Configure logging

The Kibana logging system has three main components: loggers, appenders and layouts. These components allow us to log messages according to message type and level, to control how these messages are formatted and where the final logs will be displayed or stored.

Loggers define what logging settings should be applied to a particular logger.

Appenders define where log messages are displayed (eg. stdout or console) and stored (eg. file on the disk).

Layouts define how log messages are formatted and what type of information they include.

Currently we support the following log levels: all, fatal, error, warn, info, debug, trace, off.

Levels are ordered, so all > fatal > error > warn > info > debug > trace > off.

A log record will be logged by the logger if its level is higher than or equal to the level of its logger. Otherwise, the log record is ignored.

The all and off levels can only be used in configuration and are handy shortcuts that allow you to log every log record or disable logging entirely for a specific logger. These levels can also be specified using cli arguments.

Every appender should know exactly how to format log messages before they are written to the console or file on the disk. This behavior is controlled by the layouts and configured through appender.layout configuration property for every custom appender. Currently we don’t define any default layout for the custom appenders, so one should always make the choice explicitly.

There are two types of layout supported at the moment: pattern and json.

With pattern layout it’s possible to define a string pattern with special placeholders %conversion_pattern that will be replaced with data from the actual log message. By default the following pattern is used: [%date][%level][%logger] %message.

The conversions that are provided out of the box are:

level Outputs the level of the logging event. Example of %level output: TRACE, DEBUG, INFO.

logger Outputs the name of the logger that published the logging event. Example of %logger output: server, server.http, server.http.kibana.

message Outputs the application supplied message associated with the logging event.

meta Outputs the entries of meta object data in json format, if one is present in the event. Example of %meta output:

// Meta{from: 'v7', to: 'v8'}
'{"from":"v7","to":"v8"}'
// Meta empty object
'{}'
// no Meta provided
''

date Outputs the date of the logging event. The date conversion specifier may be followed by a set of braces containing a name of predefined date format and canonical timezone name. Timezone name is expected to be one from TZ database name. Timezone defaults to the host timezone when not explicitly specified. Example of %date output:

pid Outputs the process ID.

The pattern layout also offers a highlight option that allows you to highlight some parts of the log message with different colors. Highlighting is quite handy if log messages are forwarded to a terminal with color support.

With json layout log messages will be formatted as JSON strings in ECS format that includes a timestamp, log level, logger, message text and any other metadata that may be associated with the log message itself.

Every logger has a unique name that follows a hierarchical naming rule. The logger is considered to be an ancestor of another logger if its name followed by a . is a prefix of the descendant logger. For example, a logger named a.b is an ancestor of logger a.b.c. All top-level loggers are descendants of a special root logger at the top of the logger hierarchy. The root logger always exists, is fully configured and logs to info level by default. The root logger must also be configured if any other logging configuration is specified in your kibana.yml.

You can configure log level and appenders for a specific logger. If a logger only has a log level configured, then the appenders configuration applied to the logger is inherited from the ancestor logger, up to the root logger.

Root

The root logger has a dedicated configuration node since this logger is special and should always exist. By default root is configured with info level and default appender that is also always available. This is the configuration that all custom loggers will use unless they’re re-configured explicitly.

For example to see all log messages that fall back on the root logger configuration, just add one line to the configuration:

logging.root.level: all

Or disable logging entirely with off:

logging.root.level: off

Metrics Logs

The metrics.ops logger is configured with debug level and will automatically output sample system and process information at a regular interval. The metrics that are logged are a subset of the data collected and are formatted in the log message as follows:

The log interval is the same as the interval at which system and process information is refreshed and is configurable under ops.interval:

ops.interval: 5000

The minimum interval is 100ms and defaults to 5000ms.

Request and Response Logs

The http.server.response logger is configured with debug level and will automatically output data about http requests and responses occurring on the Kibana server. The message contains some high-level information, and the corresponding log meta contains the following:

Similar to Log4j’s RollingFileAppender, this appender will log into a file, and rotate it following a rolling strategy when the configured policy triggers.

The triggering policy determines when a rollover should occur.

There are currently two policies supported: size-limit and time-interval.

Size-limit triggering policy

This policy will rotate the file when it reaches a predetermined size.

logging:
  appenders:
    rolling-file:
      type: rolling-file
      fileName: /var/logs/kibana.log
      policy:
        type: size-limit
        size: 50mb
      strategy:
        //...
      layout:
        type: pattern

The options are:

The maximum size the log file should reach before a rollover should be performed. The default value is 100mb

Time-interval triggering policy

This policy will rotate the file every given interval of time.

logging:
  appenders:
    rolling-file:
      type: rolling-file
      fileName: /var/logs/kibana.log
      policy:
        type: time-interval
        interval: 10s
        modulate: true
      strategy:
        //...
      layout:
        type: pattern

The options are:

How often a rollover should occur. The default value is 24h

Whether the interval should be adjusted to cause the next rollover to occur on the interval boundary.

For example, if modulate is true and the interval is 4h, if the current hour is 3 am then the first rollover will occur at 4 am and then next ones will occur at 8 am, noon, 4pm, etc. The default value is true.

The rolling strategy determines how the rollover should occur: both the naming of the rolled files, and their retention policy.

There is currently one strategy supported: numeric.

Numeric rolling strategy

This strategy will suffix the file with a given pattern when rolling, and will retains a fixed amount of rolled files.

logging:
  appenders:
    rolling-file:
      type: rolling-file
      fileName: /var/logs/kibana.log
      policy:
        // ...
      strategy:
        type: numeric
        pattern: '-%i'
        max: 2
      layout:
        type: pattern

For example, with this configuration:

The options are:

The suffix to append to the file path when rolling. Must include %i, as this is the value that will be converted to the file index.

For example, with fileName: /var/logs/kibana.log and pattern: '-%i', the rolling files created will be /var/logs/kibana-1.log, /var/logs/kibana-2.log, and so on. The default value is -%i

The maximum number of files to keep. Once this number is reached, oldest files will be deleted. The default value is 7

Similar to log4j’s RewriteAppender, this appender serves as a sort of middleware, modifying the provided log events before passing them along to another appender.

logging:
  appenders:
    my-rewrite-appender:
      type: rewrite
      appenders: [console, file] # name of "destination" appender(s)
      policy:
        # ...

The most common use case for the RewriteAppender is when you want to filter or censor sensitive data that may be contained in a log entry. In fact, with a default configuration, Kibana will automatically redact any authorization, cookie, or set-cookie headers when logging http requests & responses.

To configure additional rewrite rules, you’ll need to specify a RewritePolicy.

Rewrite policies exist to indicate which parts of a log record can be modified within the rewrite appender.

Meta

The meta rewrite policy can read and modify any data contained in the LogMeta before passing it along to a destination appender.

Meta policies must specify one of three modes, which indicate which action to perform on the configured properties: - update updates an existing property at the provided path. - remove removes an existing property at the provided path.

The properties are listed as a path and value pair, where path is the dot-delimited path to the target property in the LogMeta object, and value is the value to add or update in that target property. When using the remove mode, a value is not necessary.

Here’s an example of how you would replace any cookie header values with [REDACTED]:

logging:
  appenders:
    my-rewrite-appender:
      type: rewrite
      appenders: [console]
      policy:
        type: meta # indicates that we want to rewrite the LogMeta
        mode: update # will update an existing property only
        properties:
          - path: "http.request.headers.cookie" # path to property
            value: "[REDACTED]" # value to replace at path

Rewrite appenders can even be passed to other rewrite appenders to apply multiple filter policies/modes, as long as it doesn’t create a circular reference. Each rewrite appender is applied sequentially (one after the other).

logging:
  appenders:
    remove-request-headers:
      type: rewrite
      appenders: [censor-response-headers] # redirect to the next rewrite appender
      policy:
        type: meta
        mode: remove
        properties:
          - path: "http.request.headers" # remove all request headers
    censor-response-headers:
      type: rewrite
      appenders: [console] # output to console
      policy:
        type: meta
        mode: update
        properties:
          - path: "http.response.headers.set-cookie"
            value: "[REDACTED]"

logging:
  appenders:
    custom_console:
      type: console
      layout:
        type: pattern
        highlight: true
        pattern: "[%date][%level][%logger] %message %meta"
    file:
      type: file
      fileName: ./kibana.log
      layout:
        type: json
    censor:
      type: rewrite
      appenders: [custom_console, file]
      policy:
        type: meta
        mode: update
        properties:
          - path: "http.request.headers.cookie"
            value: "[REDACTED]"
  loggers:
    - name: http.server.response
      appenders: [censor] # pass these logs to our rewrite appender
      level: debug