APM Onboarding Tutorial

This guide provides step-by-step instructions on configuring and using Coralogix Application Performance Monitoring (APM). Follow these steps to effectively onboard your services, install collectors, correlate resources, and leverage advanced features.

Select a Data & TCO Pipeline

Before setting up APM, you must select and configure a data pipeline and TCO pipeline to send us your data.

Data Pipeline

You can enjoy our APM features in multiple ways, using either of our data pipelines: Events2Metrics and Span Metrics.

We recommend configuring span sampling to get the best results using Span Metrics. This allows the user to view traces, service connections, and maps in the Coralogix platform.

Feature	E2M	SpanMetrics	Span Metrics & Span Sampling
Overview	Full content	Service Map is not displayed.

Apdex thresholds are based on your collector’s buckets. | Includes Service Map, which shows connections between services or spans based on the traces that are being sent.

Metrics are presented in APM features, except for the map.

Connections between services or spans based on the traces that are being sent are presented.

Metrics are presented in APM features, except for the map. |

TCO Pipeline

Assign the Medium (Monitoring) TCO pipeline to the traces sent to Coralogix.

APM Basic Setup

To set up APM, instrument your service and install the Otel Collector.

Service Instrumentation

To get started with Coralogix APM, generate spans from your services by instrumenting them. Choose the relevant documentation for your service’s programming language:

Otel Collector Installation

Install the OpenTelemetry Collector. The Collector is an intermediary component responsible for aggregating and transmitting data, playing a key role in the overall APM workflow. Once installed, the Coralogix Events2Metrics feature automatically converts your spans and logs to metrics to be displayed in your UI. Select from any of the following configuration options:

APM should run once your service is instrumented and the Collector is installed.

Resource Correlation

Investigate service-resource correlation within your Kubernetes environment in your APM Resources tab or Kubernetes Dashboard. This correlation relies on span tags and metric labels.

Span tags are metadata attached to requests moving through a distributed system, facilitating traceability and correlation. Each microservice adds or propagates span tags, creating a trace that visualizes the request's path, aiding in identifying performance bottlenecks and dependencies.

Metric labels are key-value pairs associated with metrics collected from Kubernetes resources. These labels offer additional context to metrics, such as CPU usage or network traffic, making organizing and analyzing data based on service names, pod names, or environments easier.

Kubernetes OpenTelemetry Extension

Coralogix recommends the Kubernetes Complete Observability integration package for comprehensive Kubernetes and application observability. Deploying the Kubernetes OpenTelemetry extension enables telemetry data collection and full service-resource correlation in your APM Resources tab or Kubernetes Dashboard, without any additional steps.

Kubernetes Resource Correlation Without the Otel Extension

If you are not using the Kubernetes Complete Observability integration package, take the following steps:

Ensure your span tags include the K8s.deployment.name label. In case of using StatefulSets or DaemonSets, replace K8s.deployment.name with K8s.DaemonSets.name or K8s.StatefulSets.name, respectively.
In cases where metrics or metric labels are missing, you will be prompted to deploy our Kubernetes Complete Observability integration package or reload the missing labels and tags.

ECS Resource Correlation

Correlate ECS-EC2 or ECS Fargate performance metrics with your service as follows:

Service-resource correlation within your Kubernetes environment is presented in your APM Resources tab.

Correlation Using Prometheus

If you are using Prometheus to send us telemetry, ensure that kube state metrics and node exporter metrics are sent by Prometheus Operator to set up metric labels for Kubernetes resource correlation in your Kubernetes Dashboard. You may validate the metrics by following these instructions.

Correlate Logs with Services

Investigate the logs produced by your service by enriching your logs with a subsystem name matching the service name. Alternatively, set up the correlation by following these instructions.