Also improve the description: notime, notq, and nohost are literals,
not things to be replaced by something else -- so no angular brackets.
Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
Make compilation to work in systems which don't have sys/timex.h and its
ntp_gettime().
Reviewed-by: Sami Kerola <kerolasa@iki.fi>
Signed-off-by: Samuel Thibault <sthibault@debian.org>
This adds a concise description of a tool to its usage text.
A first form of this patch was proposed by Steven Honeyman
(see http://www.spinics.net/lists/util-linux-ng/msg09994.html).
Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
Either works on linux, but kfreebsd build fails
if we don't use the <sys/time.h> include.
According to man gettimeofday the correct include is <sys/time.h>
Signed-off-by: Andreas Henriksson <andreas@fatal.se>
Setting whole array to be completely full of nulls cannot be as quick as
making the only significant member of the array when needed.
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
There is no obvious way to make syslog(3) to print both pid or ppid, so
duplicate the libc syslog() to logger. Making the ppid printing work
using unix socket has side effect of local becoming capable to use both
rfc format output, which is hopefully seen as good thing. The
syslog_local() is format wise one-to-one copy with glibc syslog(3)
format.
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
- use allocated buffers to compose the final messages
- initialize static buffers to zero
- remove strlen-after-sprintf
Signed-off-by: Karel Zak <kzak@redhat.com>
This makes the obsolete protocol a little bit more compliant with the
internet standard, but few should care now when we have rfc5424 support,
and rfc3164 feels broken. For example it requires hostname to be not
fully qualified, which is hard to understand, and should make users to
prefer the new protocol.
Reported-by: Frank Thilo <thilo@unix-ag.org>
Bug-Debian: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=705217
CC: Andreas Henriksson <andreas@fatal.se>
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
When --file is combined with command line arguments the later has
silently been ignored. This commit makes user to be aware the logger
will not use command line arguments when --file is specified.
Reported-by: "Daniel 'DaB.' Baur" <debian@daniel.baur4.info>
Bug-Debian: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=467244
CC: Andreas Henriksson <andreas@fatal.se>
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Users wish to see the message should include also remote messages, not
only the one sent to locally via libc function.
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
When scripts send several messages they will be easier to group together
when parent process id is printed rather than id of the each logger
process.
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Add support the more recent syslog protocol and make it default. The
older BSD syslog protocol can still be used with option --rfc3164.
Protocols are meaningful only when messages are sent to remote syslog
server.
Requested-by: Kodiak Firesmith <ksf@sei.cmu.edu>
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Add logger_open(), logger_command_line(), logger_stdin(), and
logger_close() functions, and move all remaining option argument
assignments to control structure.
Proposed-by: Karel Zak <kzak@redhat.com>
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
When if clause that continues throughout whole function it usually can be
shorten to immediate action, e.g., in this case return on the spot not at
end of the function.
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
The getlogin(3) is known not to always work, and when that happens it is
reasonable to try determine user of name by looking process owner and
passwd information.
Reference: http://man7.org/linux/man-pages/man3/getlogin.3.html
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
It is fair assumption messages an user is asking to be wrote will be
attempted to be wrote as hard as possible.
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Fixing plain typos, miswordings, inconsistent periods, some missing
angular brackets, and a proper pluralization (even when it involves
a constant, because for some languages the precise value matters).
Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
Earlier use of unknown facility or priority number was accepted, and
resulted in unexpected result. For example when looking journalctl
--priority=7.8 was converted to priotity 0 and facility 1.
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
* systemd (since v209) uses only one library (when compiled
without --enable-compat-libs)
* all systemd build-sys stuff is merged into HAVE_SYSTEMD
(automake) and HAVE_LIBSYSTEMD (C macro) now
* all is controlled by --with-systemd, default is to automatically
check for systemd libs
* no more --enable-socket-activation and --enable-journald
Signed-off-by: Karel Zak <kzak@redhat.com>
This feature is hopefully mostly used to give MESSAGE_ID labels for
messages coming from scripts, making search of messages easy. The
logger(1) manual page update should give enough information how to use
--journald option.
[kzak@redhat.com: - add missing #ifdefs
- use xalloc.h]
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Signed-off-by: Karel Zak <kzak@redhat.com>
This patch adds a new option to logger that will make it look for a
priority prefix <n> at the beginning of every line. The priority is
a single decimal number formed as explained in syslog(3).
If a prefix is found logger will log the message using the found
facility and level in that prefix, if the prefix doesn't contain a
facility the default facility specified by the -p option will be used.
If no prefix is found, logger will use the priority specified by -p.
[kzak@redhat.com: - add --prio-prefix to usage() output]
Signed-off-by: Dennis H Jensen <dennis.h.jensen@siemens.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
This commit fixes error in usage() text, which claimed TCP is default
transport protocol. That was not true, and neither it should be. The
syslog messages has traditionally sent using UDP.
For the logger remains using UDP as first transport, but if it fails a
TCP connection is attempted. If an user wishes remote logging can be
forced to use either UDP or TCP. The service port for UDP is familiar
'syslog', for TCP the port 'syslog-conn' seems like reasonable default.
[kzak@redhat.com: - rename myopenlog to unix_socket(),
- always reset st to -1]
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Signed-off-by: Karel Zak <kzak@redhat.com>
When journald, from systemd, is in use logger gave following error.
$ logger --socket /dev/log test logger: connect /dev/log: Protocol wrong
type for socket
The journald supports only AF_DGRAM, and nothing else. Support for
AF_STREAM sockets was dropped because it messed up message ordering.
From an users point of view necessity to add --udp (UNIX(TM) Datagram
Protocol?) option when talking to syslog via unix socket felt confusing
and wrong. The command should know what is the socket type, and silently
use the correct one, unless the type is explicitely defined.
CC: Karel Zak <kzak@redhat.com>
Adviced-by: Lennart Poettering <lennart@poettering.net>
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Move these functions to the top of the file where they can be marked
static and the prototypes can be removed.
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
This is done for us via an atexit hook since c05a80ca63. Avoids a
useless 'Write error' on exit whenever invoking the tool.
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
Karel Zak
Benno Schulenberg
Samuel Thibault
Andreas Henriksson
Sami Kerola
Dennis H Jensen
Dave Reisner
maximilian attems