Distributed Tracing
8 articles from official documentation
Unlocking Tempo 3.1: Kafka Enhancements and TraceQL Metrics
Tempo 3.1 is a game-changer for observability, introducing crucial Kafka client improvements and TraceQL metrics. With new options for secure Kafka connections and the ability to redact sensitive trace data on-the-fly, this release addresses key pain points in production environments.
- →Implement TLS and SASL mechanisms for secure Kafka connections.
- →Utilize rack-aware fetching to optimize data transfer costs.
Maximize Your Trace Budget with Volumetric Sampling in Grafana Cloud
Struggling with trace budget limitations? Volumetric sampling in Grafana Cloud dynamically optimizes your sampling strategy, ensuring you capture the most relevant traces. This approach categorizes traces and adjusts sample rates intelligently, freeing you from the hassle of manual policies.
- →Specify a percentage target to let the volumetric policy manage trace sampling automatically.
- →Understand that the policy categorizes traces into buckets based on attributes, tracking frequency and sample rates.
Tempo 3.0: Unlocking Scalable Observability with TraceQL Metrics
Tempo 3.0 redefines observability with a Kafka-compatible architecture that enhances scalability and lowers total cost of ownership. Dive into how TraceQL metrics allow you to query performance data directly from traces, making troubleshooting a breeze.
- →Leverage the Kafka-compatible architecture for scalable trace ingestion and querying.
- →Utilize TraceQL metrics to query performance data directly from trace data.
Mastering Jaeger Tracing for Effective Deployment Observability
In production, tracing is crucial for understanding application performance and bottlenecks. Jaeger simplifies this with its all-in-one deployment model, combining essential components into a single container. Discover how to leverage Jaeger's capabilities and avoid common pitfalls.
- →Utilize the jaeger-query service to correct clock skew with the `--query.max-clock-skew-adjustment` parameter.
- →Deploy Jaeger using the all-in-one container for a simplified setup with essential components.
Mastering Tracing with Jaeger: Insights for Production
Tracing is essential for understanding how requests flow through your system. Jaeger provides a robust framework for capturing this data, but it requires careful setup. Learn how to leverage spans and traces effectively to enhance observability.
- →Understand spans as logical units of work with operation names, start times, and durations.
- →Utilize traces to visualize the execution path through your system as a directed acyclic graph.
Mastering OTLP Exporter Configuration for Tracing
Get your tracing data flowing smoothly with OTLP exporter configuration. Learn how to set up endpoints for traces, metrics, and logs with specific environment variables. This article dives into the details that can make or break your observability strategy.
- →Set the base endpoint using `OTEL_EXPORTER_OTLP_ENDPOINT` for all signal types.
- →Use specific endpoints for traces, metrics, logs, and profiles to optimize data flow.
Mastering Context Propagation for Effective Tracing
Context propagation is crucial for tracing in microservices, ensuring that signals from one service correlate with another. By using the W3C TraceContext specification, you can effectively manage context across service boundaries.
- →Understand context as an object that correlates signals between services.
- →Use the W3C TraceContext specification for standardized context propagation.
Mastering Async Workflow Observability in Dapr
Async workflows can be a black box, making observability a challenge. By leveraging W3C trace context, Dapr enhances the visibility of your workflows, allowing you to track execution across systems. Dive into how durable execution and context propagation can transform your monitoring strategy.
- →Enable durabletask-go to track workflow orchestrations effectively.
- →Serialize W3C context into workflow activity messages for better traceability.
OpenAI & Anthropic-compatible inference API — no GPU provisioning needed. 55+ models, pay-per-token with no minimums. VPC + zero data retention by default.
Try Serverless Inference →Get the daily digest
One email. 5 articles. Every morning.
No spam. Unsubscribe anytime.