chiark / gitweb /
elogind.git
9 years agoupdate TODO
Lennart Poettering [Tue, 19 Aug 2014 00:09:14 +0000 (02:09 +0200)]
update TODO

9 years agosysusers: also update /etc/shadow and /etc/gshadow when creating new system users
Lennart Poettering [Mon, 18 Aug 2014 23:57:24 +0000 (01:57 +0200)]
sysusers: also update /etc/shadow and /etc/gshadow when creating new system users

This should resolve problems with tools like "grpck" and suchlike.

9 years agohashmap: try to use the existing 64bit hash functions for dev_t if it is 64bit
Lennart Poettering [Mon, 18 Aug 2014 22:04:46 +0000 (00:04 +0200)]
hashmap: try to use the existing 64bit hash functions for dev_t if it is 64bit

9 years agobus: map sealed memfds as MAP_PRIVATE
David Herrmann [Mon, 18 Aug 2014 21:54:10 +0000 (23:54 +0200)]
bus: map sealed memfds as MAP_PRIVATE

Mapping files as MAP_SHARED is handled by the kernel as 'writable'
mapping. Always! Even with PROT_READ. Reason for that is,
mprotect(PROT_WRITE) could change the mapping underneath and currently
there is no kernel infrastructure to add protection there. This might
change in the future, but until then, map sealed files as MAP_PRIVATE so
we don't get EPERM.

9 years agoUpdate TODO
Lennart Poettering [Mon, 18 Aug 2014 21:16:18 +0000 (23:16 +0200)]
Update TODO

9 years agoutil: try to be a bit more NFS compatible when checking whether an FS is writable
Lennart Poettering [Mon, 18 Aug 2014 21:15:51 +0000 (23:15 +0200)]
util: try to be a bit more NFS compatible when checking whether an FS is writable

https://bugs.freedesktop.org/show_bug.cgi?id=81169

9 years agocore: minor modernizations
Lennart Poettering [Mon, 18 Aug 2014 20:25:24 +0000 (22:25 +0200)]
core: minor modernizations

9 years agounits: fix BindsTo= logic when applied relative to services with Type=oneshot
Lennart Poettering [Mon, 18 Aug 2014 20:21:42 +0000 (22:21 +0200)]
units: fix BindsTo= logic when applied relative to services with Type=oneshot

Start jobs for Type=oneshot units are successful when the unit state
transition activating → inactive took place. In such a case all units
that BindsTo= on it previously would continue to run, even though the unit
they dependet on was actually already gone.

9 years agoman: fix typo
Ronny Chevalier [Mon, 18 Aug 2014 19:00:23 +0000 (21:00 +0200)]
man: fix typo

9 years agobootchart: use NSEC_PER_SEC
Ronny Chevalier [Mon, 18 Aug 2014 18:59:11 +0000 (20:59 +0200)]
bootchart: use NSEC_PER_SEC

9 years agobus-control: Fix cgroup handling
Denis Kenzior [Mon, 18 Aug 2014 18:21:55 +0000 (13:21 -0500)]
bus-control: Fix cgroup handling

On systems without properly setup systemd, cg_get_root_path returns
-ENOENT.  This means that busctl doesn't display much information.

busctl monitor also fails whenever it intercepts messages.

This fix fakes creates a fake "/" root cgroup which lets busctl work
on such systems.

9 years agoman: mention that "units" are commonly system services
Lennart Poettering [Mon, 18 Aug 2014 17:08:03 +0000 (19:08 +0200)]
man: mention that "units" are commonly system services

Also, provide an example for -u.

9 years agonetworkd: fix how we generate lists in link_save()
Lennart Poettering [Mon, 18 Aug 2014 16:59:48 +0000 (18:59 +0200)]
networkd: fix how we generate lists in link_save()

https://bugs.freedesktop.org/show_bug.cgi?id=82721

9 years agotests: add missing entry to test-tables
Ronny Chevalier [Sat, 16 Aug 2014 12:19:12 +0000 (14:19 +0200)]
tests: add missing entry to test-tables

9 years agotests: add tests for time-util.c
Ronny Chevalier [Sat, 16 Aug 2014 12:19:11 +0000 (14:19 +0200)]
tests: add tests for time-util.c

add tests for:
- timezone_is_valid
- get_timezones

9 years agotests: add test-condition-util
Ronny Chevalier [Sat, 16 Aug 2014 12:19:10 +0000 (14:19 +0200)]
tests: add test-condition-util

9 years agotests: add tests for util.c
Ronny Chevalier [Sat, 16 Aug 2014 12:19:09 +0000 (14:19 +0200)]
tests: add tests for util.c

add tests for:
- is_symlink
- pid_is_unwaited
- pid_is_alive
- search_and_fopen
- search_and_fopen_nulstr
- glob_exists
- execute_directory

9 years agotests: add test for fdset_iterate
Ronny Chevalier [Sat, 16 Aug 2014 12:19:08 +0000 (14:19 +0200)]
tests: add test for fdset_iterate

9 years agotests: add tests for fileio.c
Ronny Chevalier [Sat, 16 Aug 2014 12:19:07 +0000 (14:19 +0200)]
tests: add tests for fileio.c

add tests for:
- write_string_stream
- write_string_file
- sendfile_full

9 years agotests: add missing unlink
Ronny Chevalier [Sat, 16 Aug 2014 12:19:06 +0000 (14:19 +0200)]
tests: add missing unlink

9 years agotests: add tests for socket-util.c
Ronny Chevalier [Sat, 16 Aug 2014 12:19:05 +0000 (14:19 +0200)]
tests: add tests for socket-util.c

add tests for:
- socket_address_is
- socket_address_is_netlink
- sockaddr_equal

9 years agoman: fix typo
Ronny Chevalier [Sat, 16 Aug 2014 12:24:27 +0000 (14:24 +0200)]
man: fix typo

9 years agotmpfiles: only execute chmod()/chown() when needed
Michael Olbrich [Sun, 17 Aug 2014 07:45:00 +0000 (09:45 +0200)]
tmpfiles: only execute chmod()/chown() when needed

This avoids errors like this, when the paths are already there with the
correct permissions and owner:

chmod(/var/spool) failed: Read-only file system

9 years agoMerge remote-tracking branch 'origin/master'
Lennart Poettering [Mon, 18 Aug 2014 16:12:55 +0000 (18:12 +0200)]
Merge remote-tracking branch 'origin/master'

9 years agonetworkd: warn when ignoring unsupported tuntap options
Tom Gundersen [Mon, 18 Aug 2014 10:29:45 +0000 (12:29 +0200)]
networkd: warn when ignoring unsupported tuntap options

The interface for creating tuntap devices should be ported to rtnl so it would support the same settings
as other kinds. In the meantime, the best one can do is to drop in a .link file to set the desired options.

9 years agocore: Verify systemd1 DBus method callers via polkit
Stef Walter [Wed, 6 Aug 2014 09:45:36 +0000 (11:45 +0200)]
core: Verify systemd1 DBus method callers via polkit

DBus methods that retrieve information can be called by anyone.

DBus methods that modify state of units are verified via polkit
action: org.freedesktop.systemd1.manage-units

DBus methods that modify state of unit files are verified via polkit
action: org.freedesktop.systemd1.manage-unit-files

DBus methods that reload the entire daemon state are verified via polkit
action: org.freedesktop.systemd1.reload-daemon

DBus methods that modify job state are callable from the clients
that started the job.

root (ie: CAP_SYS_ADMIN) can continue to perform all calls, property
access etc. There are several DBus methods that can only be
called by root.

Open up the dbus1 policy for the above methods.

(Heavily modified by Lennart, making use of the new
bus_verify_polkit_async() version that doesn't force us to always
pass the original callback around. Also, interactive auhentication must
be opt-in, not unconditional, hence I turned this off.)

9 years agobus-util: simplify bus_verify_polkit_async() a bit
Lennart Poettering [Mon, 18 Aug 2014 15:44:17 +0000 (17:44 +0200)]
bus-util: simplify bus_verify_polkit_async() a bit

First, let's drop the "bus" argument, we can determine it from the
message anyway.

Secondly, determine the right callback/userdata pair automatically from
what is currently is being dispatched. This should simplify things a lot
for us, since it makes it unnecessary to pass pointers through the
original handlers through all functions when we process messages, which
might require authentication.

9 years agosd-bus: add API to query which handler/callback is currently being dispatched
Lennart Poettering [Mon, 18 Aug 2014 15:41:56 +0000 (17:41 +0200)]
sd-bus: add API to query which handler/callback is currently being dispatched

9 years agomemfd: internalize functions, drop sd_memfd type
Daniel Mack [Mon, 18 Aug 2014 11:28:43 +0000 (13:28 +0200)]
memfd: internalize functions, drop sd_memfd type

Remove the sd_ prefix from internal functions and get rid of the sd_memfd
type. As a memfd is now just a native file descriptor, we can get rid of our
own wrapper type, and also use close() and dup() on them directly.

9 years agomemfd: use _cleanup_ if applicable
David Herrmann [Mon, 18 Aug 2014 11:07:43 +0000 (13:07 +0200)]
memfd: use _cleanup_ if applicable

We now have a sd_memfd_freep helper, use it if applicable.

9 years agomemfd: map unsealed files as MAP_SHARED
David Herrmann [Mon, 18 Aug 2014 11:05:48 +0000 (13:05 +0200)]
memfd: map unsealed files as MAP_SHARED

We need to map sealed files as MAP_PRIVATE so far as the kernel treats
MAP_SHARED as writable mapping (you can run mprotect(PROT_WRITE) at any
time on those). However, unsealed files must be mapped as MAP_SHARED.
Otherwise, we never end up writing to the real file.

9 years agomemfd: disallow importing memfds without sealing
David Herrmann [Mon, 18 Aug 2014 11:03:09 +0000 (13:03 +0200)]
memfd: disallow importing memfds without sealing

We use memfds for sealing. Lets not bother with memfds created without
MFD_ALLOW_SEALING for now. They're equivalent to random shmem files, so
don't bother treating them as sealable memfds.

9 years agomemfd: don't open kdbus for memfd
David Herrmann [Mon, 18 Aug 2014 10:57:55 +0000 (12:57 +0200)]
memfd: don't open kdbus for memfd

No reason to open /dev/kdbus/control if we want memfds. memfd_create() is
always available.

9 years agomemfd: internalize header
David Herrmann [Mon, 18 Aug 2014 10:57:03 +0000 (12:57 +0200)]
memfd: internalize header

Fix the memfd.h header to use handy features like #pragma, cleanup-funcs
and util.h. Also drop the EXTERN-C macros.

9 years agomemfd: fix memfd_create() syscall wrapper
Daniel Mack [Mon, 18 Aug 2014 10:24:04 +0000 (12:24 +0200)]
memfd: fix memfd_create() syscall wrapper

Unlike earlier versions, the syscall only takes 2 arguments in its
final version, not 3.

9 years agomemfd: move code from public library to src/shared
Daniel Mack [Mon, 18 Aug 2014 08:55:49 +0000 (10:55 +0200)]
memfd: move code from public library to src/shared

Don't expose generic kernel API via libsystemd, but keep the code internal
for our own usage.

9 years agoMakefile.am: test-bus-memfd went away. Kill its residues in Makefile.am
Daniel Mack [Mon, 18 Aug 2014 08:45:49 +0000 (10:45 +0200)]
Makefile.am: test-bus-memfd went away. Kill its residues in Makefile.am

9 years agokdbus: switch over to generic memfd implementation (ABI+API break)
Daniel Mack [Mon, 12 May 2014 23:28:09 +0000 (01:28 +0200)]
kdbus: switch over to generic memfd implementation (ABI+API break)

9 years agonetworkctl: use safe_qsort in case no links are present
Thomas Hindoe Paaboel Andersen [Sat, 16 Aug 2014 21:18:32 +0000 (23:18 +0200)]
networkctl: use safe_qsort in case no links are present

Unlikely to happen but still...

9 years agosd-bus,log: remove unused variables
Thomas Hindoe Paaboel Andersen [Fri, 15 Aug 2014 22:31:23 +0000 (00:31 +0200)]
sd-bus,log: remove unused variables

9 years agosd-event: return 'r' rather than '-errno'
Tom Gundersen [Fri, 15 Aug 2014 17:15:30 +0000 (19:15 +0200)]
sd-event: return 'r' rather than '-errno'

9 years agoMerge commit 'b39a2770ba55637da80e2e389222c59dbea73507'
Lennart Poettering [Fri, 15 Aug 2014 18:25:10 +0000 (20:25 +0200)]
Merge commit 'b39a2770ba55637da80e2e389222c59dbea73507'

9 years agosd-event: fix missing needs_rearm
Tom Gundersen [Fri, 15 Aug 2014 16:21:18 +0000 (18:21 +0200)]
sd-event: fix missing needs_rearm

9 years agosd-bus: add API to check if a client has privileges
Lennart Poettering [Fri, 15 Aug 2014 18:08:51 +0000 (20:08 +0200)]
sd-bus: add API to check if a client has privileges

This is a generalization of the vtable privilege check we already have,
but exported, and hence useful when preparing for a polkit change.

This will deal with the complexity that on dbus1 one cannot trust the
capability field we retrieve via the bus, since it is read via
/proc/$$/stat (and thus might be out-of-date) rather than directly from
the message (like on kdbus) or bus connection (as for uid creds on
dbus1).

Also, port over all code to this new API.

9 years agoupdate TODO
Lennart Poettering [Fri, 15 Aug 2014 16:29:21 +0000 (18:29 +0200)]
update TODO

9 years agocgroup: only generate warnings if actually writing to cgroup attributes failed
Lennart Poettering [Fri, 15 Aug 2014 16:14:37 +0000 (18:14 +0200)]
cgroup: only generate warnings if actually writing to cgroup attributes failed

9 years agomain,log: parse the log related kernel command line parameters at one place only...
Lennart Poettering [Fri, 15 Aug 2014 16:07:36 +0000 (18:07 +0200)]
main,log: parse the log related kernel command line parameters at one place only, and for all tools

Previously, we ended up parsing some of them three times: in main.c when
processing the kernel cmdline, in main.c when processing the process
cmdline (only for containers), and in log.c again.

Let's streamline this, and only parse them in log.c

In PID 1 also make sure we parse "quiet" first, and then override this
with the more specific checks in log.c

9 years agomain: minor code modernization for initializing the console
Lennart Poettering [Fri, 15 Aug 2014 16:01:52 +0000 (18:01 +0200)]
main: minor code modernization for initializing the console

9 years agoupdate TODO
Lennart Poettering [Fri, 15 Aug 2014 16:01:30 +0000 (18:01 +0200)]
update TODO

9 years agohostnamectl: actually implement location support
Lennart Poettering [Fri, 15 Aug 2014 15:06:10 +0000 (17:06 +0200)]
hostnamectl: actually implement location support

9 years agoresolve: fix compilation on LLVM+clang
David Herrmann [Fri, 15 Aug 2014 14:58:16 +0000 (16:58 +0200)]
resolve: fix compilation on LLVM+clang

LLVM+clang does not allow statement-expressions inside of
type-declarations (file-scope). Use CONST_MAX() to avoid this.

9 years agomacro: add CONST_MAX() macro
David Herrmann [Fri, 15 Aug 2014 14:54:52 +0000 (16:54 +0200)]
macro: add CONST_MAX() macro

The CONST_MAX() macro is similar to MAX(), but verifies that both
arguments have the same type and are constant expressions. Furthermore,
the result of CONST_MAX() is again a constant-expression.

CONST_MAX() avoids any statement-expressions and other non-trivial
expression-types. This avoids rather arbitrary restrictions in both GCC
and LLVM, which both either fail with statement-expressions inside
type-declarations or statement-expressions inside static-const
initializations.

If anybody knows how to circumvent this, please feel free to unify
CONST_MAX() and MAX().

9 years agomacro: const'ify MIN/MAX/... macros
David Herrmann [Fri, 15 Aug 2014 14:16:30 +0000 (16:16 +0200)]
macro: const'ify MIN/MAX/... macros

We must add 'const' to local variables in statement-expressions to
guarantee that the macros can produce constant-expressions if given such.
GCC seems to ignore this, but LLVM/clang requires it (understandably).

9 years agoresolved: fix assertion when joining llmnr mcast group
Lennart Poettering [Fri, 15 Aug 2014 14:42:06 +0000 (16:42 +0200)]
resolved: fix assertion when joining llmnr mcast group

9 years agonetworkd: print nice warnings if people configure invalid domain names
Lennart Poettering [Fri, 15 Aug 2014 14:21:19 +0000 (16:21 +0200)]
networkd: print nice warnings if people configure invalid domain names

9 years agoutil: make is_localhost() check for 'localdomain' too, so that we can use it for...
Lennart Poettering [Fri, 15 Aug 2014 14:04:46 +0000 (16:04 +0200)]
util: make is_localhost() check for 'localdomain' too, so that we can use it for both validating domains and host names

9 years agonetworkd: fix minor memory leak
Lennart Poettering [Fri, 15 Aug 2014 14:02:42 +0000 (16:02 +0200)]
networkd: fix minor memory leak

9 years agonetworkctl: show acquired system domains
Lennart Poettering [Fri, 15 Aug 2014 14:02:29 +0000 (16:02 +0200)]
networkctl: show acquired system domains

9 years agosd-network: add system-wide sd_network_get_domains() API
Lennart Poettering [Fri, 15 Aug 2014 14:02:14 +0000 (16:02 +0200)]
sd-network: add system-wide sd_network_get_domains() API

9 years agonetworkd: always write out locally configured settings first, dhcp-acquired ones...
Lennart Poettering [Fri, 15 Aug 2014 13:42:56 +0000 (15:42 +0200)]
networkd: always write out locally configured settings first, dhcp-acquired ones later

This is primarily important for the domains list, as we really should
prefer the locally configured domain over the dhcp supplied ones when we
use it as a search list.

9 years agonetworkd: remove "*" from domains list
Lennart Poettering [Fri, 15 Aug 2014 13:36:17 +0000 (15:36 +0200)]
networkd: remove "*" from domains list

Also, simplify things a bit and make sure we don't forget looking at one
of the entries.

9 years agonetworkctl: two OOM fixes
Lennart Poettering [Fri, 15 Aug 2014 13:34:56 +0000 (15:34 +0200)]
networkctl: two OOM fixes

9 years agosd-nework: be more careful with error codes, return ENODATA if you lack information
Lennart Poettering [Fri, 15 Aug 2014 13:34:33 +0000 (15:34 +0200)]
sd-nework: be more careful with error codes, return ENODATA if you lack information

9 years agosd-network: add support for wildcard domains
Tom Gundersen [Fri, 15 Aug 2014 12:49:31 +0000 (14:49 +0200)]
sd-network: add support for wildcard domains

9 years agonetworkd: add support for Domains= to .network files
Tom Gundersen [Fri, 15 Aug 2014 12:21:08 +0000 (14:21 +0200)]
networkd: add support for Domains= to .network files

This allows the search/routing domanis to be specified per link/network and be passed
on to resolved.

9 years agokernel-install/90-loaderentry.install: fixed cmdline parsing
Harald Hoyer [Fri, 15 Aug 2014 12:39:05 +0000 (14:39 +0200)]
kernel-install/90-loaderentry.install: fixed cmdline parsing

If /etc/kernel/cmdline is missing or empty, we read /proc/cmdline and
want to filter out the initrd line. Due to a bug, the whole contents was
filtered out.

9 years agocore: Rename Job.subscribed field to Job.clients
Stef Walter [Wed, 6 Aug 2014 09:53:00 +0000 (11:53 +0200)]
core: Rename Job.subscribed field to Job.clients

This reflects how this field will be used, to not only track where
to send signals, but also which callers (other than root) are allowed
to call DBus methods on the Job.

9 years agocore: Common code for DBus methods that Cancel a job
Stef Walter [Wed, 6 Aug 2014 09:34:40 +0000 (11:34 +0200)]
core: Common code for DBus methods that Cancel a job

Both ofs.Job.Cancel() and ofs.Manager.CancelJob() now use same
implementation. So we can add caller verify logic appropriately.

9 years agosd-bus: Remove bus arg from bus_verify_polkit_async_registry_free()
Stef Walter [Wed, 6 Aug 2014 05:57:43 +0000 (07:57 +0200)]
sd-bus: Remove bus arg from bus_verify_polkit_async_registry_free()

It's unneccessary, not used, and complicates callers of the
function.

9 years agotest: fix strtod test for real
David Herrmann [Fri, 15 Aug 2014 11:20:18 +0000 (13:20 +0200)]
test: fix strtod test for real

The "0,5" syntax was actually right. The real problem is, the test should
only run if the local system has the de_DE.UTF-8 locale. Therefore, skip
the tests if setlocale() fails. This is kinda ugly, as it is done
silently, but we cannot skip partial tests with the current
infrastructure. Should be fine this way.

9 years agoutil: never use ether_ntoa(), since it formats with %x, not %02x, which makes etherne...
Lennart Poettering [Fri, 15 Aug 2014 11:18:50 +0000 (13:18 +0200)]
util: never use ether_ntoa(), since it formats with %x, not %02x, which makes ethernet addresses look funny

9 years agounit: remove spurious newline
Lennart Poettering [Fri, 15 Aug 2014 10:57:46 +0000 (12:57 +0200)]
unit: remove spurious newline

9 years agonetworkctl: increase column width for link type to 18, to accomodate for 'ieee80211_r...
Lennart Poettering [Fri, 15 Aug 2014 10:57:33 +0000 (12:57 +0200)]
networkctl: increase column width for link type to 18, to accomodate for 'ieee80211_radiotap'

9 years agonetworkd: rename UseDomainName to UseDomains
Tom Gundersen [Fri, 15 Aug 2014 10:45:59 +0000 (12:45 +0200)]
networkd: rename UseDomainName to UseDomains

This option will also apply to the search domains, so make it plural.

9 years agosysctl: always write net.ipv4.conf.all.xyz= in addition to net.ipv4.conf.default...
Lennart Poettering [Fri, 15 Aug 2014 10:07:33 +0000 (12:07 +0200)]
sysctl: always write net.ipv4.conf.all.xyz= in addition to net.ipv4.conf.default.xyz=

Otherwise we have a boot-time race, where interfaces that popped up
after the sysctl service would get the settings applied, but all others
wouldn't.

9 years agocgroup: downgrade log messages about non-existant cgroup attributes to LOG_DEBUG
Lennart Poettering [Fri, 15 Aug 2014 09:56:36 +0000 (11:56 +0200)]
cgroup: downgrade log messages about non-existant cgroup attributes to LOG_DEBUG

9 years agocgroup: never try to create files in cgroupfs, only open them for writing
Lennart Poettering [Fri, 15 Aug 2014 09:55:43 +0000 (11:55 +0200)]
cgroup: never try to create files in cgroupfs, only open them for writing

This should have the benefit that cg_set_attribute() returns ENOENT
instead of EACCESS when we use it for non-existing attributes.

9 years agotest: fix strtod() test
David Herrmann [Fri, 15 Aug 2014 08:44:43 +0000 (10:44 +0200)]
test: fix strtod() test

One strtod() test is broken since:

commit 8e211000025940b770794abf5754de61b4add0af
Author: Thomas Hindoe Paaboel Andersen <phomes@gmail.com>
Date:   Mon Aug 4 23:13:31 2014 +0200

    test: use fabs on doubles

The commit was right, so no reason to revert it, but the test was broken
before and only worked by coincidence. Convert "0,5" to "0.5" so we don't
depend on locales for double conversion (or well, we depend on "C" which
seems reasonable).

9 years agonetworkctl: print local domain name in status output
Lennart Poettering [Fri, 15 Aug 2014 01:06:11 +0000 (03:06 +0200)]
networkctl: print local domain name in status output

9 years agodhcp: the localhost isn't valid as hostname either
Lennart Poettering [Fri, 15 Aug 2014 01:05:44 +0000 (03:05 +0200)]
dhcp: the localhost isn't valid as hostname either

9 years agocoredump: display libdw fail string on stack trace fail
Umut Tezduyar Lindskog [Tue, 8 Jul 2014 14:16:14 +0000 (16:16 +0200)]
coredump: display libdw fail string on stack trace fail

- systemd[1]: hello.service: main process exited, code= dumped, status=3/QUIT
- systemd-coredump[2541]: Failed to generate stack trace: Unwinding not supported for this architecture
- systemd-coredump[2541]: Process 1024 (hello) of user 154 dumped core.

9 years agoman: drop references to removed and obsolete 'systemctl load' command
Lennart Poettering [Fri, 15 Aug 2014 00:41:14 +0000 (02:41 +0200)]
man: drop references to removed and obsolete 'systemctl load' command

9 years agosocket: add bus property for bus property NoDelay
Susant Sahani [Thu, 14 Aug 2014 17:36:13 +0000 (23:06 +0530)]
socket: add bus property for bus property NoDelay

Missed to add the SD_BUS_PROPERTY for no_delay.

9 years agosocket: Add support for TCP defer accept
Susant Sahani [Thu, 14 Aug 2014 17:36:12 +0000 (23:06 +0530)]
socket: Add support for TCP defer accept

TCP_DEFER_ACCEPT Allow a listener to be awakened only when data
arrives on the socket. If TCP_DEFER_ACCEPT set on a server-side
listening socket, the TCP/IP stack will not to wait for the final
ACK packet and not to initiate the process until the first packet
of real data has arrived. After sending the SYN/ACK, the server will
then wait for a data packet from a client. Now, only three packets
will be sent over the network, and the connection establishment delay
will be significantly reduced.

9 years agosocket: Add Support for TCP keep alive variables
Susant Sahani [Thu, 14 Aug 2014 17:36:11 +0000 (23:06 +0530)]
socket: Add Support for TCP keep alive variables

The tcp keep alive variables now can be configured via conf
parameter. Follwing variables are now supported by this patch.

tcp_keepalive_intvl: The number of seconds between TCP keep-alive probes

tcp_keepalive_probes: The maximum number of TCP keep-alive probes to
send before giving up and killing the connection if no response is
obtained from the other end.

tcp_keepalive_time: The number of seconds a connection needs to be
idle before TCP begins sending out keep-alive probes.

9 years agoupdate TODO
Lennart Poettering [Thu, 14 Aug 2014 23:25:36 +0000 (01:25 +0200)]
update TODO

9 years agocore: Refuse mount on symlink
Timofey Titovets [Thu, 14 Aug 2014 23:05:47 +0000 (02:05 +0300)]
core: Refuse mount on symlink

9 years agonetworkd: don't respect domainname from DHCP by default
Tom Gundersen [Thu, 14 Aug 2014 23:08:46 +0000 (01:08 +0200)]
networkd: don't respect domainname from DHCP by default

Most routers will send garbage, so make this opt-in only.

9 years agosd-dhcp-lease: verify hostname and domainnames we receive
Tom Gundersen [Thu, 14 Aug 2014 23:04:53 +0000 (01:04 +0200)]
sd-dhcp-lease: verify hostname and domainnames we receive

9 years agoresolved: pull in domain names from sd-network
Tom Gundersen [Thu, 14 Aug 2014 20:18:22 +0000 (22:18 +0200)]
resolved: pull in domain names from sd-network

9 years agosd-network: add sd_network_linkg_get_domains()
Tom Gundersen [Thu, 14 Aug 2014 18:35:37 +0000 (20:35 +0200)]
sd-network: add sd_network_linkg_get_domains()

For now this only exposes the domain name (DHCP Option 15), and not
the search string (DHCP Option 119), which will be implemented in
a follow-up patch.

9 years agomachine_kill(): Don't kill the unit when killing the leader
Eelco Dolstra [Thu, 14 Aug 2014 17:59:16 +0000 (19:59 +0200)]
machine_kill(): Don't kill the unit when killing the leader

If "machinectl poweroff" or "machinectl reboot" is used on a
systemd-nspawn container started with --keep-unit and --register, it
should *only* send the appropriate signal to the leader PID (i.e. the
container's systemd process). It shouldn't fall through to
manager_kill_unit() to also send the signal to the unit. The latter
ends up killing systemd-nspawn, which takes down the container
prematurely.

9 years agoresolved: clarify that LLMNR scopes must have a link assigned
Lennart Poettering [Thu, 14 Aug 2014 17:56:22 +0000 (19:56 +0200)]
resolved: clarify that LLMNR scopes must have a link assigned

This is supposed to remove some compiler warnings:

http://lists.freedesktop.org/archives/systemd-devel/2014-July/021393.html

9 years agoupdate TODO
Lennart Poettering [Thu, 14 Aug 2014 17:02:48 +0000 (19:02 +0200)]
update TODO

9 years agocore: do not add default dependencies to /usr mount unit
Jon Severinsson [Mon, 21 Jul 2014 22:39:13 +0000 (00:39 +0200)]
core: do not add default dependencies to /usr mount unit

This makes no difference if /usr was mounted in the initrd,
and brings the behaviour of legacy systems closer to those
with a propper initrd.

9 years agoupdate TODO
Lennart Poettering [Thu, 14 Aug 2014 15:16:37 +0000 (17:16 +0200)]
update TODO

9 years agocore: move status line ellipsation to 50% of the line
Lennart Poettering [Thu, 14 Aug 2014 15:15:09 +0000 (17:15 +0200)]
core: move status line ellipsation to 50% of the line

http://lists.freedesktop.org/archives/systemd-devel/2014-July/021591.html

9 years agoupdate TODO
Lennart Poettering [Thu, 14 Aug 2014 14:36:25 +0000 (16:36 +0200)]
update TODO

9 years agoanalyze: some fixes to the --help text
Lennart Poettering [Thu, 14 Aug 2014 14:36:18 +0000 (16:36 +0200)]
analyze: some fixes to the --help text

9 years agoRevert "socket: add support for TCP fast Open"
Lennart Poettering [Thu, 14 Aug 2014 14:34:47 +0000 (16:34 +0200)]
Revert "socket: add support for TCP fast Open"

This reverts commit 9528592ff8d7ff361da430285deba8196e8984d5.

Apparently TFO is actually the default at least for the server side now.

Also the setsockopt doesn't actually take a bool, but a qlen integer.