{ "leansignal_demand_export": "1.0", "exported_at": "2026-07-17T00:00:00Z", "source_tenant": "leansignal-catalog", "catalog": { "slug": "java-otel-demand-extended", "integration": "java", "variant": "extended", "title": "Java (JVM) — Extended", "category": "SDK", "dashboards": 1, "panels": 14, "alerts": 9, "docs_url": "https://docs.leansignal.io/integrations/java/" }, "demand": { "ref": "demand", "name": "Java (JVM) — Extended (OpenTelemetry)", "description": "Deep JVM visibility: CPU, per-pool and non-heap memory, GC collection rate and time share, post-GC live set (the leak signal), and threads by state, plus the full alert set.", "dashboards": [ { "ref": "dash-java", "name": "Java (JVM) — Extended", "description": "JVM runtime health (heap/non-heap memory, GC duration, threads, classes, CPU) plus HTTP server request metrics (extended variant).", "data": { "display": { "name": "Java (JVM) — Extended" }, "panels": { "http_request_rate": { "kind": "Panel", "spec": { "display": { "name": "HTTP request rate", "description": "Inbound request throughput per instance." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "requests/sec" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance) (rate(http_server_request_duration_seconds_count{instance=~\"$instance\"}[5m]))", "seriesNameFormat": "{{instance}}" } } } } ] } }, "http_error_rate": { "kind": "Panel", "spec": { "display": { "name": "HTTP 5xx error rate", "description": "Share of requests answered 5xx." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent-decimal" }, "min": 0, "max": 1 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance) (rate(http_server_request_duration_seconds_count{instance=~\"$instance\",http_response_status_code=~\"5..\"}[5m])) / sum by (instance) (rate(http_server_request_duration_seconds_count{instance=~\"$instance\"}[5m]))", "seriesNameFormat": "{{instance}}" } } } } ] } }, "http_latency_p95": { "kind": "Panel", "spec": { "display": { "name": "HTTP request duration p95", "description": "95th-percentile request latency per instance." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "seconds" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "histogram_quantile(0.95, sum by (le, instance) (rate(http_server_request_duration_seconds_bucket{instance=~\"$instance\"}[5m])))", "seriesNameFormat": "{{instance}}" } } } } ] } }, "cpu_utilization": { "kind": "Panel", "spec": { "display": { "name": "JVM CPU utilization", "description": "Recent CPU utilization as reported by the JVM — an instantaneous gauge, volatile by nature; read it for diagnosis rather than alerting." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent-decimal" }, "min": 0, "max": 1 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "avg by (instance) (jvm_cpu_recent_utilization_ratio{instance=~\"$instance\"})", "seriesNameFormat": "{{instance}}" } } } } ] } }, "heap_utilization": { "kind": "Panel", "spec": { "display": { "name": "Heap utilization vs limit", "description": "Heap used as a share of -Xmx — the OOM headroom signal." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent-decimal" }, "min": 0, "max": 1 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance) (jvm_memory_used_bytes{instance=~\"$instance\",jvm_memory_type=\"heap\"}) / sum by (instance) (jvm_memory_limit_bytes{instance=~\"$instance\",jvm_memory_type=\"heap\"})", "seriesNameFormat": "{{instance}}" } } } } ] } }, "heap_memory_used": { "kind": "Panel", "spec": { "display": { "name": "Heap memory used", "description": "Absolute heap in use — the sawtooth is healthy GC behavior." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance) (jvm_memory_used_bytes{instance=~\"$instance\",jvm_memory_type=\"heap\"})", "seriesNameFormat": "{{instance}}" } } } } ] } }, "heap_after_gc": { "kind": "Panel", "spec": { "display": { "name": "Heap used after last GC", "description": "The live set — what survives collection. Steady growth here is the memory-leak signal." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance) (jvm_memory_used_after_last_gc_bytes{instance=~\"$instance\",jvm_memory_type=\"heap\"})", "seriesNameFormat": "{{instance}}" } } } } ] } }, "memory_used_by_pool": { "kind": "Panel", "spec": { "display": { "name": "Heap memory used by pool", "description": "Eden/survivor/old-gen split — where in the heap the memory lives." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance, jvm_memory_pool_name) (jvm_memory_used_bytes{instance=~\"$instance\",jvm_memory_type=\"heap\"})", "seriesNameFormat": "{{instance}} · {{jvm_memory_pool_name}}" } } } } ] } }, "nonheap_memory_used": { "kind": "Panel", "spec": { "display": { "name": "Non-heap memory used", "description": "Metaspace, code cache, and friends — growth outside -Xmx." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "bytes" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance) (jvm_memory_used_bytes{instance=~\"$instance\",jvm_memory_type=\"non_heap\"})", "seriesNameFormat": "{{instance}}" } } } } ] } }, "gc_pause_p95": { "kind": "Panel", "spec": { "display": { "name": "GC pause duration p95", "description": "Collection pause latency per collector." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "seconds" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "histogram_quantile(0.95, sum by (le, instance, jvm_gc_name) (rate(jvm_gc_duration_seconds_bucket{instance=~\"$instance\"}[5m])))", "seriesNameFormat": "{{instance}} · {{jvm_gc_name}}" } } } } ] } }, "gc_collection_rate": { "kind": "Panel", "spec": { "display": { "name": "GC collection rate", "description": "Collections per second, per collector." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "ops/sec" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance, jvm_gc_name) (rate(jvm_gc_duration_seconds_count{instance=~\"$instance\"}[5m]))", "seriesNameFormat": "{{instance}} · {{jvm_gc_name}}" } } } } ] } }, "gc_time_fraction": { "kind": "Panel", "spec": { "display": { "name": "Time spent in GC", "description": "GC seconds per wall-clock second (can exceed 100% with parallel collectors)." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "percent-decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance) (rate(jvm_gc_duration_seconds_sum{instance=~\"$instance\"}[5m]))", "seriesNameFormat": "{{instance}}" } } } } ] } }, "thread_count": { "kind": "Panel", "spec": { "display": { "name": "Live thread count", "description": "Platform threads alive — unbounded growth points at a thread leak or a stuck pool." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance) (jvm_thread_count{instance=~\"$instance\"})", "seriesNameFormat": "{{instance}}" } } } } ] } }, "thread_count_by_state": { "kind": "Panel", "spec": { "display": { "name": "Threads by state", "description": "Runnable vs blocked vs waiting — a pile-up of blocked threads points at a lock." }, "plugin": { "kind": "TimeSeriesChart", "spec": { "legend": { "position": "bottom", "mode": "table", "values": [ "last-number", "mean", "max" ] }, "yAxis": { "format": { "unit": "decimal" }, "min": 0 } } }, "queries": [ { "kind": "TimeSeriesQuery", "spec": { "plugin": { "kind": "PrometheusTimeSeriesQuery", "spec": { "query": "sum by (instance, jvm_thread_state) (jvm_thread_count{instance=~\"$instance\"})", "seriesNameFormat": "{{instance}} · {{jvm_thread_state}}" } } } } ] } } }, "layouts": [ { "kind": "Grid", "spec": { "display": { "title": "HTTP", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/http_request_rate" } }, { "x": 12, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/http_error_rate" } }, { "x": 0, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/http_latency_p95" } } ] } }, { "kind": "Grid", "spec": { "display": { "title": "Runtime", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/cpu_utilization" } } ] } }, { "kind": "Grid", "spec": { "display": { "title": "Memory", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/heap_utilization" } }, { "x": 12, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/heap_memory_used" } }, { "x": 0, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/heap_after_gc" } }, { "x": 12, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/memory_used_by_pool" } }, { "x": 0, "y": 16, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/nonheap_memory_used" } } ] } }, { "kind": "Grid", "spec": { "display": { "title": "GC", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/gc_pause_p95" } }, { "x": 12, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/gc_collection_rate" } }, { "x": 0, "y": 8, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/gc_time_fraction" } } ] } }, { "kind": "Grid", "spec": { "display": { "title": "Threads", "collapse": { "open": true } }, "items": [ { "x": 0, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/thread_count" } }, { "x": 12, "y": 0, "width": 12, "height": 8, "content": { "$ref": "#/spec/panels/thread_count_by_state" } } ] } } ], "variables": [ { "kind": "ListVariable", "spec": { "name": "instance", "display": { "name": "Instance" }, "allowAllValue": true, "allowMultiple": true, "customAllValue": ".*", "defaultValue": ".*", "plugin": { "kind": "PrometheusLabelValuesVariable", "spec": { "labelName": "instance" } } } } ], "duration": "1h", "refreshInterval": "30s", "datasources": {} } } ], "alert_rules": [ { "ref": "alert-http-5xx-error-rate", "name": "High HTTP 5xx error rate", "description": null, "query": "sum by (instance) (rate(http_server_request_duration_seconds_count{http_response_status_code=~\"5..\"}[5m])) / sum by (instance) (rate(http_server_request_duration_seconds_count[5m]))", "condition_op": "gt", "threshold": 0.05, "interval": "1m", "for": "5m", "severity": "critical", "message": "{{labels.instance}} is serving {{value | percent:1}} 5xx responses.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-http-latency-p95-high", "name": "High HTTP p95 latency", "description": null, "query": "histogram_quantile(0.95, sum by (le, instance) (rate(http_server_request_duration_seconds_bucket[5m])))", "condition_op": "gt", "threshold": 1, "interval": "1m", "for": "5m", "severity": "warning", "message": "{{labels.instance}} p95 request latency is {{value}}s.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-heap-near-limit", "name": "Heap memory near limit", "description": null, "query": "sum by (instance) (min_over_time(jvm_memory_used_bytes{jvm_memory_type=\"heap\"}[10m])) / sum by (instance) (jvm_memory_limit_bytes{jvm_memory_type=\"heap\"})", "condition_op": "gt", "threshold": 0.9, "interval": "1m", "for": "5m", "severity": "critical", "message": "{{labels.instance}} heap stays above {{value | percent:1}} of its limit even after GC — imminent OOM.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-gc-pause-p95-high", "name": "Elevated GC pause p95", "description": null, "query": "histogram_quantile(0.95, sum by (le, instance) (rate(jvm_gc_duration_seconds_bucket[5m])))", "condition_op": "gt", "threshold": 0.5, "interval": "1m", "for": "5m", "severity": "warning", "message": "{{labels.instance}} GC pause p95 is {{value}}s.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-heap-utilization-warning", "name": "High heap utilization", "description": null, "query": "sum by (instance) (min_over_time(jvm_memory_used_bytes{jvm_memory_type=\"heap\"}[10m])) / sum by (instance) (jvm_memory_limit_bytes{jvm_memory_type=\"heap\"})", "condition_op": "gt", "threshold": 0.8, "interval": "1m", "for": "10m", "severity": "warning", "message": "{{labels.instance}} heap stays above {{value | percent:1}} of its limit even after GC.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-gc-time-fraction-high", "name": "Excessive time in GC", "description": null, "query": "sum by (instance) (rate(jvm_gc_duration_seconds_sum[5m]))", "condition_op": "gt", "threshold": 0.1, "interval": "1m", "for": "10m", "severity": "warning", "message": "{{labels.instance}} spends {{value | percent:1}} of wall time in GC.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-http-latency-p99-high", "name": "Very high HTTP p99 latency", "description": null, "query": "histogram_quantile(0.99, sum by (le, instance) (rate(http_server_request_duration_seconds_bucket[5m])))", "condition_op": "gt", "threshold": 2.5, "interval": "1m", "for": "5m", "severity": "warning", "message": "{{labels.instance}} p99 request latency is {{value}}s.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-thread-count-high", "name": "High thread count", "description": null, "query": "sum by (instance) (jvm_thread_count)", "condition_op": "gt", "threshold": 300, "interval": "1m", "for": "10m", "severity": "warning", "message": "{{labels.instance}} has {{value}} live threads.", "notify_on_resolve": true, "paused": false, "channels": [] }, { "ref": "alert-post-gc-heap-high", "name": "Live heap high after GC (possible leak)", "description": null, "query": "sum by (instance) (jvm_memory_used_after_last_gc_bytes{jvm_memory_type=\"heap\"}) / sum by (instance) (jvm_memory_limit_bytes{jvm_memory_type=\"heap\"})", "condition_op": "gt", "threshold": 0.9, "interval": "1m", "for": "15m", "severity": "critical", "message": "{{labels.instance}} retains {{value | percent:1}} of heap even after GC (memory-leak signal).", "notify_on_resolve": true, "paused": false, "channels": [] } ] } }