syslogngfreen/syslog-ng

https://badges.gitter.im/Join%20Chat.svg](https://gitter.im/syslog-ng/syslog-ng?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=body_badge) https://travis-ci.org/syslog-ng/syslog-ng.svg?branch=master](https://travis-ci.org/syslog-ng/syslog-ng)

syslog-ng

syslog-ng is an enhanced log daemon, supporting a wide range of input and output methods: syslog, unstructured text, message queues, databases (SQL and NoSQL alike), and more.

Quickstart

The simplest configuration accepts system logs from /dev/log (from applications or forwarded by systemd) and writes everything to a single file:

@version: 3.29
@include "scl.conf"

log {
	source { system(); };
	destination { file("/var/log/syslog"); };
};

This one additionally processes logs from the network (TCP/514 by default):

@version: 3.29
@include "scl.conf"

log {
	source {
		system();
		network();
	};
	destination { file("/var/log/syslog"); };
};

This config is designed for structured/application logging, using local submission via JSON, and outputting in key=value format:

@version: 3.29
@include "scl.conf"

log {
	source { system(); };
	destination { file("/var/log/app.log" template("$(format-welf --subkeys .cim.)\n")); };
};

To submit a structured log using logger, you might run:

$ logger '@cim: {"name1":"value1", "name2":"value2"}'

In which case the resulting message will be:

name1=value1 name2=value2

For a brief introduction to configuring the syslog-ng application, see the quickstart guide.

Features

• Receive and send https://tools.ietf.org/html/rfc3164 and https://tools.ietf.org/html/rfc5424 style syslog messages
• Receive and send JSON formatted messages
• Work with any kind of unstructured data
• Classify and structure logs using built-in parsers (csv-parser(), db-parser(), kv-parser(), etc.)
• Normalize, crunch, and process logs as they flow through the system
• Hand over logs for further processing using files, message queues (like AMQP), or databases (like http://www.postgresql.org/ or http://www.mongodb.org/)
• Forward logs to big data tools (like https://www.elastic.co/, http://kafka.apache.org/, or http://hadoop.apache.org/)

Performance

• syslog-ng provides performance levels comparable to a large cluster when running on a single node
• In the simplest use case, it scales up to 600-800k messages per second
• But classification, parsing, and filtering still produce several tens of thousands of messages per second

Community

• syslog-ng is developed by a community of volunteers, the best way to contact us is via our http://github.com/syslog-ng/syslog-ng project, our https://gitter.im/syslog-ng/syslog-ng or our mailing list.
• syslog-ng is integrated into almost all Linux distributions and BSDs, it is also incorporated into a number of products, see our powered by syslog-ng page for more details.

***

Balabit is the original creator and largest current sponsor of the syslog-ng project. They offer support, professional services, and addons you may be interested in

Feedback

We are really interested to see who uses our software, so if you do use it and you like what you see, please tell us about it. A star on github or an email saying thanks means a lot already, but telling us about your use case, your experience, and things to improve would be much appreciated.

Just send an email to feedback (at) syslog-ng.org.

Feedback Powers Open Source.

Installation from source

Releases and precompiled tarballs are available on https://github.com/syslog-ng/syslog-ng/releases.

To compile from source, the easiest is to use dbld, a docker based, self-hosted compile/build/release infrastructure within the source tree. See dbld/README.md for more information.

For the brave souls who want to compile syslog-ng from scratch, the usual drill applies:

$ ./configure && make && make install

The extra effort in contrast with the dbld based build is the need to fetch and install all build dependencies of syslog-ng (of which there are a few).

If you don't have a configure script (because of cloning from git, for example), run ./autogen.sh to generate it.

Some of the functionality of syslog-ng is compiled only if the required development libraries are present. The configure script displays a summary of enabled features at the end of its run. For details, see the syslog-ng compiling instructions.

Installation from ***aries

***aries are available in various Linux distributions and contributors maintain packages of the latest and greatest syslog-ng version for various OSes.

Debian/Ubuntu

Simply invoke the following command as root:

# apt-get install syslog-ng

The latest versions of syslog-ng are available for a wide range of Debian and Ubuntu releases and architectures from an https://build.opensuse.org/project/show/home:laszlo_budai:syslog-ng.

For instructions on how to install syslog-ng on Debian/Ubuntu distributions, see the blog post Installing the latest syslog-ng on Ubuntu and other DEB distributions.

Fedora

syslog-ng is available as a Fedora package that you can install using yum:

# yum install syslog-ng

You can download packages for the latest versions from https://copr.fedoraproject.org/coprs/czanik/.

For instructions on how to install syslog-ng on RPM distributions, see the blog post Installing latest syslog-ng on RHEL and other RPM distributions.

If you wish to install the latest RPM package that comes from a recent commit in Git for testing purposes, read the blog post, RPM packages from syslog-ng Git HEAD.

Others

***aries for other platforms are listed on the official third party page.

Installation from Docker image

***aries are also available as a Docker image. To find out more, check out the blog post, Your central log server in Docker.

Documentation

The documentation of the latest released version of syslog-ng Open Source Edition is available here. For earlier versions, see the syslog-ng Documentation Page.

Contributing

If you would like to contribute to syslog-ng, to fix a bug or create a new module, the syslog-ng gitbook helps you take the first steps to working with the code base.