MCP Kafka Schema Reg

# Schema Registry Custom Metrics Summary ## 📊 **Custom `mcp_schema_registry_*` Metrics Added** The remote MCP server now exposes **14 custom Schema Registry metrics** with the prefix `mcp_schema_registry_*` that provide deep insights into Schema Registry operations and health. --- ## 🔢 **Counter Metrics (Operational Tracking)** ### **Operation Tracking** ```prometheus # Operations by type across all registries mcp_schema_registry_operations_total{operation="register_schema"} 25 mcp_schema_registry_operations_total{operation="get_schema"} 156 mcp_schema_registry_operations_total{operation="check_compatibility"} 42 mcp_schema_registry_operations_total{operation="list_subjects"} 89 mcp_schema_registry_operations_total{operation="export_schema"} 12 # Total operations per registry mcp_schema_registry_operations_by_registry_total{registry="production"} 198 mcp_schema_registry_operations_by_registry_total{registry="staging"} 114 ``` ### **Error Tracking** ```prometheus # Failed operations per registry mcp_schema_registry_errors_total{registry="production"} 3 mcp_schema_registry_errors_total{registry="staging"} 1 ``` ### **Specific Operation Counters** ```prometheus # Schema registrations mcp_schema_registry_registrations_total{registry="production"} 25 mcp_schema_registry_registrations_total{registry="staging"} 18 # Compatibility checks mcp_schema_registry_compatibility_checks_total{registry="production"} 42 mcp_schema_registry_compatibility_checks_total{registry="staging"} 28 # Schema exports mcp_schema_registry_exports_total{registry="production"} 12 mcp_schema_registry_exports_total{registry="staging"} 8 # Context-specific operations mcp_schema_registry_context_operations_total{registry="production",context="user-events"} 45 mcp_schema_registry_context_operations_total{registry="production",context="order-events"} 32 ``` --- ## 📏 **Gauge Metrics (Current State)** ### **Registry Inventory** ```prometheus # Current subjects count (real-time from Schema Registry) mcp_schema_registry_subjects{registry="production"} 45 mcp_schema_registry_subjects{registry="staging"} 23 # Current schemas count (real-time from Schema Registry) mcp_schema_registry_schemas{registry="production"} 127 mcp_schema_registry_schemas{registry="staging"} 67 # Available contexts count mcp_schema_registry_contexts{registry="production"} 3 mcp_schema_registry_contexts{registry="staging"} 2 ``` ### **Health Status** ```prometheus # Registry health (1=healthy, 0=unhealthy) mcp_schema_registry_status{registry="production"} 1 mcp_schema_registry_status{registry="staging"} 1 ``` ### **Performance Metrics** ```prometheus # Average response times mcp_schema_registry_response_time_seconds_avg{registry="production"} 0.089000 mcp_schema_registry_response_time_seconds_avg{registry="staging"} 0.045000 # Maximum response times mcp_schema_registry_response_time_seconds_max{registry="production"} 0.450000 mcp_schema_registry_response_time_seconds_max{registry="staging"} 0.200000 # Minimum response times mcp_schema_registry_response_time_seconds_min{registry="production"} 0.023000 mcp_schema_registry_response_time_seconds_min{registry="staging"} 0.015000 ``` --- ## 🚀 **Key Features** ### **📈 Easy to Obtain Stats** - ✅ **Real-time inventory**: Current subjects/schemas count - ✅ **Performance tracking**: Response time statistics - ✅ **Error monitoring**: Failed operations per registry - ✅ **Operation insights**: Most-used MCP tools - ✅ **Context tracking**: Per-context operation counts ### **🔄 Auto-Collection** - ✅ **Cached registry stats**: 5-minute TTL for performance - ✅ **Automatic instrumentation**: Metrics recorded on every operation - ✅ **Multi-registry aware**: Separate metrics per registry - ✅ **Context-aware**: Operations tracked by Schema Registry context ### **🎯 Production-Ready** - ✅ **Prometheus compatible**: Standard metric format - ✅ **Grafana dashboards**: Ready for visualization - ✅ **Alerting ready**: Critical thresholds for monitoring - ✅ **Performance optimized**: Minimal overhead on operations --- ## 📊 **Monitoring Use Cases** ### **🔍 Operations Monitoring** ```promql # Most frequently used operations rate(mcp_schema_registry_operations_total[5m]) # Error rate per registry rate(mcp_schema_registry_errors_total[5m]) / rate(mcp_schema_registry_operations_by_registry_total[5m]) # Schema registration rate rate(mcp_schema_registry_registrations_total[5m]) ``` ### **📈 Capacity Planning** ```promql # Subject growth rate deriv(mcp_schema_registry_subjects[1h]) # Schema proliferation mcp_schema_registry_schemas / mcp_schema_registry_subjects # Registry utilization mcp_schema_registry_operations_by_registry_total ``` ### **⚡ Performance Analysis** ```promql # Average response time trends mcp_schema_registry_response_time_seconds_avg # Slowest registries topk(5, mcp_schema_registry_response_time_seconds_max) # Performance comparison mcp_schema_registry_response_time_seconds_avg{registry="production"} / mcp_schema_registry_response_time_seconds_avg{registry="staging"} ``` --- ## 🚨 **Recommended Alerts** ### **Critical Alerts** ```yaml # Registry down - alert: SchemaRegistryDown expr: mcp_schema_registry_status == 0 for: 1m labels: severity: critical # High error rate - alert: SchemaRegistryHighErrorRate expr: rate(mcp_schema_registry_errors_total[5m]) / rate(mcp_schema_registry_operations_by_registry_total[5m]) > 0.1 for: 2m labels: severity: warning # Slow response times - alert: SchemaRegistrySlowResponse expr: mcp_schema_registry_response_time_seconds_avg > 1.0 for: 5m labels: severity: warning ``` ### **Capacity Alerts** ```yaml # Subject count growing rapidly - alert: SchemaRegistrySubjectGrowth expr: increase(mcp_schema_registry_subjects[1h]) > 100 for: 0m labels: severity: info # Schema proliferation - alert: SchemaRegistrySchemaProliferation expr: mcp_schema_registry_schemas / mcp_schema_registry_subjects > 10 for: 0m labels: severity: warning ``` --- ## 🔧 **Technical Implementation** ### **Metrics Collection** - **Real-time operations**: Recorded via `record_schema_operation()` method - **Registry stats**: Cached via `get_registry_stats()` with 5-minute TTL - **Performance data**: Response times tracked per operation - **Error tracking**: Failed operations automatically recorded ### **Performance Optimizations** - **Caching**: Registry stats cached to avoid excessive API calls - **Sampling**: Limited to first 50 subjects for schema counting - **Async-safe**: Thread-safe metrics collection - **Memory efficient**: Bounded metric storage with automatic cleanup ### **Integration Points** - **Health checks**: Registry connectivity tested and exposed - **MCP tools**: All 48 tools automatically instrumented - **Context awareness**: Operations tracked per Schema Registry context - **Multi-registry**: Separate metrics for each configured registry --- ## ✅ **Summary** **Added 14 custom Schema Registry metrics** that provide: 🎯 **Operational Insights**: Track register, get, compatibility, export operations 📊 **Real-time Inventory**: Current subjects, schemas, contexts count ⚡ **Performance Monitoring**: Response times (avg/min/max) per registry 🚨 **Error Tracking**: Failed operations with registry-level granularity 🏥 **Health Status**: Registry connectivity and status monitoring 📍 **Context Awareness**: Per-context operation tracking These metrics are **automatically collected**, **easy to obtain**, and **production-ready** for monitoring Schema Registry operations at scale! 🚀