⚡ OpenTelemetry & Micrometer AIOps

AI-Powered Observability & Log Analysis

OpenCrawling leverages native Spring Boot 4, OpenTelemetry (OTel) correlated spans, and Spring AI Model Context Protocol (MCP) tools to transform complex multi-threaded trace telemetry into automated Root Cause Analysis (RCA) reports.

🔍 Correlated OTel Traces

Every ingestion pipeline stage—Scanning, Extracting, Chunking, Embedding, and Indexing—generates distributed OpenTelemetry spans correlated by unique jobId.

          Tracing: Scanning (450ms) → Extracting (1200ms) → Chunking (300ms) → Embedding (1800ms) → Indexing (650ms)
        

🛠️ System MCP Tools

Exposes system-level Model Context Protocol tools exclusively to the Admin Copilot, enabling LLM agents to query live traces, exception stack traces, and Micrometer performance metrics.

          fetch_job_traces(jobId)
get_error_logs(jobId)
query_throughput_metrics(connectorId)
        

⚡ "Diagnose with AI" Action

System administrators click "Diagnose with AI" in oc-admin-ui to trigger live Root Cause Analysis reports, bottleneck insights, and automated resolution recommendations.

          Status: FAILED → RCA: Downstream Vector DB insertion timed out after 30s during pgvector batch flush.
        

AIOps Observability Architecture

graph TD subgraph Engine [OpenCrawling Engine] Scanner[Repository Connector Scan] TikaExtractor[Apache Tika Text Extractor] Chunker[Token Text Splitter] Embedder[Embedding Microservice] Writer[PgVector / Milvus Writer] end subgraph Telemetry [OpenTelemetry & Micrometer] TraceStore[TelemetryTraceStore - OTel Spans] ErrorLog[InMemoryLogAppender] Metrics[Micrometer Performance Counters] end subgraph AIOps [AIOps Diagnostic Layer] McpTools[ObservabilityMcpTools - System MCP] DiagnosticEngine[AIOpsDiagnosticService] RESTController[ObservabilityController /api/observability] end subgraph AdminUI [Admin Console - oc-admin-ui] DiagnoseBtn[Diagnose with AI Button] RCAModal[AIOps Diagnostic Modal] end Scanner -->|Span Record| TraceStore TikaExtractor -->|Span Record| TraceStore Chunker -->|Span Record| TraceStore Embedder -->|Span Record| TraceStore Writer -->|Span Record| TraceStore Writer -->|Exception Event| ErrorLog TraceStore --> McpTools ErrorLog --> McpTools Metrics --> McpTools McpTools --> DiagnosticEngine DiagnosticEngine --> RESTController DiagnoseBtn -->|GET /api/observability/diagnose/id| RESTController RESTController -->|DiagnosticReport JSON| RCAModal

Configuration & Environment Variables

The AIOps observability pipeline is enabled by default and can be configured or disabled dynamically using standard environment flags.

Environment Variable	Type	Default	Description
OPENCRAWLING_OBSERVABILITY_ENABLED	boolean	true	Globally toggle OTel distributed tracing, Kafka span propagation, and metrics.
OPENCRAWLING_OBSERVABILITY_SAMPLING_PROBABILITY	float	1.0	Configure trace sampling rate (from 0.0 to 1.0) to manage telemetry network overhead.
OTEL_EXPORTER_OTLP_TRACES_ENDPOINT	string	http://localhost:4318/v1/traces	The OTLP HTTP receiver endpoint of the target OpenTelemetry Collector.