Configuration reference

This string is used to ensure that only your agents can send data to your APM Server. You must have created the API key using the APM Server command line tool.

api_key is an alternative to secret_token. If both secret_token and api_key are configured, then api_key has precedence and secret_token is ignored.

If this configuration option is set to true the agent will collect and report breakdown metrics (span.self_time) used for "Time spent by span type" chart. Set it to false to disable the collection and reporting of breakdown metrics, which can reduce the overhead of the agent.

If this configuration option is set to true the agent will collect and report error events. Set it to false to disable the collection and reporting of APM error events, which can reduce the overhead of the agent.

Also see PHP errors as APM error events.

A comma-separated list of wildcard expressions to match instrumentation names which should be disabled. When an instrumentation is disabled, no spans will be created for that instrumentation. Each instrumentation has a name and any number of keywords. If the instrumentation’s name or any of its keywords match this configuration option then the instrumentation is disabled.

See Wildcard for more details on how to use wildcard expressions.

Supported instrumentations:

Examples:

If set to true, the agent will work as usual, except for any task requiring communication with the APM server. Events will be dropped and the agent won’t be able to receive central configuration, which means that any other configuration cannot be changed in this state without restarting the service. Example uses for this setting are: maintaining the ability to create traces and log trace/transaction/span IDs through the log correlation feature, and getting automatic distributed tracing via the W3C HTTP headers.

Setting to false will completely disable the agent.

The name of the environment this service is deployed in, e.g. "production" or "staging".

Environments allow you to easily filter data on a global level in the APM app. It’s important to be consistent when naming environments across agents. See environment selector in the Kibana UI for more information.

Labels from this configuration are added to all the entities produced by the agent.

The format is key=value[,key=value[,...]]. For example dept=engineering,rack=number8.

Any labels set by the application via the agent’s public API will override global labels with the same keys.

This option allows for the reported host name to be configured. If this option is not set the local machine’s host name is used.

A fallback configuration setting to control the logging level for the agent. Only used when a sink-specific option is not explicitly set. See Logging for details.

The logging level for stderr logging sink. See Logging for details.

The logging level for syslog logging sink. See Logging for details.

If this option is set to true then the agent creates spans for method executions based on sampling aka statistical profiler.

Due to the nature of how sampling profilers work, the duration of the inferred spans are not exact, but only estimations. See profiling_inferred_spans_sampling_interval to fine tune the trade-off between accuracy and overhead.

The minimum duration of an inferred span.

Note that effective minimum duration is also affected by the sampling interval so it is max(profiling_inferred_spans_min_duration, profiling_inferred_spans_sampling_interval)

This configuration option supports the duration suffixes: ms, s and m. For example: 100ms.

The frequency at which stack traces are gathered in order to construct inferred spans. The lower this is set, the more accurate the inferred spans durations will be. On the other hand higher accuracy comes at the expense of higher overhead and more inferred spans for potentially irrelevant operations (see profiling_inferred_spans_min_duration).

This configuration option supports the duration suffixes: ms, s and m. For example: 50ms.

This string is used to ensure that only your agents can send data to your APM Server. Both the agents and the APM Server have to be configured with the same secret token.

See the relevant APM Server’s documentation on how to configure APM Server’s secret token.

Use this setting if the APM Server requires a token, like in Elasticsearch Service.

secret_token is an alternative to api_key. If both secret_token and api_key are configured then api_key has precedence and secret_token is ignored.

If a request sending events to the APM server takes longer than the configured timeout, the request is canceled and the events are discarded.

The value has to be provided in duration format.

This option’s default unit is s (seconds).

If the value is 0 (or 0ms, 0s, etc.) the timeout for sending events to the APM Server is disabled.

Negative values are invalid and result in the default value being used instead.

The URL for your APM Server. The URL must be fully qualified, including protocol (http or https) and port.

This is used to keep all the errors and transactions of your service together and is the primary filter in the Elastic APM user interface.

If it’s set, this name is used to distinguish between different nodes of a service. If it’s not set, data aggregations will be done based on the container ID if the monitored application runs in a container. Otherwise data aggregations will be done based on the reported hostname (automatically discovered or manually configured using hostname).

The version of the currently deployed service. If your deployments are not versioned, the recommended value for this field is the commit identifier of the deployed revision, e.g., the output of git rev-parse HEAD.

Setting this option to true will enable span compression feature. Span compression reduces the collection, processing, and storage overhead, and removes clutter from the UI. The tradeoff is that some information such as DB statements of all the compressed spans will not be collected.

Consecutive spans that are exact match and that are under this threshold will be compressed into a single composite span. This option does not apply to composite spans. This reduces the collection, processing, and storage overhead, and removes clutter from the UI. The tradeoff is that the DB statements of all the compressed spans will not be collected.

Since it is max duration threshold setting this configuration option to 0 effectively disables this compression strategy because only spans with duration 0 will be considered eligible for compression with this strategy.

This configuration option supports the duration suffixes: ms, s and m. For example: 10ms. This option’s default unit is ms, so 5 is interpreted as 5ms.

Consecutive spans to the same destination that are under this threshold will be compressed into a single composite span. This option does not apply to composite spans. This reduces the collection, processing, and storage overhead, and removes clutter from the UI. The tradeoff is that the DB statements of all the compressed spans will not be collected.

Since it is max duration threshold setting this configuration option to 0 effectively disables this compression strategy because only spans with duration 0 will be considered eligible for compression with this strategy.

This configuration option supports the duration suffixes: ms, s and m. For example: 10ms. This option’s default unit is ms, so 5 is interpreted as 5ms.

While it might be very helpful to have stack trace attached to a span, collecting stack traces does have some overhead. This configuration controls the minimum span duration at which stack traces are collected. A higher value means lower overhead as stack trace collection is skipped for quick spans.

Set this config to:

This configuration option supports the duration suffixes: ms, s and m. For example: 10ms. This option’s default unit is ms, so 5 is interpreted as 5ms.

This option controls how many frames are included in stack traces captured by the agent.

Set this config to:

This option instructs the agent to ignore requests with certain URLs by not to creating transactions for those requests. It only affects automatic creation of transactions by the agent but user can still create transactions manually by using agent’s public API.

See Wildcard section for more details on how to use wildcard expressions.

This limits the amount of spans that are recorded per transaction. This is helpful in cases where a transaction creates a very high amount of spans, for example, thousands of SQL queries. Setting an upper limit helps prevent overloading the Agent and APM server in these edge cases.

If the value is 0 no spans will be collected.

Negative values are invalid and result in the default value being used instead.

By default, the agent will sample every transaction (e.g., a request to your service). To reduce overhead and storage requirements, set the sample rate to a value between 0.0 and 1.0. The agent still records the overall time and result for unsampled transactions, but not context information, labels, or spans.

By default, the agent verifies the SSL certificate if you use an HTTPS connection to the APM server. The verification can be disabled by changing this setting to false.

With this option, you can group several URL paths together by using wildcard expressions like /user/* - this way /user/Alice and /user/Bob will be mapped to transaction name /user/*.

See Wildcard section for more details on how to use wildcard expressions.