This filter helps automatically parse messages (or specific event fields) which are of the foo=bar variety.

For example, if you have a log message which contains ip=1.2.3.4 error=REFUSED, you can parse those automatically by configuring:

    filter {
      kv { }
    }

The above will result in a message of ip=1.2.3.4 error=REFUSED having the fields:

  • ip: 1.2.3.4
  • error: REFUSED

This is great for postfix, iptables, and other types of logs that tend towards key=value syntax.

You can configure any arbitrary strings to split your data on, in case your data is not structured using = signs and whitespace. For example, this filter can also be used to parse query parameters like foo=bar&baz=fizz by setting the field_split parameter to &.

 

Synopsis

edit

This plugin supports the following configuration options:

Required configuration options:

kv {
}

Available configuration options:

Setting Input type Required Default value

add_field

hash

No

{}

add_tag

array

No

[]

allow_duplicate_values

boolean

No

true

default_keys

hash

No

{}

exclude_keys

array

No

[]

field_split

string

No

" "

include_brackets

boolean

No

true

include_keys

array

No

[]

periodic_flush

boolean

No

false

prefix

string

No

""

recursive

boolean

No

false

remove_field

array

No

[]

remove_tag

array

No

[]

source

string

No

"message"

target

string

No

trim

string

No

trimkey

string

No

value_split

string

No

"="

Details

edit

 

add_field

edit
  • Value type is hash
  • Default value is {}

If this filter is successful, add any arbitrary fields to this event. Field names can be dynamic and include parts of the event using the %{field}.

Example:

    filter {
      kv {
        add_field => { "foo_%{somefield}" => "Hello world, from %{host}" }
      }
    }
[source,ruby]
    # You can also add multiple fields at once:
    filter {
      kv {
        add_field => {
          "foo_%{somefield}" => "Hello world, from %{host}"
          "new_field" => "new_static_value"
        }
      }
    }

If the event has field "somefield" == "hello" this filter, on success, would add field foo_hello if it is present, with the value above and the %{host} piece replaced with that value from the event. The second example would also add a hardcoded field.

add_tag

edit
  • Value type is array
  • Default value is []

If this filter is successful, add arbitrary tags to the event. Tags can be dynamic and include parts of the event using the %{field} syntax.

Example:

    filter {
      kv {
        add_tag => [ "foo_%{somefield}" ]
      }
    }
[source,ruby]
    # You can also add multiple tags at once:
    filter {
      kv {
        add_tag => [ "foo_%{somefield}", "taggedy_tag"]
      }
    }

If the event has field "somefield" == "hello" this filter, on success, would add a tag foo_hello (and the second example would of course add a taggedy_tag tag).

allow_duplicate_values

edit
  • Value type is boolean
  • Default value is true

A bool option for removing duplicate key/value pairs. When set to false, only one unique key/value pair will be preserved.

For example, consider a source like from=me from=me. [from] will map to an Array with two elements: ["me", "me"]. to only keep unique key/value pairs, you could use this configuration:

    filter {
      kv {
        allow_duplicate_values => false
      }
    }

default_keys

edit
  • Value type is hash
  • Default value is {}

A hash specifying the default keys and their values which should be added to the event in case these keys do not exist in the source field being parsed.

    filter {
      kv {
        default_keys => [ "from", "[email protected]",
                         "to", "[email protected]" ]
      }
    }

exclude_keys

edit
  • Value type is array
  • Default value is []

An array specifying the parsed keys which should not be added to the event. By default no keys will be excluded.

For example, consider a source like Hey, from=<abc>, to=def foo=bar. To exclude from and to, but retain the foo key, you could use this configuration:

    filter {
      kv {
        exclude_keys => [ "from", "to" ]
      }
    }

field_split

edit
  • Value type is string
  • Default value is " "

A string of characters to use as delimiters for parsing out key-value pairs.

These characters form a regex character class and thus you must escape special regex characters like [ or ] using \.

Example with URL Query Strings

For example, to split out the args from a url query string such as ?pin=12345~0&d=123&[email protected]&oq=bobo&ss=12345:

    filter {
      kv {
        field_split => "&?"
      }
    }

The above splits on both & and ? characters, giving you the following fields:

include_brackets

edit
  • Value type is boolean
  • Default value is true

A boolean specifying whether to include brackets as value wrappers (the default is true)

    filter {
      kv {
        include_brackets => true
      }
    }

For example, the result of this line: bracketsone=(hello world) bracketstwo=[hello world]

will be: * bracketsone: hello world * bracketstwo: hello world

instead of: * bracketsone: (hello * bracketstwo: [hello

include_keys

edit
  • Value type is array
  • Default value is []

An array specifying the parsed keys which should be added to the event. By default all keys will be added.

For example, consider a source like Hey, from=<abc>, to=def foo=bar. To include from and to, but exclude the foo key, you could use this configuration:

    filter {
      kv {
        include_keys => [ "from", "to" ]
      }
    }

periodic_flush

edit
  • Value type is boolean
  • Default value is false

Call the filter flush method at regular interval. Optional.

prefix

edit
  • Value type is string
  • Default value is ""

A string to prepend to all of the extracted keys.

For example, to prepend arg_ to all keys:

    filter { kv { prefix => "arg_" } }

recursive

edit
  • Value type is boolean
  • Default value is false

A boolean specifying whether to drill down into values and recursively get more key-value pairs from it. The extra key-value pairs will be stored as subkeys of the root key.

Default is not to recursive values.

    filter {
      kv {
        recursive => "true"
      }
    }

remove_field

edit
  • Value type is array
  • Default value is []

If this filter is successful, remove arbitrary fields from this event. Fields names can be dynamic and include parts of the event using the %{field} Example:

    filter {
      kv {
        remove_field => [ "foo_%{somefield}" ]
      }
    }
[source,ruby]
    # You can also remove multiple fields at once:
    filter {
      kv {
        remove_field => [ "foo_%{somefield}", "my_extraneous_field" ]
      }
    }

If the event has field "somefield" == "hello" this filter, on success, would remove the field with name foo_hello if it is present. The second example would remove an additional, non-dynamic field.

remove_tag

edit
  • Value type is array
  • Default value is []

If this filter is successful, remove arbitrary tags from the event. Tags can be dynamic and include parts of the event using the %{field} syntax.

Example:

    filter {
      kv {
        remove_tag => [ "foo_%{somefield}" ]
      }
    }
[source,ruby]
    # You can also remove multiple tags at once:
    filter {
      kv {
        remove_tag => [ "foo_%{somefield}", "sad_unwanted_tag"]
      }
    }

If the event has field "somefield" == "hello" this filter, on success, would remove the tag foo_hello if it is present. The second example would remove a sad, unwanted tag as well.

source

edit
  • Value type is string
  • Default value is "message"

The field to perform key=value searching on

For example, to process the not_the_message field:

    filter { kv { source => "not_the_message" } }

target

edit
  • Value type is string
  • There is no default value for this setting.

The name of the container to put all of the key-value pairs into.

If this setting is omitted, fields will be written to the root of the event, as individual fields.

For example, to place all keys into the event field kv:

    filter { kv { target => "kv" } }

trim

edit
  • Value type is string
  • There is no default value for this setting.

A string of characters to trim from the value. This is useful if your values are wrapped in brackets or are terminated with commas (like postfix logs).

These characters form a regex character class and thus you must escape special regex characters like [ or ] using \.

For example, to strip <, >, [, ] and , characters from values:

    filter {
      kv {
        trim => "<>\[\],"
      }
    }

trimkey

edit
  • Value type is string
  • There is no default value for this setting.

A string of characters to trim from the key. This is useful if your keys are wrapped in brackets or start with space.

These characters form a regex character class and thus you must escape special regex characters like [ or ] using \.

For example, to strip < > [ ] and , characters from keys:

    filter {
      kv {
        trimkey => "<>\[\],"
      }
    }

value_split

edit
  • Value type is string
  • Default value is "="

A string of characters to use as delimiters for identifying key-value relations.

These characters form a regex character class and thus you must escape special regex characters like [ or ] using \.

For example, to identify key-values such as key1:value1 key2:value2:

    filter { kv { value_split => ":" } }