Automatic discovery of apps and services 🔗
Splunk Distribution of the OpenTelemetry Collector は、データソースを自動的に識別し、メトリクスとトレースを収集して Splunk Observability Cloud に送信することができます。オートディスカバリーとゼロコードインストルメンテーションにより、Splunk Distribution of the OpenTelemetry Collector やアプリケーションをデプロイする前に手動で設定する必要はありません。
The following diagram shows how automatic discovery finds data sources:
Automatic discovery detects and collects signal data from third-party services, such as databases and web servers. Through auto discovery the Collector automatically generates a configuration snipet that you can modify and incorporate into your existing configuration to retrieve your services』 data.
注意
Auto discovery of third-party services is only available for the Collector in Linux and Kubernetes environments.
Via automatic discovery you can also enable zero-code instrumentation of back-end applications, which allows the Collector to retrieve data from application language runtimes without having to modify the source application code or adding any new installation or configuration steps. With this option you won’t have to install and configure your instrumentation agents separately. Zero-code instrumentation using the Collector’s automatic discovery feature is only available for the Java, Node.js, and .NET language runtimes. Alternatively you can deploy zero-code instrumentation independently from the Collector for 8 back-end languages, as described in バックエンドアプリケーションをインストルメンテーションして、スパンを Splunk APM に送信する.
Supported host services and applications for automatic discovery
Automatic discovery of third-party applications is supported in Linux and Kubernetes environments for the following services:
サービス |
レシーバー |
---|---|
Apache Web Server |
Apache Web Server receiver. See Apache Web Server receiver. |
MySQL |
MySQL receiver. See MySQL レシーバー |
PostgreSQL |
PostgreSQL receiver. See PostgreSQLレシーバー |
OracleDB |
Oracle DB receiver. See Oracle Databaseレシーバー |
NGINX |
NGINX receiver. See NGINX receiver |
RabbitMQ |
RabbitMQ receiver. See RabbitMQ receiver. |
Redis |
Redis receiver. See Redis レシーバー |
MongoDB |
MongoDB receiver. See MongoDB レシーバー |
Kafka |
Kafka metrics receiver. See Kafka メトリクスレシーバー |
Microsoft SQL Server |
Microsoft SQL Server receiver. See Microsoft SQL server receiver |
Read more:
Linux: Automatic discovery for third-party services in Linux
Kubernetes: Automatic discovery for third-party services in Kubernetes
Discovered services in the UI
If you’re in a Kubernetes environment and are using the Collector version 0.109.0 or higher, log in to Splunk Observability Cloud and go to the Discovered services tab in the Data Management menu to see which of your services and third-party applications the Collector has detected and can monitor automatically.

Supported language runtimes for zero-code instrumentation
Zero-code instrumentation via the Collector’s automatic discovery is available for the following languages:
Application language |
Linux対応 |
Windows対応 |
Kubernetes対応 |
---|---|---|---|
Java |
Yes. See Zero-code instrumentation for back-end applications in Linux |
いいえ |
はい。Zero-code instrumentation for back-end applications in Kubernetes を参照してください |
.NET |
Yes. See Zero-code instrumentation for back-end applications in Linux |
はい。Zero-code instrumentation for back-end applications in Windows を参照してください |
はい。Zero-code instrumentation for back-end applications in Kubernetes を参照してください |
Node.js |
Yes. See Zero-code instrumentation for back-end applications in Linux |
いいえ |
はい。Zero-code instrumentation for back-end applications in Kubernetes を参照してください |