unless requested


@@ -593,8 +565,6 @@ Features:
   when we start a service in order to avoid confusion when a user
   assumes starting a service is enough to make it accessible
 
   when we start a service in order to avoid confusion when a user
   assumes starting a service is enough to make it accessible
 

-* support User= and Group= attributes for AF_UNIX sockets. (difficult, requires NSS from PID 1?)
-

 * Make it possible to set the keymap independently from the font on
   the kernel cmdline. Right now setting one resets also the other.
 
 * Make it possible to set the keymap independently from the font on
   the kernel cmdline. Right now setting one resets also the other.
 


@@ -607,8 +577,6 @@ Features:
   - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
   - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
 
   - readahead: when bumping /sys readahead variable save mtime and compare later to detect changes
   - readahead: make use of EXT4_IOC_MOVE_EXT, as used by http://e4rat.sourceforge.net/
 

-* add support for /bin/mount -s
-

 * GC unreferenced jobs (such as .device jobs)
 
 * write blog stories about:
 * GC unreferenced jobs (such as .device jobs)
 
 * write blog stories about:


@@ -634,8 +602,6 @@ Features:
 
 * allow port=0 in .socket units
 
 
 * allow port=0 in .socket units
 

-* support systemd.mask= on the kernel command line.
-

 * recreate systemd's D-Bus private socket file on SIGUSR2
 
 * Support --test based on current system state
 * recreate systemd's D-Bus private socket file on SIGUSR2
 
 * Support --test based on current system state


@@ -674,17 +640,13 @@ Features:
 * networkd:
    - add more keys to [Route] and [Address] sections
    - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)
 * networkd:
    - add more keys to [Route] and [Address] sections
    - add support for more DHCPv4 options (and, longer term, other kinds of dynamic config)

-   - send hostname to DHCP server

    - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
    - add reduced [Link] support to .network files
    - add Scope= parsing option for [Network]
    - properly handle routerless dhcp leases
    - add proper initrd support (in particular generate .network/.link files based on /proc/cmdline)
    - add reduced [Link] support to .network files
    - add Scope= parsing option for [Network]
    - properly handle routerless dhcp leases

-   - set lifetime on the address acquired from dhcp
-   - patch kernel to support module alias for tunnel device (ipip/sit/gre),
-     then remove remove libkmod dependency and CAP_SYS_MODULE
-   - add veth netdev support (c.f. http://shorewall.net/bridge-Shorewall-perl.html#veth)
-   - add tun/tap netdev support

    - add more attribute support for SIT tunnel
    - add more attribute support for SIT tunnel

+   - work with non-ethernet devices
+   - add support for more bond options

 
 * networkd-wait-online:
    - make operstates to wait for configurable?
 
 * networkd-wait-online:
    - make operstates to wait for configurable?


@@ -692,7 +654,23 @@ Features:
 * dhcp:
    - figure out how much we can increase Maximum Message Size
    - export timezone information
 * dhcp:
    - figure out how much we can increase Maximum Message Size
    - export timezone information

-   - FORCERENEW
+   - support RFC4702 (pass FQDN)
+
+* dhcp6:
+   - add functions to set previously stored IPv6 addresses on startup and get
+     them at shutdown; store them in client->ia_na
+   - 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 elapsed time option
+   - implement dhcpv6 authentication
+   - investigate the usefulness of Confirm messages; i.e. are there any
+     situations where the link changes without any loss in carrier detection
+     or interface down
+   - some servers don't do rapid commit without a filled in IA_NA, verify
+     this behavior

 
 External:
 
 
 External:
 


@@ -728,6 +706,9 @@ External:
 
 * register catalog database signature as file magic
 
 
 * register catalog database signature as file magic
 

+* zsh shell completion: <command> <verb> -<TAB> should complete options, but currently
+  does not
+

 Regularly:
 
 * look for close() vs. close_nointr() vs. close_nointr_nofail()
 Regularly:
 
 * look for close() vs. close_nointr() vs. close_nointr_nofail()