Observability Service on Apache Dubbo

Observability

Mon, 01 Jan 0001 00:00:00 +0000

The most direct way to manage Dubbo is through the visual interface provided by the Dubbo control plane (dubbo-control-plane). In the previous article on Quick Start - Deploying Dubbo Applications we also used it to check the status after the service was started.

dubbo-control-plane supports visual displays, monitoring cluster status, and real-time delivery of traffic control rules.

Installing the Console

To experience the effects, we first need to install dubbo-control-plane. Here are the specific steps to install dubbo-control-plane in a Linux environment:

Querying Metrics Monitoring Indicators from Prometheus

Mon, 01 Jan 0001 00:00:00 +0000

Prerequisites

This article demonstrates how to deploy Prometheus in a Kubernetes environment and achieve monitoring data statistics and querying for the Dubbo cluster. You need to complete or possess the following:

A local or remote Kubernetes cluster
Ensure that Prometheus is correctly installed
Deploy the sample application and enable metrics collection
Use the Prometheus dashboard to query data metrics

Ensure Prometheus is Running Correctly

Verify that Prometheus has been correctly deployed

kubectl -n dubbo-system get svc prometheus-server
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
prometheus-server ClusterIP 10.109.160.254 <none> 9090/TCP 4m

Deploy Sample

kubectl apply -f https://raw.githubusercontent.com/apache/dubbo-samples/master/4-governance/dubbo-samples-metrics-spring-boot/Deployment.yml

Wait for the sample application to run normally, and confirm the application status using the following command:

Visualizing Cluster Metrics with Grafana

Mon, 01 Jan 0001 00:00:00 +0000

The recommended way to visualize metrics is to use Grafana to configure the observability monitoring dashboard for Dubbo.

Before You Begin

An accessible Kubernetes cluster
Properly installed and configured Prometheus service
Installed Grafana
Deployed Sample Application and enabled metrics collection

Confirm Component Status

Kubernetes

Ensure Prometheus is running

$ kubectl -n dubbo-system get svc prometheus
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
prometheus ClusterIP 10.0.250.230 <none> 9090/TCP 180s

Ensure Grafana is running

$ kubectl -n dubbo-system get svc grafana
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
grafana ClusterIP 10.0.244.130 <none> 3000/TCP 180s

Deploy Sample

kubectl apply -f https://raw.githubusercontent.com/apache/dubbo-samples/master/4-governance/dubbo-samples-metrics-spring-boot/Deployment.yml

Wait for the sample application to run, and confirm the application status with the following command:

Log Management

Mon, 01 Jan 0001 00:00:00 +0000

Supported Log Frameworks

Dubbo supports the following logging frameworks, and users can configure based on the logging framework actually used in their business applications.

Third-party Log Framework	Priority	Description
Log4j	Highest (used by default)	Direct adaptation of Log4j, requires adding log4j-core, log4j-api dependencies and log4j.properties
SLF4J	Second highest (currently recommended)	Supports log4j, log4j2, logback, etc. For logback, add slf4j-api, logback-classic, logback-core dependencies and logback.xml
Log4j2	Second lowest	Direct adaptation of Log4j2, requires adding log4j2-core dependency and log4j2.xml configuration
Common Logging (jcl is common logging)	Second lowest (used if neither Log4j nor SLF4J is in the project)	Rarely used in projects
JDK log	Lowest (last resort)	Rarely used in projects

Note

Regardless of the logging framework used, in addition to configuring on the Dubbo side, ensure the application includes the correct logging framework dependencies and configuration files.

Using slf4j

For Spring Boot users, enable slf4j logging by adding the following configuration to application.yaml or application.properties: