Log Analytics Articles

Fixing Elastic Streams processing failures without dropping data
StreamsLog Analytics

Fixing Elastic Streams processing failures without dropping data

When your Streams ingest pipeline breaks, failed documents land in the failure store, not the floor. Here's how to use those exact failures to fix your pipeline without re-ingesting from the source.

Luca Wintergerst

Connecting Cursor to Production Logs via the Elastic MCP Server
APMAI AssistantLog Analytics

Connecting Cursor to Production Logs via the Elastic MCP Server

Learn how to connect Cursor to your Elastic APM data using the Elastic Agent Builder MCP server, so you can debug production errors and make UI decisions backed by real usage data without leaving your editor.

Jeffrey Rengifo

How Streams Generates a Log Pipeline in Seconds
StreamsLog AnalyticsMachine Learning

How Streams Generates a Log Pipeline in Seconds

Streams generates a complete, tested log processing pipeline from a single click. Here's the two-stage mechanism behind it: deterministic fingerprinting, a reasoning agent that iterates against real data, and hard validation thresholds that enforce quality before you see the result.

Luca Wintergerst

How to cut Elasticsearch log storage costs with LogsDB
Log AnalyticsElastic Architecture Enhancements

How to cut Elasticsearch log storage costs with LogsDB

Learn how to enable LogsDB index mode in Elasticsearch and measure real storage savings. We compare a standard index against a LogsDB index using Apache logs and show how much storage you can reclaim.

Jeffrey Rengifo

Elasticsearch over the years — how LogsDB cuts index size by up to 75% at no throughput cost
Log AnalyticsElastic Architecture Enhancements

Elasticsearch over the years — how LogsDB cuts index size by up to 75% at no throughput cost

By default, Elasticsearch is optimized for retrieval, not storage. LogsDB changes that. Here's the layered architecture behind a 77% index size reduction.

Luca Wintergerst

Automated Error Triage: From Reactive to Autonomous
Log Analytics

Automated Error Triage: From Reactive to Autonomous

Learn how to automate error triage by using Elasticsearch log clustering and AI agents, turning production logs into actionable root cause reports.

Joe Reuter

Agent Skills for Elastic Observability
OpenTelemetryLog AnalyticsMetrics

Agent Skills for Elastic Observability

Learn how Agent Skills for Elastic Observability help SREs and developers run observability workflows through natural language to instrument apps with OpenTelemetry, search logs, manage SLOs, understand service health, and help with LLM observability.

Bahubali Shetti

Process Kubernetes logs with ease using Elastic Streams
StreamsKubernetesLog Analytics

Process Kubernetes logs with ease using Elastic Streams

Learn how to process Kubernetes logs with Elastic Streams using conditional blocks, AI-generated Grok patterns, and selective drops to reduce noise and storage cost.

Luca Wintergerst

Troubleshooting Kafka-Logstash-Elasticsearch Performance Issues in delay-sensitive platforms
Log Analytics

Troubleshooting Kafka-Logstash-Elasticsearch Performance Issues in delay-sensitive platforms

Learn how to troubleshoot ingestion bottlenecks in data pipelines built with Kafka, Logstash and Elasticsearch.

Abdelwahhab-Satta

Log Processing UX Design in Elastic Streams
Log AnalyticsStreams

Log Processing UX Design in Elastic Streams

Explore log processing in Elastic Streams and the design decisions behind the Processing UX that make log data more accessible, consistent, and actionable.

Boris Kirov

Patri Pascual

Automated log parsing in Streams with ML
Log AnalyticsGenAI

Automated log parsing in Streams with ML

Learn how a hybrid ML approach achieved 94% log parsing and 91% log partitioning accuracy through automation experiments with log format fingerprinting in Streams.

Nastia Havriushenko

Streams Processing: Stop Fighting with Grok. Parse Your Logs in Streams.
Log AnalyticsGenAI

Streams Processing: Stop Fighting with Grok. Parse Your Logs in Streams.

Learn how Streams Processing works under the hood and how to use it to build, test, and deploy parsing logic on live data quickly.

Luca Wintergerst