Logstash is a data processing pipeline that processes data. It can receive, collect, parse, transform and forward log events. This repository includes various pattern for the Logstash filter grok.

The grok filter is included in a default Logstash installation. It is used to parse log events and split messages into multiple fields. Instead of writing regular expressions, users use predefined patterns to parse logs. Besides the included patterns, custom patterns can be added to extend the functionality.

1. Installation
2. Examples
   • Icinga 2 Main Log
   • Icinga 2 Debug Log
   • Icinga 2 Startup Log

Custom patterns need to be accessible by the Logstash daemon. It does not matter where you put the files on the file system, as long as they are readable to the logstash user.

mkdir /etc/logstash/patterns
cd /etc/logstash/patterns
git clone https://github.com/Icinga/logstash-grok-pattern.git icinga

To use custom patterns, include the directory with the patterns_dir paramter in your grok filter:

grok {
  patterns_dir   => ["/etc/logstash/patterns/icinga"]
  ...
}

The following examples demonstrate how the patterns can be used to parse Icinga log files. These are just examples, you are free to use the patterns in which way you want.

The main log of Icinga 2 includes general information about the behaviour of the process, each component of it and all enabled features. Logs are split into three fields: severity,facility and message:

This example is based on a Logstash file input plugin. Other inputs can be used as well. The date filter moves the timestamp of the log event to the field @timestamp, which is used by default in Kibana to sort events.

input {
  file {
    path => "/var/log/icinga2/icinga2.log"
    type => "icinga.main"
    codec => multiline {
      pattern             => "^\["
      negate              => true
      what                => previous
      auto_flush_interval => 2
    }
  }
}

filter {
  if [type] == "icinga.main" {
    grok {
      patterns_dir   => ["/etc/logstash/patterns/icinga"]
      match          => ["message", "%{ICINGA_MAIN}"]
      remove_field   => ["message"]
      add_tag        => ["filter.grok.icinga.main"]
      tag_on_failure => ["_grokparsefailure", "filter.icinga.main.grok.failure"]
    }

    date {
      match          => ["icinga.main.timestamp", "yyyy-MM-dd HH:mm:ss Z"]
      target         => "@timestamp"
      remove_field   => ["icinga.main.timestamp"]
      tag_on_failure => ["_dateparsefailur", "filter.icinga.debug.date.failure"]
    }
  }
}

output {
  stdout {
    codec => "rubydebug"
  }
}

The main log of Icinga 2 includes very detailed information about the behaviour of the process, each component of it and all enabled features. Logs are split into three fields: severity,facility and message:

This example is based on a Logstash file input plugin. Other inputs can be used as well. The date filter moves the timestamp of the log event to the field @timestamp, which is used by default in Kibana to sort events.

input {
  file {
    path => "/var/log/icinga2/debug.log"
    type => "icinga.debug"
    codec => multiline {
      pattern             => "^\["
      negate              => true
      what                => previous
      auto_flush_interval => 2
    }
  }
}

filter {
  if [type] == "icinga.debug" {
    grok {
      patterns_dir   => ["/etc/logstash/patterns/icinga"]
      match          => ["message", "%{ICINGA_DEBUG}"]
      remove_field   => ["message"]
      add_tag        => ["filter.grok.icinga.debug"]
      tag_on_failure => ["_grokparsefailure", "filter.icinga.debug.grok.failure"]
    }

    date {
      match          => ["icinga.debug.timestamp", "yyyy-MM-dd HH:mm:ss Z"]
      target         => "@timestamp"
      remove_field   => ["icinga.debug.timestamp"]
      tag_on_failure => ["_dateparsefailur", "filter.icinga.debug.date.failure"]
    }
  }
}

output {
  stdout {
    codec => "rubydebug"
  }
}

The startup log of Icinga 2 is generated each time when the daemon is restarted or reloaded. It includes information about the amount of objects on startup, which features are enabled, connection to the database and suchlike. The startup log does not include a timestamp, the file is rewritten completely every time.

This example is based on a Logstash file input plugin. Other inputs can be used as well.

input {
  file {
    path           => "/var/log/icinga2/startup.log"
    type           => "icinga.startup"
    start_position => "beginning"
    sincedb_path   => "/dev/null"
    codec          => multiline {
      pattern             => "^[a-z]*\/[a-zA-Z]*:"
      negate              => true
      what                => previous
      auto_flush_interval => 2
    }
  }
}

filter {
  if [type] == "icinga.startup" {
    grok {
      patterns_dir   => ["/etc/logstash/patterns/icinga"]
      match          => ["message", "%{ICINGA_STARTUP}"]
      remove_field   => ["message"]
      add_tag        => ["filter.grok.icinga.startup"]
      tag_on_failure => ["_grokparsefailure", "filter.grok.icinga.startup.failure"]
    }
  }
}

output {
  stdout {
    codec => "rubydebug"
  }
}