ReadMeFeed / Community knowledge
Real questions. Useful conversations. Find the people who know your stack.
Add business context like tenantId to New Relic APM traces using the agent's request-scoped addCustomAttribute API. Covers worked examples, propagation, cardinality limits, common mistakes, and verification steps.
Mixed Lucene and KQL filters create inconsistent dashboard results. Learn why explicit KQL predicates improve reviewability and how to avoid common mapping and wildcard traps.
Running Jaeger with all-in-one memory storage works for demos but fails in production. Learn how to choose OTLP ingestion, durable backends, and sampling before scaling.
Dashboard variables from Prometheus labels are powerful, but high-cardinality labels and panel repetition can silently multiply queries. Learn how to build a curated variable chain, check cardinality before saving, and why variables are not alert boundaries.
Shared Dynatrace tenants often page the wrong team. Tag-driven management zones paired with scoped alerting profiles route problems to the right owners—if tags stay consistent.
Learn how to eliminate Prometheus dashboard timeouts by using Recording Rules to pre-compute expensive PromQL expressions and reduce query-time latency.
Attach the OpenTelemetry Java agent at JVM startup to get traces with zero code changes — here's the exact configuration, how to verify spans arrive, and the silent-failure mistakes to avoid.
Dynatrace's OneAgent auto-instruments services and stitches PurePath traces across tiers. Here's how the model works, a reproducible demo, and the trade-off versus OpenTelemetry.
Fixed sampling wastes storage on healthy traffic while missing rare errors. Jaeger's Adaptive Sampling dynamically adjusts per-operation rates via a collector feedback loop—here's how to configure it, where it breaks, and how to verify it works.
newrelic
kibana
elasticsearch
opentelemetry
jaegertracing
grafana
prometheus
dynatrace
java