Jaeger Integration
Import Jaeger traces into TigerOps via OTLP. Correlated RED metrics, service topology maps, and AI-powered slow span detection — all from your existing Jaeger instrumentation.
How It Works
Enable the OTLP Exporter
Configure Jaeger Collector to forward spans to TigerOps via the OTLP gRPC or HTTP endpoint. Jaeger 1.35+ supports native OTLP export with no additional components.
Configure Span Sampling
Set your head-based or tail-based sampling strategy. TigerOps accepts 100% of sampled spans and applies intelligent compression for high-volume services.
Map Services to Metric Labels
TigerOps automatically maps Jaeger service names and operation names to Prometheus metric labels, enabling RED metric computation directly from trace data.
Correlate Traces with Metrics and Logs
When a slow trace is detected, TigerOps surfaces correlated infrastructure metrics and log lines from the same time window — giving full incident context in one view.
What You Get Out of the Box
OTLP Trace Ingestion
Receive Jaeger spans via OTLP gRPC (port 4317) or OTLP HTTP (port 4318). Full OpenTelemetry semantic conventions are honored for automatic service topology mapping.
Service Dependency Maps
TigerOps builds real-time service topology graphs from trace parent-child relationships. Identify which services are on the critical path for each user-facing operation.
RED Metrics from Traces
TigerOps derives Rate, Errors, and Duration (RED) metrics from ingested spans automatically. No separate instrumentation needed for service-level SLO dashboards.
Slow Span Detection
AI-powered P99 latency baselining per operation detects slow spans before they become user-reported issues. Drill from the alert directly into the offending trace.
Cross-Signal Correlation
Trace IDs in log lines are automatically linked to Jaeger spans in TigerOps. Go from a log error directly to the distributed trace that produced it.
Trace Retention & Search
Configurable trace retention with full-text search across span tags, service names, and operation names. Find the specific trace that caused an incident from days ago.
Jaeger Collector OTLP Configuration
Configure Jaeger Collector to export spans to TigerOps via OTLP gRPC.
# jaeger-collector.yaml — OTLP export to TigerOps
# Jaeger 1.35+ supports native OTLP export
# Option A: Jaeger Collector with OTLP exporter flag
# docker run jaegertracing/jaeger-collector:1.50 \
# --collector.otlp.enabled=true \
# --reporter.grpc.host-port=otlp.ingest.atatus.net:4317 \
# --reporter.grpc.tls.enabled=true \
# --reporter.grpc.tls.cert=/etc/tls/client.crt \
# --reporter.grpc.tls.key=/etc/tls/client.key
# Option B: OpenTelemetry Collector as a bridge (recommended)
# otelcol-config.yaml:
receivers:
jaeger:
protocols:
grpc:
endpoint: 0.0.0.0:14250
thrift_http:
endpoint: 0.0.0.0:14268
thrift_compact:
endpoint: 0.0.0.0:6831
processors:
batch:
send_batch_size: 1000
timeout: 5s
memory_limiter:
limit_mib: 512
exporters:
otlp/tigerops:
endpoint: otlp.ingest.atatus.net:4317
headers:
authorization: "Bearer ${TIGEROPS_API_KEY}"
tls:
insecure: false
service:
pipelines:
traces:
receivers: [jaeger]
processors: [memory_limiter, batch]
exporters: [otlp/tigerops]Common Questions
Does TigerOps replace Jaeger or work alongside it?
TigerOps can work as the backend for Jaeger (replacing Jaeger's storage backend) or alongside it by receiving the same spans via OTLP. The OTLP approach lets you run both in parallel during evaluation.
Which Jaeger versions support OTLP export?
Jaeger Collector 1.35 and later supports native OTLP gRPC and HTTP export. For older versions, use the OpenTelemetry Collector as a bridge — configure the jaeger receiver and otlp exporter pointing to TigerOps.
Can TigerOps ingest Jaeger Thrift format spans directly?
Yes. TigerOps accepts Jaeger spans in Thrift compact protocol format via the OpenTelemetry Collector's jaeger receiver. Configure the collector to receive Thrift from your SDKs and export to TigerOps via OTLP.
How does TigerOps handle high trace volume (millions of spans/minute)?
TigerOps uses adaptive tail-based sampling on ingestion to store representative traces while controlling storage costs. You configure the sampling policy (error traces always, latency outliers always, remainder sampled at N%) in the TigerOps dashboard.
Can I query traces using TraceQL or Jaeger's query language?
TigerOps supports a superset of Jaeger's query API and a TraceQL-compatible query language. Existing Jaeger UI queries can be ported directly. TigerOps also adds metric-based trace search not available in Jaeger.
Your Jaeger Traces Deserve Better Correlation
RED metrics, service maps, and cross-signal correlation from Jaeger spans. Connect in minutes.