Manage multiline messages

edit

The files harvested by Filebeat may contain messages that span multiple lines of text. For example, multiline messages are common in files that contain Java stack traces. In order to correctly handle these multiline events, you need to configure multiline settings in the filebeat.yml file to specify which lines are part of a single event.

If you are sending multiline events to Logstash, use the options described here to handle multiline events before sending the event data to Logstash. Trying to implement multiline event handling in Logstash (for example, by using the Logstash multiline codec) may result in the mixing of streams and corrupted data.

Also read Avoid YAML formatting problems and Regular expression support to avoid common mistakes.

Configuration options

edit

You can specify the following options in the filebeat.inputs section of the filebeat.yml config file to control how Filebeat deals with messages that span multiple lines.

The following example shows how to configure Filebeat to handle a multiline message where the first line of the message begins with a bracket ([).

multiline.type: pattern
multiline.pattern: '^\['
multiline.negate: true
multiline.match: after

Filebeat takes all the lines that do not start with [ and combines them with the previous line that does. For example, you could use this configuration to join the following lines of a multiline message into a single event:

[beat-logstash-some-name-832-2015.11.28] IndexNotFoundException[no such index]
    at org.elasticsearch.cluster.metadata.IndexNameExpressionResolver$WildcardExpressionResolver.resolve(IndexNameExpressionResolver.java:566)
    at org.elasticsearch.cluster.metadata.IndexNameExpressionResolver.concreteIndices(IndexNameExpressionResolver.java:133)
    at org.elasticsearch.cluster.metadata.IndexNameExpressionResolver.concreteIndices(IndexNameExpressionResolver.java:77)
    at org.elasticsearch.action.admin.indices.delete.TransportDeleteIndexAction.checkBlock(TransportDeleteIndexAction.java:75)
multiline.type
Defines which aggregation method to use. The default is pattern. The other option is count which lets you aggregate constant number of lines.
multiline.pattern
Specifies the regular expression pattern to match. Note that the regexp patterns supported by Filebeat differ somewhat from the patterns supported by Logstash. See Regular expression support for a list of supported regexp patterns. Depending on how you configure other multiline options, lines that match the specified regular expression are considered either continuations of a previous line or the start of a new multiline event. You can set the negate option to negate the pattern.
multiline.negate
Defines whether the pattern is negated. The default is false.
multiline.match

Specifies how Filebeat combines matching lines into an event. The settings are after or before. The behavior of these settings depends on what you specify for negate:

Setting for negate Setting for match Result Example pattern: ^b

false

after

Consecutive lines that match the pattern are appended to the previous line that doesn’t match.

Lines a b b c b b become "abb" and "cbb"

false

before

Consecutive lines that match the pattern are prepended to the next line that doesn’t match.

Lines b b a b b c become "bba" and "bbc"

true

after

Consecutive lines that don’t match the pattern are appended to the previous line that does match.

Lines b a c b d e become "bac" and "bde"

true

before

Consecutive lines that don’t match the pattern are prepended to the next line that does match.

Lines a c b d e b become "acb" and "deb"

The after setting is equivalent to previous in Logstash, and before is equivalent to next.

multiline.flush_pattern
Specifies a regular expression, in which the current multiline will be flushed from memory, ending the multiline-message.
multiline.max_lines
The maximum number of lines that can be combined into one event. If the multiline message contains more than max_lines, any additional lines are discarded. The default is 500.
multiline.timeout
After the specified timeout, Filebeat sends the multiline event even if no new pattern is found to start a new event. The default is 5s.
multiline.count_lines
The number of lines to aggregate into a single event.
multiline.skip_newline
When set, multiline events are concatenated without a line separator.

Examples of multiline configuration

edit

The examples in this section cover the following use cases:

  • Combining a Java stack trace into a single event
  • Combining C-style line continuations into a single event
  • Combining multiple lines from time-stamped events
Java stack traces
edit

Java stack traces consist of multiple lines, with each line after the initial line beginning with whitespace, as in this example:

Exception in thread "main" java.lang.NullPointerException
        at com.example.myproject.Book.getTitle(Book.java:16)
        at com.example.myproject.Author.getBookTitles(Author.java:25)
        at com.example.myproject.Bootstrap.main(Bootstrap.java:14)

To consolidate these lines into a single event in Filebeat, use the following multiline configuration:

multiline.type: pattern
multiline.pattern: '^[[:space:]]'
multiline.negate: false
multiline.match: after

This configuration merges any line that begins with whitespace up to the previous line.

Here’s a Java stack trace that presents a slightly more complex example:

Exception in thread "main" java.lang.IllegalStateException: A book has a null property
       at com.example.myproject.Author.getBookIds(Author.java:38)
       at com.example.myproject.Bootstrap.main(Bootstrap.java:14)
Caused by: java.lang.NullPointerException
       at com.example.myproject.Book.getId(Book.java:22)
       at com.example.myproject.Author.getBookIds(Author.java:35)
       ... 1 more

To consolidate these lines into a single event in Filebeat, use the following multiline configuration:

multiline.type: pattern
multiline.pattern: '^[[:space:]]+(at|\.{3})[[:space:]]+\b|^Caused by:'
multiline.negate: false
multiline.match: after

In this example, the pattern matches the following lines:

  • a line that begins with spaces followed by the word at or ...
  • a line that begins with the words Caused by:
Line continuations
edit

Several programming languages use the backslash (\) character at the end of a line to denote that the line continues, as in this example:

printf ("%10.10ld  \t %10.10ld \t %s\
  %f", w, x, y, z );

To consolidate these lines into a single event in Filebeat, use the following multiline configuration:

multiline.type: pattern
multiline.pattern: '\\$'
multiline.negate: false
multiline.match: before

This configuration merges any line that ends with the \ character with the line that follows.

Timestamps
edit

Activity logs from services such as Elasticsearch typically begin with a timestamp, followed by information on the specific activity, as in this example:

[2015-08-24 11:49:14,389][INFO ][env                      ] [Letha] using [1] data paths, mounts [[/
(/dev/disk1)]], net usable_space [34.5gb], net total_space [118.9gb], types [hfs]

To consolidate these lines into a single event in Filebeat, use the following multiline configuration:

multiline.type: pattern
multiline.pattern: '^\[[0-9]{4}-[0-9]{2}-[0-9]{2}'
multiline.negate: true
multiline.match: after

This configuration uses the negate: true and match: after settings to specify that any line that does not match the specified pattern belongs to the previous line.

Application events
edit

Sometimes your application logs contain events, that begin and end with custom markers, such as the following example:

[2015-08-24 11:49:14,389] Start new event
[2015-08-24 11:49:14,395] Content of processing something
[2015-08-24 11:49:14,399] End event

To consolidate this as a single event in Filebeat, use the following multiline configuration:

multiline.type: pattern
multiline.pattern: 'Start new event'
multiline.negate: true
multiline.match: after
multiline.flush_pattern: 'End event'

The flush_pattern option, specifies a regex at which the current multiline will be flushed. If you think of the pattern option specifying the beginning of an event, the flush_pattern option will specify the end or last line of the event.

This example will not work correctly if start/end log blocks are mixed with non-multiline logs, or if different start/end log blocks overlap with each other. For instance, Some other log log lines in the following example will be merged into a single multiline document because they neither match multiline.pattern nor multiline.flush_pattern, and multiline.negate is set to true.

[2015-08-24 11:49:14,389] Start new event
[2015-08-24 11:49:14,395] Content of processing something
[2015-08-24 11:49:14,399] End event
[2015-08-24 11:50:14,389] Some other log
[2015-08-24 11:50:14,395] Some other log
[2015-08-24 11:50:14,399] Some other log
[2015-08-24 11:51:14,389] Start new event
[2015-08-24 11:51:14,395] Content of processing something
[2015-08-24 11:51:14,399] End event

Test your regexp pattern for multiline

edit

To make it easier for you to test the regexp patterns in your multiline config, we’ve created a Go Playground. You can simply plug in the regexp pattern along with the multiline.negate setting that you plan to use, and paste a sample message between the content backticks (` `). Then click Run, and you’ll see which lines in the message match your specified configuration. For example:

go playground