Lennart Poettering [Thu, 14 Aug 2014 01:30:40 +0000 (03:30 +0200)]
update TODO
Susant Sahani [Mon, 28 Jul 2014 06:48:29 +0000 (12:18 +0530)]
socket: add support for tcp nagle
This patch adds support for TCP TCP_NODELAY socket option. This can be
configured via NoDelay conf parameter. TCP Nagle's algorithm works by
combining a number of small outgoing messages, and sending them all at
once. This controls the TCP_NODELAY socket option.
Lennart Poettering [Thu, 14 Aug 2014 00:59:02 +0000 (02:59 +0200)]
logind: add new session type "web" for PAM web clients, such as cockpit
On request of Stef Walter.
Dave Reisner [Tue, 11 Mar 2014 14:41:22 +0000 (10:41 -0400)]
util: allow strappenda to take any number of args
This makes strappenda3 redundant, so we remove its usage and
definition. Add a few tests along the way for sanity.
Tom Gundersen [Wed, 13 Aug 2014 23:59:58 +0000 (01:59 +0200)]
sd-event: drop _likely_()
This is not certain to be likely.
Lennart says: a frequent usecase is invoking some function regularly in intervals
in such a case every single iteration we'll have to rearm
Tobias Geerinckx-Rice [Mon, 4 Aug 2014 20:15:52 +0000 (22:15 +0200)]
timer: order OnCalendar units after timer-sync.target if DefaultDependencies=no
Avoids triggering timers prematurely on systems with significantly inaccurate
clocks, or some embedded platforms that lack one entirely.
Tom Gundersen [Wed, 13 Aug 2014 23:35:16 +0000 (01:35 +0200)]
udev: link-config - fix crash due to missing hwaddr
Reported by: master.nosferatu@gmail.com
Tom Gundersen [Wed, 13 Aug 2014 23:28:35 +0000 (01:28 +0200)]
test: dhcp-server - fix test
Reported by Corey Hammerton on G+
Tom Gundersen [Wed, 13 Aug 2014 23:26:55 +0000 (01:26 +0200)]
sd-dhcp-server: linebreaks
Don't overflow unnecessarily.
Tom Gundersen [Wed, 13 Aug 2014 22:22:27 +0000 (00:22 +0200)]
sd-event: do not arm timers unnecessarily
Rather than recalculating the next timeout on every loop, we only do it when something changed.
Lennart Poettering [Wed, 13 Aug 2014 23:23:20 +0000 (01:23 +0200)]
networkctl: also use the same color logic when running "networkctl status" without arguments
Lennart Poettering [Wed, 13 Aug 2014 23:18:37 +0000 (01:18 +0200)]
networkctl: add the same color logic to "list" and "status" outputs
And always put operational state first, setup state second.
Lennart Poettering [Wed, 13 Aug 2014 23:10:08 +0000 (01:10 +0200)]
networkctl: name setup state variable setup_state
Lennart Poettering [Wed, 13 Aug 2014 23:08:46 +0000 (01:08 +0200)]
networkctl: update column header to new 'setup' state naming
Lennart Poettering [Wed, 13 Aug 2014 23:00:57 +0000 (01:00 +0200)]
resolved: enable LLMNR
THis was accidentally broken, as we truned off LLMNR far to frequently,
where we only wanted to turn off LLMNr on IPV6 on kernels lacking
support for it.
Lennart Poettering [Wed, 13 Aug 2014 23:00:15 +0000 (01:00 +0200)]
resolved: allow passing on which protocol, family and interface to look something up
Also, return on which protocol/family/interface we found something.
Umut Tezduyar Lindskog [Tue, 5 Aug 2014 10:17:09 +0000 (12:17 +0200)]
ldconfig: add configure option to disable
Tom Gundersen [Wed, 13 Aug 2014 20:55:49 +0000 (22:55 +0200)]
sd-networkd: rename link_get_state to link_get_setup_state
Suggested by Kay and Lennart.
Tom Gundersen [Wed, 13 Aug 2014 20:44:35 +0000 (22:44 +0200)]
sd-network: rename operstates 'down' -> 'off' and 'up' -> 'no-carrier'
Suggested by Lennart and Kay.
Tom Gundersen [Wed, 13 Aug 2014 20:37:45 +0000 (22:37 +0200)]
sd-network: /_get_link_/_link_get_/
The link is the 'object', so make this in line with our usual naming convention.
Suggested by Kay and Lennart.
Daniel Buch [Wed, 13 Aug 2014 20:02:27 +0000 (22:02 +0200)]
networkctl: color status dump without link name
Lets mimic colored operational state dump as if link name is appiled
Lennart Poettering [Wed, 13 Aug 2014 19:07:07 +0000 (21:07 +0200)]
timesyncd: don't trip up if networkd isn't running and can't tell us any NTP servers
Eelco Dolstra [Wed, 6 Aug 2014 11:14:51 +0000 (13:14 +0200)]
journald: Fix off-by-one error in "Missed X kernel messages" warning
On receiving a message, "kernel_seqnum" is set to "serial + 1". So
subtracting 1 will cause messages like "Missed 0 kernel messages",
which should be "Missed 1 kernel messages".
Lennart Poettering [Wed, 13 Aug 2014 18:02:38 +0000 (20:02 +0200)]
sd-network: bring comments up-to-date
Lennart Poettering [Wed, 13 Aug 2014 18:01:19 +0000 (20:01 +0200)]
mount-setup: fix counting of early mounts without SMACK
http://lists.freedesktop.org/archives/systemd-devel/2014-August/021772.html
Lennart Poettering [Wed, 13 Aug 2014 18:01:00 +0000 (20:01 +0200)]
update TODO
Lennart Poettering [Wed, 13 Aug 2014 16:53:05 +0000 (18:53 +0200)]
journald: also increase the SendBuffer of /dev/log to 8M
http://lists.freedesktop.org/archives/systemd-devel/2014-August/021825.html
Harald Hoyer [Wed, 13 Aug 2014 14:45:53 +0000 (16:45 +0200)]
switch-root: not only mkdir the parents but the target dir
switch-root would only mkdir the parents of the mount move directories.
With this patch, the mount move target directory is created to make
switch-root to an empty root directory work.
Denis Kenzior [Sat, 9 Aug 2014 07:06:21 +0000 (02:06 -0500)]
bus-bloom: Fix bloom filter calculation
i is being used incorrectly. It is used to refer to the number of
indexes calculated so far (out of k). However, it is also incremented
when a new hash key is being used. This means that the results are
inconsistent with the desired behavior described in PORTING-DBUS1
document.
The expected result is that for the default values of m and k (512, 8)
the 1st hash key should produce 4 indexes. The second hash key is used
for the next 4 and overall 8 indexes into m are calculated.
The current behavior results in 6 indexes being calculated, 4 coming
from hash key 1 and 2 others from hash key 5.
Tom Gundersen [Wed, 13 Aug 2014 13:43:57 +0000 (15:43 +0200)]
TODO
Tom Gundersen [Wed, 13 Aug 2014 13:40:59 +0000 (15:40 +0200)]
TODO
Tom Gundersen [Wed, 13 Aug 2014 13:36:11 +0000 (15:36 +0200)]
networkctl: add colors to admin state and fix alignment
It is useful to color in the admin state both to easily spot failed links, but also to quickly
distinguish between links that are fully configured and in degraded mode (only IPv4LL) or in
degraded mode and still waiting for DHCP.
Tom Gundersen [Wed, 13 Aug 2014 13:34:27 +0000 (15:34 +0200)]
sd-network: rename state INITALIZING to PENDING and expose as any other state
This is the state when we are waiting for udev to initialize the device, and waiting for
libudev and rtnl to be in sync. In the future we probably will also be waiting for nl80211.
At this point we do not yet have enough information to know whether or not networkd should
be handling the device.
Lennart Poettering [Wed, 13 Aug 2014 13:00:12 +0000 (15:00 +0200)]
resolved: skip IPv6 LLMNR if IPv6 is not available
David Herrmann [Mon, 11 Aug 2014 16:17:54 +0000 (18:17 +0200)]
login: share VT-signal handler between sessions
sd-event does not allow multiple handlers for a single signal. However,
logind sets up signal handlers for each session with VT_PROCESS set (that
is, it has an active controller). Therefore, registering multiple such
controllers will fail.
Lets make the VT-handler global, as it's mostly trivial, anyway. This way,
the sessions don't have to take care of that and we can simply acknowledge
all VT-switch requests as we always did.
Tom Gundersen [Tue, 12 Aug 2014 21:12:09 +0000 (23:12 +0200)]
networkd: dhcp4 - fix copy-paste error
Reported by poma.
Tom Gundersen [Tue, 12 Aug 2014 18:50:13 +0000 (20:50 +0200)]
networkd: disable ipv4ll default routes by default
This causes machines without connectivity to hang where they would otherwise fail. Keep it
opt-in for now, but consider whether we sholud just drop it.
Tom Gundersen [Tue, 12 Aug 2014 18:00:34 +0000 (20:00 +0200)]
sd-network: split the operstate 'unknown' into 'down' and 'up'
Tom Gundersen [Fri, 8 Aug 2014 10:12:17 +0000 (12:12 +0200)]
networkd: link - split out dhcp4 handling
Tom Gundersen [Tue, 12 Aug 2014 11:15:11 +0000 (13:15 +0200)]
udev: link-config - fix naming policy
Don't exit the name-finding loop when the 'kernel' policy is detected. We should
still find a fallback name if possible in the (very likely) case that no kernel
name is set at all.
Tom Gundersen [Tue, 12 Aug 2014 11:13:12 +0000 (13:13 +0200)]
udev: link-config - whitespace
Don't overflow 80 cols unneccessarily (we are ok with doing it when it improves readability, but that's not the case here).
Tom Gundersen [Wed, 6 Aug 2014 13:54:25 +0000 (15:54 +0200)]
networkd: link - fix IN_SET usage
Tom Gundersen [Wed, 6 Aug 2014 13:54:03 +0000 (15:54 +0200)]
networkd: link - split out ipv4ll handling
Tom Gundersen [Wed, 6 Aug 2014 11:48:11 +0000 (13:48 +0200)]
networkd: link - make check for whether addresses/routes are being configured implicit
Tom Gundersen [Wed, 6 Aug 2014 10:50:53 +0000 (12:50 +0200)]
networkd: split out networkd-link.h
Lennart Poettering [Tue, 12 Aug 2014 17:50:36 +0000 (19:50 +0200)]
update TODO
Lennart Poettering [Tue, 12 Aug 2014 17:49:15 +0000 (19:49 +0200)]
man: minimal updates to resolved.conf man page
Lennart Poettering [Tue, 12 Aug 2014 17:48:24 +0000 (19:48 +0200)]
timesyncd: beef up NTP server selection logic, and acquire NTP servers from DHCP
Lennart Poettering [Tue, 12 Aug 2014 17:33:09 +0000 (19:33 +0200)]
resolved: return exit code of event loop
Lennart Poettering [Tue, 12 Aug 2014 17:32:55 +0000 (19:32 +0200)]
resolved: unify logic how we flush out DNS servers we learnt
Lennart Poettering [Tue, 12 Aug 2014 17:29:38 +0000 (19:29 +0200)]
sd-resolve: scale number of threads by queries currently being processed, rather than number of queries currently referenced
This way a query that blocks for a longer time, and whose reference has
been dropped won't block subsequent lookups from finishing.
Lennart Poettering [Tue, 12 Aug 2014 17:29:27 +0000 (19:29 +0200)]
sd-resolve: fix allocation if query ids, never reuse them
Lennart Poettering [Tue, 12 Aug 2014 17:28:32 +0000 (19:28 +0200)]
sd-resolve: keep floating query referenced while we access it
Lennart Poettering [Tue, 12 Aug 2014 14:58:56 +0000 (16:58 +0200)]
timesyncd: split up into multiple source file
The source file got much too large, hence split up the sources into
multiple per-object files, similar in style to resolved.
Lennart Poettering [Tue, 12 Aug 2014 14:19:37 +0000 (16:19 +0200)]
networkctl: make networkctl status without link name show all local IP addresses
Lennart Poettering [Tue, 12 Aug 2014 14:12:53 +0000 (16:12 +0200)]
update TODO
Lennart Poettering [Tue, 12 Aug 2014 14:03:45 +0000 (16:03 +0200)]
networkctl: when outputing a list fo links, sort them by ifindex
Lennart Poettering [Tue, 12 Aug 2014 13:41:13 +0000 (15:41 +0200)]
update TODO
Lennart Poettering [Tue, 12 Aug 2014 13:41:01 +0000 (15:41 +0200)]
networkctl: add a bit of color to the output
Lennart Poettering [Tue, 12 Aug 2014 13:19:30 +0000 (15:19 +0200)]
networkctl: add new switch "-a" to "networkctl status" to show verbose status of all local links
Lennart Poettering [Tue, 12 Aug 2014 13:08:09 +0000 (15:08 +0200)]
networkctl: properly format MAC addresses
Lennart Poettering [Tue, 12 Aug 2014 13:05:21 +0000 (15:05 +0200)]
sd-network: add API to output all collected system-wide NTP and DNS servers
Also, output the collected information in "networkctl".
Lennart Poettering [Tue, 12 Aug 2014 12:54:32 +0000 (14:54 +0200)]
timesyncd: move the generic sockaddr_equal() call to socket-util.h
Lennart Poettering [Tue, 12 Aug 2014 11:01:58 +0000 (13:01 +0200)]
networkd: expose a global list of DNS and NTP servers in the state file
Lennart Poettering [Tue, 12 Aug 2014 10:21:10 +0000 (12:21 +0200)]
resolved: filter out duplicate DNS servers when writing resolv.conf
Lennart Poettering [Tue, 12 Aug 2014 09:55:06 +0000 (11:55 +0200)]
networkd: update manager_save() to use fflush_and_check() to simplify things a bit
Thomas Hindoe Paaboel Andersen [Tue, 12 Aug 2014 02:42:33 +0000 (04:42 +0200)]
resolved: initialize counter
introduced in:
a407657425a3e47fd2b559cd3bc800f791303f63
Lennart Poettering [Tue, 12 Aug 2014 00:33:37 +0000 (02:33 +0200)]
networkd: fix build
Lennart Poettering [Tue, 12 Aug 2014 00:12:05 +0000 (02:12 +0200)]
sd-network: move sd-network API into libsystemd proper
In contrast to the DHCP/IPv4LL/ICMP6 APIs sd-network is not a protocol
implementation but a client API for networkd, hence move it into
libsystemd proper.
Lennart Poettering [Tue, 12 Aug 2014 00:04:17 +0000 (02:04 +0200)]
sd-network: remove client-side dhcp API
The networkd should abstract the difference between DHCP supplied and
configured data, and hence the DHCP lease concept should not exposed on
the client side.
Should we want to support arbitrary DHCP fields one day, we can add a
new sd_network_get_link_dhcp_field() call or so.
Lennart Poettering [Mon, 11 Aug 2014 23:41:42 +0000 (01:41 +0200)]
networkd: add minimal client tool "networkd" to query network status
In the long run this should become a full fledged client to networkd
(but not before networkd learns bus support). For now, just pull
interesting data out of networkd, udev, and rtnl and present it to the
user, in a simple but useful output.
Lennart Poettering [Mon, 11 Aug 2014 23:43:23 +0000 (01:43 +0200)]
udev: modernize net_id builtin a bit
Lennart Poettering [Mon, 11 Aug 2014 20:56:52 +0000 (22:56 +0200)]
sd-network: rename the per-ifindex calls to sd_network_get_link_xxx()
This way we can introduce system-wide versions of these calls that are
called the same way, but without the "link" in the name.
Lennart Poettering [Mon, 11 Aug 2014 20:44:51 +0000 (22:44 +0200)]
networkd: monopolize in_addr utility functions in shared/in-addr-util.h
Primarily, this means we get rid of net_parse_inaddr(), and replace it
everywhere with in_addr_from_string() and in_addr_from_string_auto().
These functions do not clobber the callers arguments on failure, which
is more close to our usual coding style.
Lennart Poettering [Mon, 11 Aug 2014 20:04:05 +0000 (22:04 +0200)]
timesyncd: when we don't know anything about the network state, assume we are online
This should provide better compatibility with systems that lack networkd
Lennart Poettering [Mon, 11 Aug 2014 18:15:14 +0000 (20:15 +0200)]
build-sys: add missing makefile symlinks
Lennart Poettering [Mon, 11 Aug 2014 18:14:27 +0000 (20:14 +0200)]
man: drop any reference to "syslog" as log target
Lennart Poettering [Mon, 11 Aug 2014 18:13:38 +0000 (20:13 +0200)]
udev: never bypass our own logging framework and call vsyslog() directly from udev tools
Lennart Poettering [Mon, 11 Aug 2014 18:08:08 +0000 (20:08 +0200)]
log: never ever log to syslog from PID 1, log to the journal again
We don't support journal-less systems anyway, so let's avoid the
confusion.
Lennart Poettering [Mon, 11 Aug 2014 17:28:53 +0000 (19:28 +0200)]
man: improve documentation for StandardOutput= and StandardInput=
Olivier Brunel [Fri, 8 Aug 2014 18:45:43 +0000 (20:45 +0200)]
login: set_controller should fail if prepare_vt fails
If controllers can expect logind to have "prepared" the VT (e.g. set it to
graphics mode, etc) then TakeControl() should fail if said preparation
failed (and session_restore_vt() was called).
(David: fixed up !CONFIG_VT case and errno-numbers)
Lennart Poettering [Mon, 11 Aug 2014 16:35:54 +0000 (18:35 +0200)]
fix a couple of more lazy "return -1"
Fix should strictly follow the rule to return negative errno-style
error codes from functions, hence let's fix more "return -1"-style
lazinesses.
Lennart Poettering [Mon, 11 Aug 2014 16:24:43 +0000 (18:24 +0200)]
journal-remote: fix parsing of fd command line argument
Lennart Poettering [Mon, 11 Aug 2014 16:24:28 +0000 (18:24 +0200)]
initctl: don't eat up error code
Lennart Poettering [Mon, 11 Aug 2014 16:24:08 +0000 (18:24 +0200)]
bootchart: use the bool type where appropriate
Lennart Poettering [Mon, 11 Aug 2014 16:23:47 +0000 (18:23 +0200)]
bootchart: it's not OK to return -1 from a main program
Lennart Poettering [Mon, 11 Aug 2014 14:41:05 +0000 (16:41 +0200)]
man: document that we look for both the instance's and the template's .d/ directory when looking for unit file drop-ins.
Lennart Poettering [Mon, 11 Aug 2014 13:32:09 +0000 (15:32 +0200)]
util: avoid considering dpkg temporary files relevant for anything
https://bugs.freedesktop.org/show_bug.cgi?id=82453
Lennart Poettering [Sun, 10 Aug 2014 21:40:48 +0000 (23:40 +0200)]
timesyncd: always use CLOCK_BOOTTIME if we can
After all we want to compare a monotonically increasing clock with the
remote clock, hence we shouldn't ignore system suspend periods.
Lennart Poettering [Sun, 10 Aug 2014 21:40:18 +0000 (23:40 +0200)]
man: update resolved man pages a bit
Lennart Poettering [Sun, 10 Aug 2014 21:36:41 +0000 (23:36 +0200)]
man: extend timesycnd man page a bit
Lennart Poettering [Sun, 10 Aug 2014 21:10:08 +0000 (23:10 +0200)]
resolved: when picking a new hostname make sure two hosts pick different ones
This way we can avoid always picking the same replacement hostnames when
picking one.
Lennart Poettering [Sun, 10 Aug 2014 20:48:16 +0000 (22:48 +0200)]
resolved: actually, the peer with the lower IP address wins conflicts
Lennart Poettering [Sun, 10 Aug 2014 20:28:12 +0000 (22:28 +0200)]
resolved: make sure we don't mark the wrong zone RRs conflicting
Lennart Poettering [Wed, 6 Aug 2014 15:21:00 +0000 (17:21 +0200)]
resolved: be a bit more communicative about conflicts
Lennart Poettering [Wed, 6 Aug 2014 15:00:14 +0000 (17:00 +0200)]
update TODO
Lennart Poettering [Wed, 6 Aug 2014 14:59:48 +0000 (16:59 +0200)]
resolved: verify all RRs when we come back from suspend
Lennart Poettering [Wed, 6 Aug 2014 14:32:55 +0000 (16:32 +0200)]
resolved: destroy outstanding queries if the clients that initiated them die
Lennart Poettering [Wed, 6 Aug 2014 14:32:17 +0000 (16:32 +0200)]
resolved: properly pass aborted transaction result back to clients
Lennart Poettering [Wed, 6 Aug 2014 14:15:35 +0000 (16:15 +0200)]
resolved: implement full LLMNR conflict detection logic