You are viewing docs on Elastic's new documentation system, currently in technical preview. For all other Elastic docs, visit elastic.co/guide.

Elastic APM

Monitor, detect, and diagnose complex application performance issues.

Version
8.4.2 (View all)
Compatible Kibana version(s)
8.4.0 or higher
Supported Serverless project types

Security
Observability
Subscription level
Basic

The APM integration installs Elasticsearch templates and ingest node pipelines for APM data.

To learn more about the APM Integration architecture, see APM Components.

Quick start

Ready to jump in? Read the APM quick start.

How to use this integration

Add the APM integration to an Elastic Agent policy to create an apm input. Any Elastic Agents set up with this policy will run an APM Server binary locally. Don't forget to configure the APM Server host, especially if it needs to be accessed from outside, like when running in Docker. Then, configure your APM agents to communicate with APM Server.

If you have Real User Monitoring (RUM) enabled, you must run Elastic Agent centrally. Otherwise, you can run it on edge machines by downloading and installing Elastic Agent on the same machines that your instrumented services run.

Data Streams

When using the APM integration, apm events are indexed into data streams. Data stream names contain the event type, service name, and a user-configurable namespace.

There is no specific recommendation for what to use as a namespace; it is intentionally flexible. You might use the environment, like production, testing, or development, or you could namespace data by business unit. It is your choice.

See APM data streams for more information.

Compatibility

The APM integration requires Kibana and Elasticsearch 7.12.x+ with at least the basic license. The APM integration version should match the Elastic Stack Major.Minor version. For example, the APM integration version 7.16.2 should be run with the Elastic Stack 7.16.x.

IMPORTANT: If you run APM Server with Elastic Agent manually in standalone mode, you must install the APM integration, otherwise the APM Server will not ingest any events.

Traces

Traces are comprised of spans and transactions.

Traces are written to traces-apm-* data streams, except for RUM traces, which are written to traces-apm.rum-*.

Exported fields

FieldDescriptionType
@timestamp
Date/time when the event originated. This is the date/time extracted from the event, typically representing when the event was generated by the source. If the event source has no original timestamp, this value is typically populated by the first time the event was received by the pipeline. Required field for all events.
date
agent.ephemeral_id
Ephemeral identifier of this agent (if one exists). This id normally changes across restarts, but agent.id does not.
keyword
agent.name
Custom name of the agent. This is a name that can be given to an agent. This can be helpful if for example two Filebeat instances are running on the same host but a human readable separation is needed on which Filebeat instance data is coming from. If no name is given, the name is often left empty.
keyword
agent.version
Version of the agent.
keyword
child.id
The ID(s) of the child event(s).
keyword
client.domain
Client domain.
keyword
client.geo.city_name
City name.
keyword
client.geo.continent_name
Name of the continent.
keyword
client.geo.country_iso_code
Country ISO code.
keyword
client.geo.country_name
Country name.
keyword
client.geo.location
Longitude and latitude.
geo_point
client.geo.region_iso_code
Region ISO code.
keyword
client.geo.region_name
Region name.
keyword
client.ip
IP address of the client (IPv4 or IPv6).
ip
client.port
Port of the client.
long
cloud.account.id
The cloud account or organization id used to identify different entities in a multi-tenant environment. Examples: AWS account id, Google Cloud ORG Id, or other unique identifier.
keyword
cloud.account.name
The cloud account name or alias used to identify different entities in a multi-tenant environment. Examples: AWS account name, Google Cloud ORG display name.
keyword
cloud.availability_zone
Availability zone in which this host, resource, or service is located.
keyword
cloud.instance.id
Instance ID of the host machine.
keyword
cloud.instance.name
Instance name of the host machine.
keyword
cloud.machine.type
Machine type of the host machine.
keyword
cloud.origin.account.id
The cloud account or organization id used to identify different entities in a multi-tenant environment.
keyword
cloud.origin.provider
Name of the cloud provider.
keyword
cloud.origin.region
Region in which this host, resource, or service is located.
keyword
cloud.origin.service.name
The cloud service name is intended to distinguish services running on different platforms within a provider.
keyword
cloud.project.id
The cloud project identifier. Examples: Google Cloud Project id, Azure Project id.
keyword
cloud.project.name
The cloud project name. Examples: Google Cloud Project name, Azure Project name.
keyword
cloud.provider
Name of the cloud provider. Example values are aws, azure, gcp, or digitalocean.
keyword
cloud.region
Region in which this host, resource, or service is located.
keyword
cloud.service.name
The cloud service name is intended to distinguish services running on different platforms within a provider, eg AWS EC2 vs Lambda, GCP GCE vs App Engine, Azure VM vs App Server. Examples: app engine, app service, cloud run, fargate, lambda.
keyword
container.id
Unique container id.
keyword
data_stream.dataset
The field can contain anything that makes sense to signify the source of the data. Examples include nginx.access, prometheus, endpoint etc. For data streams that otherwise fit, but that do not have dataset set we use the value "generic" for the dataset value. event.dataset should have the same value as data_stream.dataset. Beyond the Elasticsearch data stream naming criteria noted above, the dataset value has additional restrictions: * Must not contain - * No longer than 100 characters
constant_keyword
data_stream.namespace
A user defined namespace. Namespaces are useful to allow grouping of data. Many users already organize their indices this way, and the data stream naming scheme now provides this best practice as a default. Many users will populate this field with default. If no value is used, it falls back to default. Beyond the Elasticsearch index naming criteria noted above, namespace value has the additional restrictions: * Must not contain - * No longer than 100 characters
constant_keyword
data_stream.type
An overarching type for the data stream. Currently allowed values are "logs" and "metrics". We expect to also add "traces" and "synthetics" in the near future.
constant_keyword
destination.address
Some event destination addresses are defined ambiguously. The event will sometimes list an IP, a domain or a unix socket. You should always store the raw address in the .address field. Then it should be duplicated to .ip or .domain, depending on which one it is.
keyword
destination.ip
IP address of the destination (IPv4 or IPv6).
ip
destination.port
Port of the destination.
long
ecs.version
ECS version this event conforms to. ecs.version is a required field and must exist in all events. When querying across multiple indices -- which may conform to slightly different ECS versions -- this field lets integrations adjust to the schema version of the events.
keyword
event.outcome
This is one of four ECS Categorization Fields, and indicates the lowest level in the ECS category hierarchy. event.outcome simply denotes whether the event represents a success or a failure from the perspective of the entity that produced the event. Note that when a single transaction is described in multiple events, each event may populate different values of event.outcome, according to their perspective. Also note that in the case of a compound event (a single event that contains multiple logical events), this field should be populated with the value that best captures the overall success or failure from the perspective of the event producer. Further note that not all events will have an associated outcome. For example, this field is generally not populated for metric events, events with event.type:info, or any events for which an outcome does not make logical sense.
keyword
faas.coldstart
Boolean indicating whether the function invocation was a coldstart or not.
boolean
faas.execution
Request ID of the function invocation.
keyword
faas.id
A unique identifier of the invoked serverless function.
keyword
faas.name
The lambda function name.
keyword
faas.trigger.request_id
The ID of the origin trigger request.
keyword
faas.trigger.type
The trigger type.
keyword
faas.version
The lambda function version.
keyword
host.architecture
Operating system architecture.
keyword
host.hostname
Hostname of the host. It normally contains what the hostname command returns on the host machine.
keyword
host.ip
Host ip addresses.
ip
host.name
Name of the host. It can contain what hostname returns on Unix systems, the fully qualified domain name, or a name specified by the user. The sender decides which value to use.
keyword
host.os.platform
Operating system platform (such centos, ubuntu, windows).
keyword
http.request.headers
The canonical headers of the monitored HTTP request.
object
http.request.method
HTTP request method. Prior to ECS 1.6.0 the following guidance was provided: "The field value must be normalized to lowercase for querying." As of ECS 1.6.0, the guidance is deprecated because the original case of the method may be useful in anomaly detection. Original case will be mandated in ECS 2.0.0
keyword
http.request.referrer
Referrer for this HTTP request.
keyword
http.response.finished
Used by the Node agent to indicate when in the response life cycle an error has occurred.
boolean
http.response.headers
The canonical headers of the monitored HTTP response.
object
http.response.status_code
HTTP response status code.
long
http.version
HTTP version.
keyword
kubernetes.namespace
Kubernetes namespace
keyword
kubernetes.node.name
Kubernetes node name
keyword
kubernetes.pod.name
Kubernetes pod name
keyword
kubernetes.pod.uid
Kubernetes Pod UID
keyword
labels
Custom key/value pairs. Can be used to add meta information to events. Should not contain nested objects. All values are stored as keyword. Example: docker and k8s labels.
object
network.carrier.icc
ISO country code, eg. US
keyword
network.carrier.mcc
Mobile country code
keyword
network.carrier.mnc
Mobile network code
keyword
network.carrier.name
Carrier name, eg. Vodafone, T-Mobile, etc.
keyword
network.connection.subtype
Detailed network connection sub-type, e.g. "LTE", "CDMA"
keyword
network.connection.type
Network connection type, eg. "wifi", "cell"
keyword
numeric_labels
Custom key/value pairs. Can be used to add meta information to events. Should not contain nested objects. All values are stored as scaled_float.
object
observer.ephemeral_id
Ephemeral identifier of the APM Server.
keyword
observer.hostname
Hostname of the observer.
keyword
observer.id
Unique identifier of the APM Server.
keyword
observer.name
Custom name of the observer. This is a name that can be given to an observer. This can be helpful for example if multiple firewalls of the same model are used in an organization. If no custom name is needed, the field can be left empty.
keyword
observer.type
The type of the observer the data is coming from. There is no predefined list of observer types. Some examples are forwarder, firewall, ids, ips, proxy, poller, sensor, APM server.
keyword
observer.version
Observer version.
keyword
parent.id
The ID of the parent event.
keyword
process.args
Array of process arguments, starting with the absolute path to the executable. May be filtered to protect sensitive information.
keyword
process.pid
Process id.
long
process.ppid
Parent process' pid.
long
process.title
Process title. The proctitle, some times the same as process name. Can also be different: for example a browser setting its title to the web page currently opened.
keyword
process.title.text
Multi-field of process.title.
match_only_text
processor.event
Processor event.
keyword
processor.name
Processor name.
constant_keyword
service.environment
Identifies the environment where the service is running. If the same service runs in different environments (production, staging, QA, development, etc.), the environment can identify other instances of the same service. Can also group services and applications from the same environment.
keyword
service.framework.name
Name of the framework used.
keyword
service.framework.version
Version of the framework used.
keyword
service.language.name
Name of the programming language used.
keyword
service.language.version
Version of the programming language used.
keyword
service.name
Name of the service data is collected from. The name of the service is normally user given. This allows for distributed services that run on multiple hosts to correlate the related instances based on the name. In the case of Elasticsearch the service.name could contain the cluster name. For Beats the service.name is by default a copy of the service.type field if no name is specified.
keyword
service.node.name
Name of a service node. This allows for two nodes of the same service running on the same host to be differentiated. Therefore, service.node.name should typically be unique across nodes of a given service. In the case of Elasticsearch, the service.node.name could contain the unique node name within the Elasticsearch cluster. In cases where the service doesn't have the concept of a node name, the host name or container name can be used to distinguish running instances that make up this service. If those do not provide uniqueness (e.g. multiple instances of the service running on the same host) - the node name can be manually set.
keyword
service.origin.id
Immutable id of the service emitting this event.
keyword
service.origin.name
Immutable name of the service emitting this event.
keyword
service.origin.version
The version of the service the data was collected from.
keyword
service.runtime.name
Name of the runtime used.
keyword
service.runtime.version
Version of the runtime used.
keyword
service.target.name
Target service for which data is collected.
keyword
service.target.type
Type of the target service for which data is collected
keyword
service.version
Version of the service the data was collected from. This allows to look at a data set only for a specific version of a service.
keyword
session.id
The ID of the session to which the event belongs.
keyword
session.sequence
The sequence number of the event within the session to which the event belongs.
long
source.domain
Source domain.
keyword
source.ip
IP address of the source (IPv4 or IPv6).
ip
source.nat.ip
Translated ip of source based NAT sessions (e.g. internal client to internet) Typically connections traversing load balancers, firewalls, or routers.
ip
source.nat.port
Translated port of source based NAT sessions. (e.g. internal client to internet) Typically used with load balancers, firewalls, or routers.
long
source.port
Port of the source.
long
span.action
The specific kind of event within the sub-type represented by the span (e.g. query, connect)
keyword
span.composite.compression_strategy
The compression strategy that was used.
keyword
span.composite.count
Number of compressed spans the composite span represents.
long
span.composite.sum.us
Sum of the durations of the compressed spans, in microseconds.
long
span.db.link
Database link.
keyword
span.db.rows_affected
Number of rows affected by the database statement.
long
span.destination.service.name
Identifier for the destination service (e.g. 'http://elastic.co', 'elasticsearch', 'rabbitmq') DEPRECATED: this field will be removed in a future release
keyword
span.destination.service.resource
Identifier for the destination service resource being operated on (e.g. 'http://elastic.co:80', 'elasticsearch', 'rabbitmq/queue_name')
keyword
span.destination.service.type
Type of the destination service (e.g. 'db', 'elasticsearch'). Should typically be the same as span.type. DEPRECATED: this field will be removed in a future release
keyword
span.duration.us
Duration of the span, in microseconds.
long
span.id
Unique identifier of the span within the scope of its trace. A span represents an operation within a transaction, such as a request to another service, or a database query.
keyword
span.kind
"The kind of span: CLIENT, SERVER, PRODUCER, CONSUMER, or INTERNAL."
keyword
span.links.span.id
Unique identifier of the linked span.
keyword
span.links.trace.id
Unique identifier of the linked trace.
keyword
span.message.age.ms
Age of a message in milliseconds.
long
span.message.queue.name
Name of the message queue or topic where the message is published or received.
keyword
span.name
Generic designation of a span in the scope of a transaction.
keyword
span.subtype
A further sub-division of the type (e.g. postgresql, elasticsearch)
keyword
span.sync
Indicates whether the span was executed synchronously or asynchronously.
boolean
span.type
Keyword of specific relevance in the service's domain (eg: 'db.postgresql.query', 'template.erb', 'cache', etc).
keyword
timestamp.us
Timestamp of the event in microseconds since Unix epoch.
long
trace.id
Unique identifier of the trace. A trace groups multiple events like transactions that belong together. For example, a user request handled by multiple inter-connected services.
keyword
transaction.duration.us
Total duration of this transaction, in microseconds.
long
transaction.experience.cls
The Cumulative Layout Shift metric
scaled_float
transaction.experience.fid
The First Input Delay metric
scaled_float
transaction.experience.longtask.count
The total number of of longtasks
long
transaction.experience.longtask.max
The max longtask duration
scaled_float
transaction.experience.longtask.sum
The sum of longtask durations
scaled_float
transaction.experience.tbt
The Total Blocking Time metric
scaled_float
transaction.id
Unique identifier of the transaction within the scope of its trace. A transaction is the highest level of work measured within a service, such as a request to a server.
keyword
transaction.marks
A user-defined mapping of groups of marks in milliseconds.
object
transaction.message.age.ms
Age of a message in milliseconds.
long
transaction.message.queue.name
Name of the message queue or topic where the message is published or received.
keyword
transaction.name
Generic designation of a transaction in the scope of a single service (eg. 'GET /users/:id').
keyword
transaction.name.text
Multi-field of transaction.name.
text
transaction.result
The result of the transaction. HTTP status code for HTTP-related transactions.
keyword
transaction.sampled
Transactions that are 'sampled' will include all available information. Transactions that are not sampled will not have spans or context.
boolean
transaction.span_count.dropped
The total amount of dropped spans for this transaction.
long
transaction.type
Keyword of specific relevance in the service's domain (eg. 'request', 'backgroundjob', etc)
keyword
url.domain
Domain of the url, such as "www.elastic.co". In some cases a URL may refer to an IP and/or port directly, without a domain name. In this case, the IP address would go to the domain field. If the URL contains a literal IPv6 address enclosed by [ and ] (IETF RFC 2732), the [ and ] characters should also be captured in the domain field.
keyword
url.fragment
Portion of the url after the #, such as "top". The # is not part of the fragment.
keyword
url.full
If full URLs are important to your use case, they should be stored in url.full, whether this field is reconstructed or present in the event source.
wildcard
url.full.text
Multi-field of url.full.
match_only_text
url.original
Unmodified original url as seen in the event source. Note that in network monitoring, the observed URL may be a full URL, whereas in access logs, the URL is often just represented as a path. This field is meant to represent the URL as it was observed, complete or not.
wildcard
url.original.text
Multi-field of url.original.
match_only_text
url.path
Path of the request, such as "/search".
wildcard
url.port
Port of the request, such as 443.
long
url.query
The query field describes the query string of the request, such as "q=elasticsearch". The ? is excluded from the query string. If a URL contains no ?, there is no query field. If there is a ? but no query, the query field exists with an empty string. The exists query can be used to differentiate between the two cases.
keyword
url.scheme
Scheme of the request, such as "https". Note: The : is not part of the scheme.
keyword
user.domain
Name of the directory the user is a member of. For example, an LDAP or Active Directory domain name.
keyword
user.email
User email address.
keyword
user.id
Unique identifier of the user.
keyword
user.name
Short name or login of the user.
keyword
user.name.text
Multi-field of user.name.
match_only_text
user_agent.device.name
Name of the device.
keyword
user_agent.name
Name of the user agent.
keyword
user_agent.original
Unparsed user_agent string.
keyword
user_agent.original.text
Multi-field of user_agent.original.
match_only_text
user_agent.os.family
OS family (such as redhat, debian, freebsd, windows).
keyword
user_agent.os.full
Operating system name, including the version or code name.
keyword
user_agent.os.full.text
Multi-field of user_agent.os.full.
match_only_text
user_agent.os.kernel
Operating system kernel version as a raw string.
keyword
user_agent.os.name
Operating system name, without the version.
keyword
user_agent.os.name.text
Multi-field of user_agent.os.name.
match_only_text
user_agent.os.platform
Operating system platform (such centos, ubuntu, windows).
keyword
user_agent.os.version
Operating system version as a raw string.
keyword
user_agent.version
Version of the user agent.
keyword

Application Metrics

Application metrics are comprised of custom, application-specific metrics, basic system metrics such as CPU and memory usage, and runtime metrics such as JVM garbage collection statistics.

Application metrics are written to service-specific metrics-apm.app.*-* data streams.

Exported fields

FieldDescriptionType
@timestamp
Date/time when the event originated. This is the date/time extracted from the event, typically representing when the event was generated by the source. If the event source has no original timestamp, this value is typically populated by the first time the event was received by the pipeline. Required field for all events.
date
agent.ephemeral_id
Ephemeral identifier of this agent (if one exists). This id normally changes across restarts, but agent.id does not.
keyword
agent.name
Custom name of the agent. This is a name that can be given to an agent. This can be helpful if for example two Filebeat instances are running on the same host but a human readable separation is needed on which Filebeat instance data is coming from. If no name is given, the name is often left empty.
keyword
agent.version
Version of the agent.
keyword
client.domain
Client domain.
keyword
client.geo.city_name
City name.
keyword
client.geo.continent_name
Name of the continent.
keyword
client.geo.country_iso_code
Country ISO code.
keyword
client.geo.country_name
Country name.
keyword
client.geo.location
Longitude and latitude.
geo_point
client.geo.region_iso_code
Region ISO code.
keyword
client.geo.region_name
Region name.
keyword
client.ip
IP address of the client (IPv4 or IPv6).
ip
client.port
Port of the client.
long
cloud.account.id
The cloud account or organization id used to identify different entities in a multi-tenant environment. Examples: AWS account id, Google Cloud ORG Id, or other unique identifier.
keyword
cloud.account.name
The cloud account name or alias used to identify different entities in a multi-tenant environment. Examples: AWS account name, Google Cloud ORG display name.
keyword
cloud.availability_zone
Availability zone in which this host, resource, or service is located.
keyword
cloud.instance.id
Instance ID of the host machine.
keyword
cloud.instance.name
Instance name of the host machine.
keyword
cloud.machine.type
Machine type of the host machine.
keyword
cloud.project.id
The cloud project identifier. Examples: Google Cloud Project id, Azure Project id.
keyword
cloud.project.name
The cloud project name. Examples: Google Cloud Project name, Azure Project name.
keyword
cloud.provider
Name of the cloud provider. Example values are aws, azure, gcp, or digitalocean.
keyword
cloud.region
Region in which this host, resource, or service is located.
keyword
cloud.service.name
The cloud service name is intended to distinguish services running on different platforms within a provider, eg AWS EC2 vs Lambda, GCP GCE vs App Engine, Azure VM vs App Server. Examples: app engine, app service, cloud run, fargate, lambda.
keyword
container.id
Unique container id.
keyword
data_stream.dataset
The field can contain anything that makes sense to signify the source of the data. Examples include nginx.access, prometheus, endpoint etc. For data streams that otherwise fit, but that do not have dataset set we use the value "generic" for the dataset value. event.dataset should have the same value as data_stream.dataset. Beyond the Elasticsearch data stream naming criteria noted above, the dataset value has additional restrictions: * Must not contain - * No longer than 100 characters
constant_keyword
data_stream.namespace
A user defined namespace. Namespaces are useful to allow grouping of data. Many users already organize their indices this way, and the data stream naming scheme now provides this best practice as a default. Many users will populate this field with default. If no value is used, it falls back to default. Beyond the Elasticsearch index naming criteria noted above, namespace value has the additional restrictions: * Must not contain - * No longer than 100 characters
constant_keyword
data_stream.type
An overarching type for the data stream. Currently allowed values are "logs" and "metrics". We expect to also add "traces" and "synthetics" in the near future.
constant_keyword
destination.address
Some event destination addresses are defined ambiguously. The event will sometimes list an IP, a domain or a unix socket. You should always store the raw address in the .address field. Then it should be duplicated to .ip or .domain, depending on which one it is.
keyword
destination.ip
IP address of the destination (IPv4 or IPv6).
ip
destination.port
Port of the destination.
long
ecs.version
ECS version this event conforms to. ecs.version is a required field and must exist in all events. When querying across multiple indices -- which may conform to slightly different ECS versions -- this field lets integrations adjust to the schema version of the events.
keyword
event.outcome
This is one of four ECS Categorization Fields, and indicates the lowest level in the ECS category hierarchy. event.outcome simply denotes whether the event represents a success or a failure from the perspective of the entity that produced the event. Note that when a single transaction is described in multiple events, each event may populate different values of event.outcome, according to their perspective. Also note that in the case of a compound event (a single event that contains multiple logical events), this field should be populated with the value that best captures the overall success or failure from the perspective of the event producer. Further note that not all events will have an associated outcome. For example, this field is generally not populated for metric events, events with event.type:info, or any events for which an outcome does not make logical sense.
keyword
host.architecture
Operating system architecture.
keyword
host.hostname
Hostname of the host. It normally contains what the hostname command returns on the host machine.
keyword
host.ip
Host ip addresses.
ip
host.name
Name of the host. It can contain what hostname returns on Unix systems, the fully qualified domain name, or a name specified by the user. The sender decides which value to use.
keyword
host.os.platform
Operating system platform (such centos, ubuntu, windows).
keyword
kubernetes.namespace
Kubernetes namespace
keyword
kubernetes.node.name
Kubernetes node name
keyword
kubernetes.pod.name
Kubernetes pod name
keyword
kubernetes.pod.uid
Kubernetes Pod UID
keyword
labels
Custom key/value pairs. Can be used to add meta information to events. Should not contain nested objects. All values are stored as keyword. Example: docker and k8s labels.
object
metricset.name
Name of the set of metrics.
keyword
network.connection.type
Network connection type, eg. "wifi", "cell"
keyword
numeric_labels
Custom key/value pairs. Can be used to add meta information to events. Should not contain nested objects. All values are stored as scaled_float.
object
observer.ephemeral_id
Ephemeral identifier of the APM Server.
keyword
observer.hostname
Hostname of the observer.
keyword
observer.id
Unique identifier of the APM Server.
keyword
observer.name
Custom name of the observer. This is a name that can be given to an observer. This can be helpful for example if multiple firewalls of the same model are used in an organization. If no custom name is needed, the field can be left empty.
keyword
observer.type
The type of the observer the data is coming from. There is no predefined list of observer types. Some examples are forwarder, firewall, ids, ips, proxy, poller, sensor, APM server.
keyword
observer.version
Observer version.
keyword
process.args
Array of process arguments, starting with the absolute path to the executable. May be filtered to protect sensitive information.
keyword
process.pid
Process id.
long
process.ppid
Parent process' pid.
long
process.title
Process title. The proctitle, some times the same as process name. Can also be different: for example a browser setting its title to the web page currently opened.
keyword
process.title.text
Multi-field of process.title.
match_only_text
processor.event
Processor event.
constant_keyword
processor.name
Processor name.
constant_keyword
service.environment
Identifies the environment where the service is running. If the same service runs in different environments (production, staging, QA, development, etc.), the environment can identify other instances of the same service. Can also group services and applications from the same environment.
keyword
service.framework.name
Name of the framework used.
keyword
service.framework.version
Version of the framework used.
keyword
service.language.name
Name of the programming language used.
keyword
service.language.version
Version of the programming language used.
keyword
service.name
Name of the service data is collected from. The name of the service is normally user given. This allows for distributed services that run on multiple hosts to correlate the related instances based on the name. In the case of Elasticsearch the service.name could contain the cluster name. For Beats the service.name is by default a copy of the service.type field if no name is specified.
keyword
service.node.name
Name of a service node. This allows for two nodes of the same service running on the same host to be differentiated. Therefore, service.node.name should typically be unique across nodes of a given service. In the case of Elasticsearch, the service.node.name could contain the unique node name within the Elasticsearch cluster. In cases where the service doesn't have the concept of a node name, the host name or container name can be used to distinguish running instances that make up this service. If those do not provide uniqueness (e.g. multiple instances of the service running on the same host) - the node name can be manually set.
keyword
service.runtime.name
Name of the runtime used.
keyword
service.runtime.version
Version of the runtime used.
keyword
service.version
Version of the service the data was collected from. This allows to look at a data set only for a specific version of a service.
keyword
source.domain
Source domain.
keyword
source.ip
IP address of the source (IPv4 or IPv6).
ip
source.nat.ip
Translated ip of source based NAT sessions (e.g. internal client to internet) Typically connections traversing load balancers, firewalls, or routers.
ip
source.nat.port
Translated port of source based NAT sessions. (e.g. internal client to internet) Typically used with load balancers, firewalls, or routers.
long
source.port
Port of the source.
long
user.domain
Name of the directory the user is a member of. For example, an LDAP or Active Directory domain name.
keyword
user.email
User email address.
keyword
user.id
Unique identifier of the user.
keyword
user.name
Short name or login of the user.
keyword
user.name.text
Multi-field of user.name.
match_only_text
user_agent.device.name
Name of the device.
keyword
user_agent.name
Name of the user agent.
keyword
user_agent.original
Unparsed user_agent string.
keyword
user_agent.original.text
Multi-field of user_agent.original.
match_only_text
user_agent.os.family
OS family (such as redhat, debian, freebsd, windows).
keyword
user_agent.os.full
Operating system name, including the version or code name.
keyword
user_agent.os.full.text
Multi-field of user_agent.os.full.
match_only_text
user_agent.os.kernel
Operating system kernel version as a raw string.
keyword
user_agent.os.name
Operating system name, without the version.
keyword
user_agent.os.name.text
Multi-field of user_agent.os.name.
match_only_text
user_agent.os.platform
Operating system platform (such centos, ubuntu, windows).
keyword
user_agent.os.version
Operating system version as a raw string.
keyword
user_agent.version
Version of the user agent.
keyword

Internal Metrics

Internal metrics comprises metrics produced by Elastic APM agents and Elastic APM server for powering various Kibana charts in the APM app, such as "Time spent by span type".

Internal metrics are written to metrics-apm.internal-* data streams.

Exported fields

FieldDescriptionTypeUnitMetric Type
@timestamp
Date/time when the event originated. This is the date/time extracted from the event, typically representing when the event was generated by the source. If the event source has no original timestamp, this value is typically populated by the first time the event was received by the pipeline. Required field for all events.
date
agent.ephemeral_id
Ephemeral identifier of this agent (if one exists). This id normally changes across restarts, but agent.id does not.
keyword
agent.name
Custom name of the agent. This is a name that can be given to an agent. This can be helpful if for example two Filebeat instances are running on the same host but a human readable separation is needed on which Filebeat instance data is coming from. If no name is given, the name is often left empty.
keyword
agent.version
Version of the agent.
keyword
agent_config_applied
Value for agent_config_applied
long
client.domain
Client domain.
keyword
client.geo.city_name
City name.
keyword
client.geo.continent_name
Name of the continent.
keyword
client.geo.country_iso_code
Country ISO code.
keyword
client.geo.country_name
Country name.
keyword
client.geo.location
Longitude and latitude.
geo_point
client.geo.region_iso_code
Region ISO code.
keyword
client.geo.region_name
Region name.
keyword
client.ip
IP address of the client (IPv4 or IPv6).
ip
client.port
Port of the client.
long
cloud.account.id
The cloud account or organization id used to identify different entities in a multi-tenant environment. Examples: AWS account id, Google Cloud ORG Id, or other unique identifier.
keyword
cloud.account.name
The cloud account name or alias used to identify different entities in a multi-tenant environment. Examples: AWS account name, Google Cloud ORG display name.
keyword
cloud.availability_zone
Availability zone in which this host, resource, or service is located.
keyword
cloud.instance.id
Instance ID of the host machine.
keyword
cloud.instance.name
Instance name of the host machine.
keyword
cloud.machine.type
Machine type of the host machine.
keyword
cloud.project.id
The cloud project identifier. Examples: Google Cloud Project id, Azure Project id.
keyword
cloud.project.name
The cloud project name. Examples: Google Cloud Project name, Azure Project name.
keyword
cloud.provider
Name of the cloud provider. Example values are aws, azure, gcp, or digitalocean.
keyword
cloud.region
Region in which this host, resource, or service is located.
keyword
cloud.service.name
The cloud service name is intended to distinguish services running on different platforms within a provider, eg AWS EC2 vs Lambda, GCP GCE vs App Engine, Azure VM vs App Server. Examples: app engine, app service, cloud run, fargate, lambda.
keyword
clr.gc.count
The total number of GC collections that have occurred.
long
counter
clr.gc.gen0size
The size of the generation 0 heap.
long
byte
gauge
clr.gc.gen1size
The size of the generation 1 heap.
long
byte
gauge
clr.gc.gen2size
The size of the generation 2 heap.
long
byte
gauge
clr.gc.gen3size
The size of the generation 3 heap - also known as Large Object Heap (LOH).
long
byte
gauge
clr.gc.time
The approximate accumulated collection elapsed time in milliseconds.
long
ms
counter
container.id
Unique container id.
keyword
data_stream.dataset
The field can contain anything that makes sense to signify the source of the data. Examples include nginx.access, prometheus, endpoint etc. For data streams that otherwise fit, but that do not have dataset set we use the value "generic" for the dataset value. event.dataset should have the same value as data_stream.dataset. Beyond the Elasticsearch data stream naming criteria noted above, the dataset value has additional restrictions: * Must not contain - * No longer than 100 characters
constant_keyword
data_stream.namespace
A user defined namespace. Namespaces are useful to allow grouping of data. Many users already organize their indices this way, and the data stream naming scheme now provides this best practice as a default. Many users will populate this field with default. If no value is used, it falls back to default. Beyond the Elasticsearch index naming criteria noted above, namespace value has the additional restrictions: * Must not contain - * No longer than 100 characters
constant_keyword
data_stream.type
An overarching type for the data stream. Currently allowed values are "logs" and "metrics". We expect to also add "traces" and "synthetics" in the near future.
constant_keyword
destination.address
Some event destination addresses are defined ambiguously. The event will sometimes list an IP, a domain or a unix socket. You should always store the raw address in the .address field. Then it should be duplicated to .ip or .domain, depending on which one it is.
keyword
destination.ip
IP address of the destination (IPv4 or IPv6).
ip
destination.port
Port of the destination.
long
ecs.version
ECS version this event conforms to. ecs.version is a required field and must exist in all events. When querying across multiple indices -- which may conform to slightly different ECS versions -- this field lets integrations adjust to the schema version of the events.
keyword
event.outcome
This is one of four ECS Categorization Fields, and indicates the lowest level in the ECS category hierarchy. event.outcome simply denotes whether the event represents a success or a failure from the perspective of the entity that produced the event. Note that when a single transaction is described in multiple events, each event may populate different values of event.outcome, according to their perspective. Also note that in the case of a compound event (a single event that contains multiple logical events), this field should be populated with the value that best captures the overall success or failure from the perspective of the event producer. Further note that not all events will have an associated outcome. For example, this field is generally not populated for metric events, events with event.type:info, or any events for which an outcome does not make logical sense.
keyword
faas.billed_duration
The number of milliseconds for which the FaaS end user is billed.
long
ms
counter
faas.coldstart
Boolean indicating whether the function invocation was a coldstart or not.
boolean
faas.coldstart_duration
The number of milliseconds elapsed during cold start.
float
ms
counter
faas.duration
The total number of milliseconds elapsed during the execution of the FaaS invocation.
float
ms
counter
faas.execution
Request ID of the function invocation.
keyword
faas.id
A unique identifier of the invoked serverless function.
keyword
faas.name
The lambda function name.
keyword
faas.timeout
The FaaS invocation timeout, expressed in milliseconds.
long
ms
counter
faas.trigger.request_id
The ID of the origin trigger request.
keyword
faas.trigger.type
The trigger type.
keyword
faas.version
The lambda function version.
keyword
golang.goroutines
The number of goroutines that currently exist.
long
counter
golang.heap.allocations.active
Bytes in non-idle spans.
long
byte
gauge
golang.heap.allocations.allocated
Bytes allocated and not yet freed (same as Alloc from runtime.MemStats).
long
byte
gauge
golang.heap.allocations.frees
The number of frees.
long
counter
golang.heap.allocations.idle
Bytes in idle spans.
long
byte
gauge
golang.heap.allocations.mallocs
The number of mallocs.
long
counter
golang.heap.allocations.objects
The total number of allocated objects.
long
counter
golang.heap.allocations.total
Bytes allocated (even if freed) throughout the lifetime.
long
byte
counter
golang.heap.gc.cpu_fraction
Fraction of CPU time used by garbage collection.
scaled_float
percent
gauge
golang.heap.gc.next_gc_limit
Target heap size of the next garbage collection cycle.
long
byte
gauge
golang.heap.gc.total_count
The total number of garbage collections.
long
counter
golang.heap.gc.total_pause.ns
The total garbage collection duration in nanoseconds.
long
nanos
counter
golang.heap.system.obtained
Via HeapSys from runtime.MemStats, bytes obtained from system. heap_sys = heap_idle + heap_inuse.
long
byte
gauge
golang.heap.system.released
Bytes released to the OS.
long
byte
gauge
golang.heap.system.stack
Bytes of stack memory obtained from the OS.
long
byte
gauge
golang.heap.system.total
Total bytes obtained from system (sum of XxxSys from runtime.MemStats).
long
byte
gauge
host.architecture
Operating system architecture.
keyword
host.hostname
Hostname of the host. It normally contains what the hostname command returns on the host machine.
keyword
host.ip
Host ip addresses.
ip
host.name
Name of the host. It can contain what hostname returns on Unix systems, the fully qualified domain name, or a name specified by the user. The sender decides which value to use.
keyword
host.os.platform
Operating system platform (such centos, ubuntu, windows).
keyword
jvm.gc.alloc
long
byte
gauge
jvm.gc.count
long
gauge
jvm.gc.time
long
ms
gauge
jvm.memory.heap.committed
long
byte
gauge
jvm.memory.heap.max
long
byte
gauge
jvm.memory.heap.pool.committed
long
byte
gauge
jvm.memory.heap.pool.max
long
byte
gauge
jvm.memory.heap.pool.used
long
byte
gauge
jvm.memory.heap.used
long
byte
gauge
jvm.memory.non_heap.committed
long
byte
gauge
jvm.memory.non_heap.max
long
byte
gauge
jvm.memory.non_heap.used
long
byte
gauge
jvm.thread.count
long
gauge
kubernetes.namespace
Kubernetes namespace
keyword
kubernetes.node.name
Kubernetes node name
keyword
kubernetes.pod.name
Kubernetes pod name
keyword
kubernetes.pod.uid
Kubernetes Pod UID
keyword
labels
Custom key/value pairs. Can be used to add meta information to events. Should not contain nested objects. All values are stored as keyword. Example: docker and k8s labels.
object
metricset.name
Name of the set of metrics.
keyword
network.connection.type
Network connection type, eg. "wifi", "cell"
keyword
nodejs.eventloop.delay.avg.ms
scaled_float
ms
gauge
nodejs.handles.active
long
gauge
nodejs.memory.arrayBuffers.bytes
long
byte
gauge
nodejs.memory.external.bytes
long
byte
gauge
nodejs.memory.heap.allocated.bytes
long
byte
gauge
nodejs.memory.heap.used.bytes
long
byte
gauge
nodejs.requests.active
long
gauge
numeric_labels
Custom key/value pairs. Can be used to add meta information to events. Should not contain nested objects. All values are stored as scaled_float.
object
observer.ephemeral_id
Ephemeral identifier of the APM Server.
keyword
observer.hostname
Hostname of the observer.
keyword
observer.id
Unique identifier of the APM Server.
keyword
observer.name
Custom name of the observer. This is a name that can be given to an observer. This can be helpful for example if multiple firewalls of the same model are used in an organization. If no custom name is needed, the field can be left empty.
keyword
observer.type
The type of the observer the data is coming from. There is no predefined list of observer types. Some examples are forwarder, firewall, ids, ips, proxy, poller, sensor, APM server.
keyword
observer.version
Observer version.
keyword
process.args
Array of process arguments, starting with the absolute path to the executable. May be filtered to protect sensitive information.
keyword
process.pid
Process id.
long
process.ppid
Parent process' pid.
long
process.title
Process title. The proctitle, some times the same as process name. Can also be different: for example a browser setting its title to the web page currently opened.
keyword
process.title.text
Multi-field of process.title.
match_only_text
processor.event
Processor event.
constant_keyword
processor.name
Processor name.
constant_keyword
ruby.gc.count
long
counter
ruby.gc.time
scaled_float
counter
ruby.heap.allocations.total
long
gauge
ruby.heap.slots.free
long
gauge
ruby.heap.slots.live
long
gauge
ruby.threads
long
gauge
service.environment
Identifies the environment where the service is running. If the same service runs in different environments (production, staging, QA, development, etc.), the environment can identify other instances of the same service. Can also group services and applications from the same environment.
keyword
service.framework.name
Name of the framework used.
keyword
service.framework.version
Version of the framework used.
keyword
service.language.name
Name of the programming language used.
keyword
service.language.version
Version of the programming language used.
keyword
service.name
Name of the service data is collected from. The name of the service is normally user given. This allows for distributed services that run on multiple hosts to correlate the related instances based on the name. In the case of Elasticsearch the service.name could contain the cluster name. For Beats the service.name is by default a copy of the service.type field if no name is specified.
keyword
service.node.name
Name of a service node. This allows for two nodes of the same service running on the same host to be differentiated. Therefore, service.node.name should typically be unique across nodes of a given service. In the case of Elasticsearch, the service.node.name could contain the unique node name within the Elasticsearch cluster. In cases where the service doesn't have the concept of a node name, the host name or container name can be used to distinguish running instances that make up this service. If those do not provide uniqueness (e.g. multiple instances of the service running on the same host) - the node name can be manually set.
keyword
service.runtime.name
Name of the runtime used.
keyword
service.runtime.version
Version of the runtime used.
keyword
service.target.name
Target service for which data is collected.
keyword
service.target.type
Type of the target service for which data is collected
keyword
service.version
Version of the service the data was collected from. This allows to look at a data set only for a specific version of a service.
keyword
source.domain
Source domain.
keyword
source.ip
IP address of the source (IPv4 or IPv6).
ip
source.nat.ip
Translated ip of source based NAT sessions (e.g. internal client to internet) Typically connections traversing load balancers, firewalls, or routers.
ip
source.nat.port
Translated port of source based NAT sessions. (e.g. internal client to internet) Typically used with load balancers, firewalls, or routers.
long
source.port
Port of the source.
long
span.destination.service.resource
Identifier for the destination service resource being operated on (e.g. 'http://elastic.co:80', 'elasticsearch', 'rabbitmq/queue_name')
keyword
span.destination.service.response_time.count
Number of aggregated outgoing requests.
long
span.destination.service.response_time.sum.us
Aggregated duration of outgoing requests, in microseconds.
long
micros
span.name
Generic designation of a span in the scope of a transaction.
keyword
span.self_time.count
Number of aggregated spans.
long
span.self_time.sum.us
Aggregated span duration, excluding the time periods where a direct child was running, in microseconds.
long
micros
span.subtype
A further sub-division of the type (e.g. postgresql, elasticsearch)
keyword
span.type
Keyword of specific relevance in the service's domain (eg: 'db.postgresql.query', 'template.erb', 'cache', etc).
keyword
system.cpu.total.norm.pct
The percentage of CPU time spent by the process since the last event. This value is normalized by the number of CPU cores and it ranges from 0 to 100%.
scaled_float
percent
gauge
system.memory.actual.free
Actual free memory in bytes. It is calculated based on the OS. On Linux it consists of the free memory plus caches and buffers. On OSX it is a sum of free memory and the inactive memory. On Windows, it is equal to system.memory.free.
long
byte
gauge
system.memory.total
Total memory.
long
byte
gauge
system.process.cgroup.cpu.cfs.period.us
CFS period in microseconds.
long
micros
counter
system.process.cgroup.cpu.cfs.quota.us
CFS quota in microseconds.
long
micros
counter
system.process.cgroup.cpu.stats.periods
Number of periods seen by the CPU.
long
counter
system.process.cgroup.cpu.stats.throttled.ns
Nanoseconds spent throttled seen by the CPU.
long
nanos
counter
system.process.cgroup.cpu.stats.throttled.periods
Number of throttled periods seen by the CPU.
long
counter
system.process.cgroup.cpuacct.total.ns
Total CPU time for the current cgroup CPU in nanoseconds.
long
nanos
counter
system.process.cgroup.memory.mem.limit.bytes
Memory limit for the current cgroup slice.
long
byte
gauge
system.process.cgroup.memory.mem.usage.bytes
Memory usage by the current cgroup slice.
long
byte
gauge
system.process.cgroup.memory.stats.inactive_file.bytes
File-backed memory on inactive LRU list, in bytes.
long
byte
gauge
system.process.cpu.system.norm.pct
The percentage of CPU time spent by the process executing kernel code since the last event. This value is normalized by the number of CPU cores and it ranges from 0 to 100%.
scaled_float
percent
gauge
system.process.cpu.total.norm.pct
The percentage of CPU time spent by the process since the last event. This value is normalized by the number of CPU cores and it ranges from 0 to 100%.
scaled_float
percent
gauge
system.process.cpu.user.norm.pct
The percentage of CPU time spent by the process executing application code since the last event. This value is normalized by the number of CPU cores and it ranges from 0 to 100%.
scaled_float
percent
gauge
system.process.memory.rss.bytes
The Resident Set Size. The amount of memory the process occupied in main memory (RAM).
long
byte
gauge
system.process.memory.size
The total virtual memory the process has.
long
byte
gauge
tags
List of keywords used to tag each event.
keyword
timeseries.instance
Time series instance ID
keyword
transaction.duration.histogram
Pre-aggregated histogram of transaction durations.
histogram
transaction.name
Generic designation of a transaction in the scope of a single service (eg. 'GET /users/:id').
keyword
transaction.result
The result of the transaction. HTTP status code for HTTP-related transactions.
keyword
transaction.root
Identifies metrics for root transactions. This can be used for calculating metrics for traces.
boolean
transaction.sampled
Transactions that are 'sampled' will include all available information. Transactions that are not sampled will not have spans or context.
boolean
transaction.self_time.count
Number of aggregated transactions.
long
transaction.self_time.sum.us
Aggregated transaction duration, excluding the time periods where a direct child was running, in microseconds.
long
micros
transaction.type
Keyword of specific relevance in the service's domain (eg. 'request', 'backgroundjob', etc)
keyword
user.domain
Name of the directory the user is a member of. For example, an LDAP or Active Directory domain name.
keyword
user.email
User email address.
keyword
user.id
Unique identifier of the user.
keyword
user.name
Short name or login of the user.
keyword
user.name.text
Multi-field of user.name.
match_only_text
user_agent.device.name
Name of the device.
keyword
user_agent.name
Name of the user agent.
keyword
user_agent.original
Unparsed user_agent string.
keyword
user_agent.original.text
Multi-field of user_agent.original.
match_only_text
user_agent.os.family
OS family (such as redhat, debian, freebsd, windows).
keyword
user_agent.os.full
Operating system name, including the version or code name.
keyword
user_agent.os.full.text
Multi-field of user_agent.os.full.
match_only_text
user_agent.os.kernel
Operating system kernel version as a raw string.
keyword
user_agent.os.name
Operating system name, without the version.
keyword
user_agent.os.name.text
Multi-field of user_agent.os.name.
match_only_text
user_agent.os.platform
Operating system platform (such centos, ubuntu, windows).
keyword
user_agent.os.version
Operating system version as a raw string.
keyword
user_agent.version
Version of the user agent.
keyword

Application errors

Application errors comprises error/exception events occurring in an application.

Application errors are written to logs-apm.error.* data stream.

Exported fields

FieldDescriptionType
@timestamp
Date/time when the event originated. This is the date/time extracted from the event, typically representing when the event was generated by the source. If the event source has no original timestamp, this value is typically populated by the first time the event was received by the pipeline. Required field for all events.
date
agent.ephemeral_id
Ephemeral identifier of this agent (if one exists). This id normally changes across restarts, but agent.id does not.
keyword
agent.name
Custom name of the agent. This is a name that can be given to an agent. This can be helpful if for example two Filebeat instances are running on the same host but a human readable separation is needed on which Filebeat instance data is coming from. If no name is given, the name is often left empty.
keyword
agent.version
Version of the agent.
keyword
client.domain
Client domain.
keyword
client.geo.city_name
City name.
keyword
client.geo.continent_name
Name of the continent.
keyword
client.geo.country_iso_code
Country ISO code.
keyword
client.geo.country_name
Country name.
keyword
client.geo.location
Longitude and latitude.
geo_point
client.geo.region_iso_code
Region ISO code.
keyword
client.geo.region_name
Region name.
keyword
client.ip
IP address of the client (IPv4 or IPv6).
ip
client.port
Port of the client.
long
cloud.account.id
The cloud account or organization id used to identify different entities in a multi-tenant environment. Examples: AWS account id, Google Cloud ORG Id, or other unique identifier.
keyword
cloud.account.name
The cloud account name or alias used to identify different entities in a multi-tenant environment. Examples: AWS account name, Google Cloud ORG display name.
keyword
cloud.availability_zone
Availability zone in which this host, resource, or service is located.
keyword
cloud.instance.id
Instance ID of the host machine.
keyword
cloud.instance.name
Instance name of the host machine.
keyword
cloud.machine.type
Machine type of the host machine.
keyword
cloud.project.id
The cloud project identifier. Examples: Google Cloud Project id, Azure Project id.
keyword
cloud.project.name
The cloud project name. Examples: Google Cloud Project name, Azure Project name.
keyword
cloud.provider
Name of the cloud provider. Example values are aws, azure, gcp, or digitalocean.
keyword
cloud.region
Region in which this host, resource, or service is located.
keyword
cloud.service.name
The cloud service name is intended to distinguish services running on different platforms within a provider, eg AWS EC2 vs Lambda, GCP GCE vs App Engine, Azure VM vs App Server. Examples: app engine, app service, cloud run, fargate, lambda.
keyword
container.id
Unique container id.
keyword
data_stream.dataset
The field can contain anything that makes sense to signify the source of the data. Examples include nginx.access, prometheus, endpoint etc. For data streams that otherwise fit, but that do not have dataset set we use the value "generic" for the dataset value. event.dataset should have the same value as data_stream.dataset. Beyond the Elasticsearch data stream naming criteria noted above, the dataset value has additional restrictions: * Must not contain - * No longer than 100 characters
constant_keyword
data_stream.namespace
A user defined namespace. Namespaces are useful to allow grouping of data. Many users already organize their indices this way, and the data stream naming scheme now provides this best practice as a default. Many users will populate this field with default. If no value is used, it falls back to default. Beyond the Elasticsearch index naming criteria noted above, namespace value has the additional restrictions: * Must not contain - * No longer than 100 characters
constant_keyword
data_stream.type
An overarching type for the data stream. Currently allowed values are "logs" and "metrics". We expect to also add "traces" and "synthetics" in the near future.
constant_keyword
destination.address
Some event destination addresses are defined ambiguously. The event will sometimes list an IP, a domain or a unix socket. You should always store the raw address in the .address field. Then it should be duplicated to .ip or .domain, depending on which one it is.
keyword
destination.ip
IP address of the destination (IPv4 or IPv6).
ip
destination.port
Port of the destination.
long
ecs.version
ECS version this event conforms to. ecs.version is a required field and must exist in all events. When querying across multiple indices -- which may conform to slightly different ECS versions -- this field lets integrations adjust to the schema version of the events.
keyword
error.culprit
Function call which was the primary perpetrator of this event.
keyword
error.exception.code
The error code set when the error happened, e.g. database error code.
keyword
error.exception.handled
Indicator whether the error was caught somewhere in the code or not.
boolean
error.exception.message
The original error message.
text
error.exception.module
The module namespace of the original error.
keyword
error.exception.type
The type of the original error, e.g. the Java exception class name.
keyword
error.grouping_key
Hash of select properties of the logged error for grouping purposes.
keyword
error.grouping_name
Name to associate with an error group. Errors belonging to the same group (same grouping_key) may have differing values for grouping_name. Consumers may choose one arbitrarily.
keyword
error.id
Unique identifier for the error.
keyword
error.log.level
The severity of the record.
keyword
error.log.logger_name
The name of the logger instance used.
keyword
error.log.message
The additionally logged error message.
text
error.log.param_message
A parametrized message. E.g. 'Could not connect to %s'. The property message is still required, and should be equal to the param_message, but with placeholders replaced. In some situations the param_message is used to group errors together.
keyword
error.stack_trace
The stack trace of this error in plain text.
wildcard
error.stack_trace.text
Multi-field of error.stack_trace.
match_only_text
event.outcome
This is one of four ECS Categorization Fields, and indicates the lowest level in the ECS category hierarchy. event.outcome simply denotes whether the event represents a success or a failure from the perspective of the entity that produced the event. Note that when a single transaction is described in multiple events, each event may populate different values of event.outcome, according to their perspective. Also note that in the case of a compound event (a single event that contains multiple logical events), this field should be populated with the value that best captures the overall success or failure from the perspective of the event producer. Further note that not all events will have an associated outcome. For example, this field is generally not populated for metric events, events with event.type:info, or any events for which an outcome does not make logical sense.
keyword
host.architecture
Operating system architecture.
keyword
host.hostname
Hostname of the host. It normally contains what the hostname command returns on the host machine.
keyword
host.ip
Host ip addresses.
ip
host.name
Name of the host. It can contain what hostname returns on Unix systems, the fully qualified domain name, or a name specified by the user. The sender decides which value to use.
keyword
host.os.platform
Operating system platform (such centos, ubuntu, windows).
keyword
http.request.headers
The canonical headers of the monitored HTTP request.
object
http.request.method
HTTP request method. Prior to ECS 1.6.0 the following guidance was provided: "The field value must be normalized to lowercase for querying." As of ECS 1.6.0, the guidance is deprecated because the original case of the method may be useful in anomaly detection. Original case will be mandated in ECS 2.0.0
keyword
http.request.referrer
Referrer for this HTTP request.
keyword
http.response.finished
Used by the Node agent to indicate when in the response life cycle an error has occurred.
boolean
http.response.headers
The canonical headers of the monitored HTTP response.
object
http.response.status_code
HTTP response status code.
long
http.version
HTTP version.
keyword
kubernetes.namespace
Kubernetes namespace
keyword
kubernetes.node.name
Kubernetes node name
keyword
kubernetes.pod.name
Kubernetes pod name
keyword
kubernetes.pod.uid
Kubernetes Pod UID
keyword
labels
Custom key/value pairs. Can be used to add meta information to events. Should not contain nested objects. All values are stored as keyword. Example: docker and k8s labels.
object
message
For log events the message field contains the log message, optimized for viewing in a log viewer. For structured logs without an original message field, other fields can be concatenated to form a human-readable summary of the event. If multiple messages exist, they can be combined into one message.
match_only_text
network.carrier.icc
ISO country code, eg. US
keyword
network.carrier.mcc
Mobile country code
keyword
network.carrier.mnc
Mobile network code
keyword
network.carrier.name
Carrier name, eg. Vodafone, T-Mobile, etc.
keyword
network.connection.subtype
Detailed network connection sub-type, e.g. "LTE", "CDMA"
keyword
network.connection.type
Network connection type, eg. "wifi", "cell"
keyword
numeric_labels
Custom key/value pairs. Can be used to add meta information to events. Should not contain nested objects. All values are stored as scaled_float.
object
observer.ephemeral_id
Ephemeral identifier of the APM Server.
keyword
observer.hostname
Hostname of the observer.
keyword
observer.id
Unique identifier of the APM Server.
keyword
observer.name
Custom name of the observer. This is a name that can be given to an observer. This can be helpful for example if multiple firewalls of the same model are used in an organization. If no custom name is needed, the field can be left empty.
keyword
observer.type
The type of the observer the data is coming from. There is no predefined list of observer types. Some examples are forwarder, firewall, ids, ips, proxy, poller, sensor, APM server.
keyword
observer.version
Observer version.
keyword
parent.id
The ID of the parent event.
keyword
process.args
Array of process arguments, starting with the absolute path to the executable. May be filtered to protect sensitive information.
keyword
process.pid
Process id.
long
process.ppid
Parent process' pid.
long
process.title
Process title. The proctitle, some times the same as process name. Can also be different: for example a browser setting its title to the web page currently opened.
keyword
process.title.text
Multi-field of process.title.
match_only_text
processor.event
Processor event.
constant_keyword
processor.name
Processor name.
constant_keyword
service.environment
Identifies the environment where the service is running. If the same service runs in different environments (production, staging, QA, development, etc.), the environment can identify other instances of the same service. Can also group services and applications from the same environment.
keyword
service.framework.name
Name of the framework used.
keyword
service.framework.version
Version of the framework used.
keyword
service.language.name
Name of the programming language used.
keyword
service.language.version
Version of the programming language used.
keyword
service.name
Name of the service data is collected from. The name of the service is normally user given. This allows for distributed services that run on multiple hosts to correlate the related instances based on the name. In the case of Elasticsearch the service.name could contain the cluster name. For Beats the service.name is by default a copy of the service.type field if no name is specified.
keyword
service.node.name
Name of a service node. This allows for two nodes of the same service running on the same host to be differentiated. Therefore, service.node.name should typically be unique across nodes of a given service. In the case of Elasticsearch, the service.node.name could contain the unique node name within the Elasticsearch cluster. In cases where the service doesn't have the concept of a node name, the host name or container name can be used to distinguish running instances that make up this service. If those do not provide uniqueness (e.g. multiple instances of the service running on the same host) - the node name can be manually set.
keyword
service.runtime.name
Name of the runtime used.
keyword
service.runtime.version
Version of the runtime used.
keyword
service.version
Version of the service the data was collected from. This allows to look at a data set only for a specific version of a service.
keyword
source.domain
Source domain.
keyword
source.ip
IP address of the source (IPv4 or IPv6).
ip
source.nat.ip
Translated ip of source based NAT sessions (e.g. internal client to internet) Typically connections traversing load balancers, firewalls, or routers.
ip
source.nat.port
Translated port of source based NAT sessions. (e.g. internal client to internet) Typically used with load balancers, firewalls, or routers.
long
source.port
Port of the source.
long
span.id
Unique identifier of the span within the scope of its trace. A span represents an operation within a transaction, such as a request to another service, or a database query.
keyword
timestamp.us
Timestamp of the event in microseconds since Unix epoch.
long
trace.id
Unique identifier of the trace. A trace groups multiple events like transactions that belong together. For example, a user request handled by multiple inter-connected services.
keyword
transaction.id
Unique identifier of the transaction within the scope of its trace. A transaction is the highest level of work measured within a service, such as a request to a server.
keyword
transaction.name
Keyword of designation of a transaction in the scope of a single service, eg: 'GET /users/:id'.
keyword
transaction.sampled
Transactions that are 'sampled' will include all available information. Transactions that are not sampled will not have spans or context.
boolean
transaction.type
Keyword of specific relevance in the service's domain (eg. 'request', 'backgroundjob', etc)
keyword
url.domain
Domain of the url, such as "www.elastic.co". In some cases a URL may refer to an IP and/or port directly, without a domain name. In this case, the IP address would go to the domain field. If the URL contains a literal IPv6 address enclosed by [ and ] (IETF RFC 2732), the [ and ] characters should also be captured in the domain field.
keyword
url.fragment
Portion of the url after the #, such as "top". The # is not part of the fragment.
keyword
url.full
If full URLs are important to your use case, they should be stored in url.full, whether this field is reconstructed or present in the event source.
wildcard
url.full.text
Multi-field of url.full.
match_only_text
url.original
Unmodified original url as seen in the event source. Note that in network monitoring, the observed URL may be a full URL, whereas in access logs, the URL is often just represented as a path. This field is meant to represent the URL as it was observed, complete or not.
wildcard
url.original.text
Multi-field of url.original.
match_only_text
url.path
Path of the request, such as "/search".
wildcard
url.port
Port of the request, such as 443.
long
url.query
The query field describes the query string of the request, such as "q=elasticsearch". The ? is excluded from the query string. If a URL contains no ?, there is no query field. If there is a ? but no query, the query field exists with an empty string. The exists query can be used to differentiate between the two cases.
keyword
url.scheme
Scheme of the request, such as "https". Note: The : is not part of the scheme.
keyword
user.domain
Name of the directory the user is a member of. For example, an LDAP or Active Directory domain name.
keyword
user.email
User email address.
keyword
user.id
Unique identifier of the user.
keyword
user.name
Short name or login of the user.
keyword
user.name.text
Multi-field of user.name.
match_only_text
user_agent.device.name
Name of the device.
keyword
user_agent.name
Name of the user agent.
keyword
user_agent.original
Unparsed user_agent string.
keyword
user_agent.original.text
Multi-field of user_agent.original.
match_only_text
user_agent.os.family
OS family (such as redhat, debian, freebsd, windows).
keyword
user_agent.os.full
Operating system name, including the version or code name.
keyword
user_agent.os.full.text
Multi-field of user_agent.os.full.
match_only_text
user_agent.os.kernel
Operating system kernel version as a raw string.
keyword
user_agent.os.name
Operating system name, without the version.
keyword
user_agent.os.name.text
Multi-field of user_agent.os.name.
match_only_text
user_agent.os.platform
Operating system platform (such centos, ubuntu, windows).
keyword
user_agent.os.version
Operating system version as a raw string.
keyword
user_agent.version
Version of the user agent.
keyword

Changelog

VersionDetailsKibana version(s)

8.4.2

Bug fix View pull request
Added observer translation to data stream migration pipeline

8.4.0 or higher

8.4.0

Enhancement View pull request
Added support for dynamically mapping summary metrics

Enhancement View pull request
Added span.name to internal_metrics data stream, for service_destination metrics

Enhancement View pull request
Added field mappings for Lambda metrics to internal metrics data stream

8.4.0 or higher

8.3.3

Bug fix View pull request
Support TLSv1.3 and remove TLSv1.0 in the default TLS supported protocols

Bug fix View pull request
Update APM integration description

Enhancement View pull request
Update APM integration screenshots

8.3.0 or higher

8.3.0

Enhancement View pull request
Field mapping for target.name and target.type added to traces data stream

Enhancement View pull request
Remove the release tag

Enhancement View pull request
Added config option for pprof_enabled

Enhancement View pull request
Added field mappings for system, process, and runtime metrics to internal metrics data stream

Bug fix View pull request
Added tags field mapping to internal_metrics data stream

8.3.0 or higher

8.2.0

Enhancement View pull request
Field mapping for source.nat.ip and source.nat.port added to data streams

Enhancement View pull request
updated traces and rum_traces ingest pipelines to translate event.duration to <event>.duration.us

Enhancement View pull request
added span.links fields to traces and rum_traces data streams

Enhancement View pull request
Added field mapping for faas.name and faas.version to internal_metrics data stream

Enhancement View pull request
removed observer.version_major field

Bug fix View pull request
added field mapping for url.original to traces, rum_traces, app_logs, and error_logs

Enhancement View pull request
added field mapping for error.stack_trace to error_logs

8.2.0 or higher

8.1.0

Enhancement View pull request
Added field mapping for faas.coldstart and faas.trigger.type

Enhancement View pull request
Added field tail_sampling_enabled

Bug fix View pull request
added error.id field to error_logs data stream

Enhancement View pull request
Added field mapping for faas.id to internal_metrics data stream

8.1.0 or higher

8.0.0

Enhancement View pull request
support setting download-agent-version

Bug fix View pull request
java attacher config uses discovery-rules, not discovery_rules

Enhancement View pull request
add java_attacher support

Enhancement View pull request
the auth.anonymous.rate_limit.{event_limit,ip_limit} defaults are now the same as in the default rate_limit struct.

Enhancement View pull request
Updated fields @timestamp, data_stream.*, and labels to reference ecs

Enhancement View pull request
Ingested labels are now stored as event.{labels,numeric_labels}

Enhancement View pull request
added new traces-apm.rum and individual ILM policies per data stream

Bug fix View pull request
fixed the sampled_traces data stream's ILM policy name

Enhancement View pull request
added app_logs data stream

Enhancement View pull request
updated ingest pipelines to reject events from apm-servers newer than installed integration

Enhancement View pull request
added event.{outcome,severity} and log.level to app_logs data stream

8.0.0 or higher

7.16.1

Bug fix View pull request
Added agent_config_applied mapping to metrics-apm.internal data stream.

7.16.1 or higher

7.16.0

Enhancement View pull request
updated package version to align with stack version

Bug fix View pull request
added client.geo fields to internal_metrics

Enhancement View pull request
removed unused fields

Enhancement View pull request
changed processor.{name,event} to constant_keyword where possible

Enhancement View pull request
changed to data-stream specific ingest pipelines

Enhancement View pull request
added cluster privilege to package for stack monitoring

7.16.0 or higher

0.5.0

Enhancement View pull request
added index sorting for internal metrics

Enhancement View pull request
added histogram dynamic_template to app metrics data stream

Enhancement View pull request
removed warm phase from ILM policies

Enhancement View pull request
added privileges to tail-sampled traces data stream

Enhancement View pull request
added tail-sampling config vars

—

0.4.0

Breaking change View pull request
added anonymous auth config, replace some RUM config

Breaking change View pull request
updated to use new apm-server.auth config

—

0.3.0

Enhancement View pull request
added apm-server.url config

Enhancement View pull request
removed apm-server.kibana.api_key config

—

0.2.0

Enhancement View pull request
added support for apm-server.rum.allow_service_names

Enhancement View pull request
added support a configurable default service environment

—

0.1.0

Enhancement View pull request
initial release

—

On this page