Logging

edit

You can use Elasticsearch’s application logs to monitor your cluster and diagnose issues. If you run Elasticsearch as a service, the default location of the logs varies based on your platform and installation method:

On Docker, log messages go to the console and are handled by the configured Docker logging driver. To access logs, run docker logs.

If you run Elasticsearch from the command line, Elasticsearch prints logs to the standard output (stdout).

Logging configuration

edit

Elastic strongly recommends using the Log4j 2 configuration that is shipped by default.

Elasticsearch uses Log4j 2 for logging. Log4j 2 can be configured using the log4j2.properties file. Elasticsearch exposes three properties, ${sys:es.logs.base_path}, ${sys:es.logs.cluster_name}, and ${sys:es.logs.node_name} that can be referenced in the configuration file to determine the location of the log files. The property ${sys:es.logs.base_path} will resolve to the log directory, ${sys:es.logs.cluster_name} will resolve to the cluster name (used as the prefix of log filenames in the default configuration), and ${sys:es.logs.node_name} will resolve to the node name (if the node name is explicitly set).

For example, if your log directory (path.logs) is /var/log/elasticsearch and your cluster is named production then ${sys:es.logs.base_path} will resolve to /var/log/elasticsearch and ${sys:es.logs.base_path}${sys:file.separator}${sys:es.logs.cluster_name}.log will resolve to /var/log/elasticsearch/production.log.

######## Server JSON ############################
appender.rolling.type = RollingFile 
appender.rolling.name = rolling
appender.rolling.fileName = ${sys:es.logs.base_path}${sys:file.separator}${sys:es.logs.cluster_name}_server.json 
appender.rolling.layout.type = ECSJsonLayout 
appender.rolling.layout.dataset = elasticsearch.server 
appender.rolling.filePattern = ${sys:es.logs.base_path}${sys:file.separator}${sys:es.logs.cluster_name}-%d{yyyy-MM-dd}-%i.json.gz 
appender.rolling.policies.type = Policies
appender.rolling.policies.time.type = TimeBasedTriggeringPolicy 
appender.rolling.policies.time.interval = 1 
appender.rolling.policies.time.modulate = true 
appender.rolling.policies.size.type = SizeBasedTriggeringPolicy 
appender.rolling.policies.size.size = 256MB 
appender.rolling.strategy.type = DefaultRolloverStrategy
appender.rolling.strategy.fileIndex = nomax
appender.rolling.strategy.action.type = Delete 
appender.rolling.strategy.action.basepath = ${sys:es.logs.base_path}
appender.rolling.strategy.action.condition.type = IfFileName 
appender.rolling.strategy.action.condition.glob = ${sys:es.logs.cluster_name}-* 
appender.rolling.strategy.action.condition.nested_condition.type = IfAccumulatedFileSize 
appender.rolling.strategy.action.condition.nested_condition.exceeds = 2GB 
################################################

Configure the RollingFile appender

Log to /var/log/elasticsearch/production_server.json

Use JSON layout.

dataset is a flag populating the event.dataset field in a ECSJsonLayout. It can be used to distinguish different types of logs more easily when parsing them.

Roll logs to /var/log/elasticsearch/production-yyyy-MM-dd-i.json; logs will be compressed on each roll and i will be incremented

Use a time-based roll policy

Roll logs on a daily basis

Align rolls on the day boundary (as opposed to rolling every twenty-four hours)

Using a size-based roll policy

Roll logs after 256 MB

Use a delete action when rolling logs

Only delete logs matching a file pattern

The pattern is to only delete the main logs

Only delete if we have accumulated too many compressed logs

The size condition on the compressed logs is 2 GB

######## Server -  old style pattern ###########
appender.rolling_old.type = RollingFile
appender.rolling_old.name = rolling_old
appender.rolling_old.fileName = ${sys:es.logs.base_path}${sys:file.separator}${sys:es.logs.cluster_name}_server.log 
appender.rolling_old.layout.type = PatternLayout
appender.rolling_old.layout.pattern = [%d{ISO8601}][%-5p][%-25c{1.}] [%node_name]%marker %m%n
appender.rolling_old.filePattern = ${sys:es.logs.base_path}${sys:file.separator}${sys:es.logs.cluster_name}-%d{yyyy-MM-dd}-%i.old_log.gz

The configuration for old style pattern appenders. These logs will be saved in *.log files and if archived will be in * .log.gz files. Note that these should be considered deprecated and will be removed in the future.

Log4j’s configuration parsing gets confused by any extraneous whitespace; if you copy and paste any Log4j settings on this page, or enter any Log4j configuration in general, be sure to trim any leading and trailing whitespace.

Note than you can replace .gz by .zip in appender.rolling.filePattern to compress the rolled logs using the zip format. If you remove the .gz extension then logs will not be compressed as they are rolled.

If you want to retain log files for a specified period of time, you can use a rollover strategy with a delete action.

appender.rolling.strategy.type = DefaultRolloverStrategy 
appender.rolling.strategy.action.type = Delete 
appender.rolling.strategy.action.basepath = ${sys:es.logs.base_path} 
appender.rolling.strategy.action.condition.type = IfFileName 
appender.rolling.strategy.action.condition.glob = ${sys:es.logs.cluster_name}-* 
appender.rolling.strategy.action.condition.nested_condition.type = IfLastModified 
appender.rolling.strategy.action.condition.nested_condition.age = 7D 

Configure the DefaultRolloverStrategy

Configure the Delete action for handling rollovers

The base path to the Elasticsearch logs

The condition to apply when handling rollovers

Delete files from the base path matching the glob ${sys:es.logs.cluster_name}-*; this is the glob that log files are rolled to; this is needed to only delete the rolled Elasticsearch logs but not also delete the deprecation and slow logs

A nested condition to apply to files matching the glob

Retain logs for seven days

Multiple configuration files can be loaded (in which case they will get merged) as long as they are named log4j2.properties and have the Elasticsearch config directory as an ancestor; this is useful for plugins that expose additional loggers. The logger section contains the java packages and their corresponding log level. The appender section contains the destinations for the logs. Extensive information on how to customize logging and all the supported appenders can be found on the Log4j documentation.

Configuring logging levels

edit

Log4J 2 log messages include a level field, which is one of the following (in order of increasing verbosity):

  • FATAL
  • ERROR
  • WARN
  • INFO
  • DEBUG
  • TRACE

By default Elasticsearch includes all messages at levels INFO, WARN, ERROR and FATAL in its logs, but filters out messages at levels DEBUG and TRACE. This is the recommended configuration. Do not filter out messages at INFO or higher log levels or else you may not be able to understand your cluster’s behaviour or troubleshoot common problems. Do not enable logging at levels DEBUG or TRACE unless you are following instructions elsewhere in this manual which call for more detailed logging, or you are an expert user who will be reading the Elasticsearch source code to determine the meaning of the logs.

Messages are logged by a hierarchy of loggers which matches the hierarchy of Java packages and classes in the Elasticsearch source code. Every logger has a corresponding dynamic setting which can be used to control the verbosity of its logs. The setting’s name is the fully-qualified name of the package or class, prefixed with logger..

You may set each logger’s verbosity to the name of a log level, for instance DEBUG, which means that messages from this logger at levels up to the specified one will be included in the logs. You may also use the value OFF to suppress all messages from the logger.

For example, the org.elasticsearch.discovery package contains functionality related to the discovery process, and you can control the verbosity of its logs with the logger.org.elasticsearch.discovery setting. To enable DEBUG logging for this package, use the Cluster update settings API as follows:

resp = client.cluster.put_settings(
    body={"persistent": {"logger.org.elasticsearch.discovery": "DEBUG"}},
)
print(resp)
response = client.cluster.put_settings(
  body: {
    persistent: {
      'logger.org.elasticsearch.discovery' => 'DEBUG'
    }
  }
)
puts response
PUT /_cluster/settings
{
  "persistent": {
    "logger.org.elasticsearch.discovery": "DEBUG"
  }
}

To reset this package’s log verbosity to its default level, set the logger setting to null:

resp = client.cluster.put_settings(
    body={"persistent": {"logger.org.elasticsearch.discovery": None}},
)
print(resp)
response = client.cluster.put_settings(
  body: {
    persistent: {
      'logger.org.elasticsearch.discovery' => nil
    }
  }
)
puts response
PUT /_cluster/settings
{
  "persistent": {
    "logger.org.elasticsearch.discovery": null
  }
}

Other ways to change log levels include:

  1. elasticsearch.yml:

    logger.org.elasticsearch.discovery: DEBUG

    This is most appropriate when debugging a problem on a single node.

  2. log4j2.properties:

    logger.discovery.name = org.elasticsearch.discovery
    logger.discovery.level = debug

    This is most appropriate when you already need to change your Log4j 2 configuration for other reasons. For example, you may want to send logs for a particular logger to another file. However, these use cases are rare.

Elasticsearch’s application logs are intended for humans to read and interpret. Different versions of Elasticsearch may report information in these logs in different ways, perhaps adding extra detail, removing unnecessary information, formatting the same information in different ways, renaming the logger or adjusting the log level for specific messages. Do not rely on the contents of the application logs remaining precisely the same between versions.

To prevent leaking sensitive information in logs, Elasticsearch suppresses certain log messages by default even at the highest verbosity levels. To disable this protection on a node, set the Java system property es.insecure_network_trace_enabled to true. This feature is primarily intended for test systems which do not contain any sensitive information. If you set this property on a system which contains sensitive information, you must protect your logs from unauthorized access.

Deprecation logging

edit

Elasticsearch also writes deprecation logs to the log directory. These logs record a message when you use deprecated Elasticsearch functionality. You can use the deprecation logs to update your application before upgrading Elasticsearch to a new major version.

By default, Elasticsearch rolls and compresses deprecation logs at 1GB. The default configuration preserves a maximum of five log files: four rolled logs and an active log.

Elasticsearch emits deprecation log messages at the CRITICAL level. Those messages are indicating that a used deprecation feature will be removed in a next major version. Deprecation log messages at the WARN level indicates that a less critical feature was used, it won’t be removed in next major version, but might be removed in the future.

To stop writing deprecation log messages, set logger.deprecation.level to OFF in log4j2.properties :

logger.deprecation.level = OFF

Alternatively, you can change the logging level dynamically:

resp = client.cluster.put_settings(
    body={"persistent": {"logger.org.elasticsearch.deprecation": "OFF"}},
)
print(resp)
response = client.cluster.put_settings(
  body: {
    persistent: {
      'logger.org.elasticsearch.deprecation' => 'OFF'
    }
  }
)
puts response
PUT /_cluster/settings
{
  "persistent": {
    "logger.org.elasticsearch.deprecation": "OFF"
  }
}

Refer to Configuring logging levels.

You can identify what is triggering deprecated functionality if X-Opaque-Id was used as an HTTP header. The user ID is included in the X-Opaque-ID field in deprecation JSON logs.

{
  "type": "deprecation",
  "timestamp": "2019-08-30T12:07:07,126+02:00",
  "level": "WARN",
  "component": "o.e.d.r.a.a.i.RestCreateIndexAction",
  "cluster.name": "distribution_run",
  "node.name": "node-0",
  "message": "[types removal] Using include_type_name in create index requests is deprecated. The parameter will be removed in the next major version.",
  "x-opaque-id": "MY_USER_ID",
  "cluster.uuid": "Aq-c-PAeQiK3tfBYtig9Bw",
  "node.id": "D7fUYfnfTLa2D7y-xw6tZg"
}

Deprecation logs can be indexed into .logs-deprecation.elasticsearch-default data stream cluster.deprecation_indexing.enabled setting is set to true.

Deprecation logs throttling

edit

Deprecation logs are deduplicated based on a deprecated feature key and x-opaque-id so that if a feature is repeatedly used, it will not overload the deprecation logs. This applies to both indexed deprecation logs and logs emitted to log files. You can disable the use of x-opaque-id in throttling by changing cluster.deprecation_indexing.x_opaque_id_used.enabled to false, refer to this class javadoc for more details.

JSON log format

edit

To make parsing Elasticsearch logs easier, logs are now printed in a JSON format. This is configured by a Log4J layout property appender.rolling.layout.type = ECSJsonLayout. This layout requires a dataset attribute to be set which is used to distinguish logs streams when parsing.

appender.rolling.layout.type = ECSJsonLayout
appender.rolling.layout.dataset = elasticsearch.server

Each line contains a single JSON document with the properties configured in ECSJsonLayout. See this class javadoc for more details. However if a JSON document contains an exception, it will be printed over multiple lines. The first line will contain regular properties and subsequent lines will contain the stacktrace formatted as a JSON array.

You can still use your own custom layout. To do that replace the line appender.rolling.layout.type with a different layout. See sample below:

appender.rolling.type = RollingFile
appender.rolling.name = rolling
appender.rolling.fileName = ${sys:es.logs.base_path}${sys:file.separator}${sys:es.logs.cluster_name}_server.log
appender.rolling.layout.type = PatternLayout
appender.rolling.layout.pattern = [%d{ISO8601}][%-5p][%-25c{1.}] [%node_name]%marker %.-10000m%n
appender.rolling.filePattern = ${sys:es.logs.base_path}${sys:file.separator}${sys:es.logs.cluster_name}-%d{yyyy-MM-dd}-%i.log.gz