mongodash 2.0.0 → 2.1.0

package/grafana/reactive_tasks.json

@@ -0,0 +1,765 @@
+{
+    "annotations": {
+        "list": [
+            {
+                "builtIn": 1,
+                "datasource": {
+                    "type": "grafana",
+                    "uid": "-- Grafana --"
+                },
+                "enable": true,
+                "hide": true,
+                "iconColor": "rgba(0, 211, 255, 1)",
+                "name": "Annotations & Alerts",
+                "type": "dashboard"
+            }
+        ]
+    },
+    "editable": true,
+    "fiscalYearStartMonth": 0,
+    "graphTooltip": 0,
+    "id": null,
+    "links": [],
+    "liveNow": false,
+    "panels": [
+        {
+            "collapsed": false,
+            "gridPos": {
+                "h": 1,
+                "w": 24,
+                "x": 0,
+                "y": 0
+            },
+            "id": 10,
+            "panels": [],
+            "title": "System Health & KPIs",
+            "type": "row"
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "description": "Percentage of tasks completed successfully vs failed.",
+            "gridPos": {
+                "h": 6,
+                "w": 4,
+                "x": 0,
+                "y": 1
+            },
+            "id": 1,
+            "options": {
+                "colorMode": "value",
+                "graphMode": "none",
+                "justifyMode": "auto",
+                "orientation": "auto",
+                "reduceOptions": {
+                    "calcs": [
+                        "lastNotNull"
+                    ],
+                    "fields": "",
+                    "values": false
+                },
+                "textMode": "auto",
+                "wideLayout": true
+            },
+            "pluginVersion": "10.0.0",
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "sum(rate(reactive_tasks_duration_seconds_count{status=\"success\"}[5m])) / sum(rate(reactive_tasks_duration_seconds_count[5m])) * 100",
+                    "legendFormat": "Health Score",
+                    "range": true,
+                    "refId": "A"
+                }
+            ],
+            "title": "Health Score",
+            "type": "stat",
+            "fieldConfig": {
+                "defaults": {
+                    "max": 100,
+                    "min": 0,
+                    "thresholds": {
+                        "mode": "absolute",
+                        "steps": [
+                            {
+                                "color": "red",
+                                "value": null
+                            },
+                            {
+                                "color": "yellow",
+                                "value": 95
+                            },
+                            {
+                                "color": "green",
+                                "value": 99
+                            }
+                        ]
+                    },
+                    "unit": "percent"
+                }
+            }
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "description": "Max age of any pending task in the system.",
+            "gridPos": {
+                "h": 6,
+                "w": 4,
+                "x": 4,
+                "y": 1
+            },
+            "id": 2,
+            "options": {
+                "colorMode": "background",
+                "graphMode": "none",
+                "justifyMode": "auto",
+                "orientation": "auto",
+                "reduceOptions": {
+                    "calcs": [
+                        "lastNotNull"
+                    ],
+                    "fields": "",
+                    "values": false
+                },
+                "textMode": "auto",
+                "wideLayout": true
+            },
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "max(reactive_tasks_global_lag_seconds)",
+                    "legendFormat": "Global Lag",
+                    "range": true,
+                    "refId": "A"
+                }
+            ],
+            "title": "Global Max Lag",
+            "type": "stat",
+            "fieldConfig": {
+                "defaults": {
+                    "thresholds": {
+                        "mode": "absolute",
+                        "steps": [
+                            {
+                                "color": "green",
+                                "value": null
+                            },
+                            {
+                                "color": "yellow",
+                                "value": 10
+                            },
+                            {
+                                "color": "red",
+                                "value": 60
+                            }
+                        ]
+                    },
+                    "unit": "s"
+                }
+            }
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "description": "Total operations per second across all tasks.",
+            "gridPos": {
+                "h": 6,
+                "w": 6,
+                "x": 8,
+                "y": 1
+            },
+            "id": 9,
+            "options": {
+                "colorMode": "value",
+                "graphMode": "area",
+                "justifyMode": "auto",
+                "orientation": "auto",
+                "reduceOptions": {
+                    "calcs": [
+                        "lastNotNull"
+                    ],
+                    "fields": "",
+                    "values": false
+                },
+                "textMode": "auto",
+                "wideLayout": true
+            },
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "sum(rate(reactive_tasks_duration_seconds_count[1m]))",
+                    "legendFormat": "Ops/sec",
+                    "range": true,
+                    "refId": "A"
+                }
+            ],
+            "title": "Total Throughput",
+            "type": "stat",
+            "fieldConfig": {
+                "defaults": {
+                    "unit": "ops"
+                }
+            }
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "description": "Number of tasks currently being processed (status=processing).",
+            "gridPos": {
+                "h": 6,
+                "w": 5,
+                "x": 14,
+                "y": 1
+            },
+            "id": 11,
+            "options": {
+                "colorMode": "value",
+                "graphMode": "none",
+                "justifyMode": "auto",
+                "orientation": "auto",
+                "reduceOptions": {
+                    "calcs": [
+                        "lastNotNull"
+                    ],
+                    "fields": "",
+                    "values": false
+                },
+                "textMode": "auto",
+                "wideLayout": true
+            },
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "sum(reactive_tasks_queue_depth{status=\"processing\"})",
+                    "legendFormat": "Active Workers",
+                    "range": true,
+                    "refId": "A"
+                }
+            ],
+            "title": "Active Workers",
+            "type": "stat",
+            "fieldConfig": {
+                "defaults": {
+                    "color": {
+                        "mode": "palette-classic"
+                    }
+                }
+            }
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "description": "Total number of tasks waiting to be picked up (status=pending).",
+            "gridPos": {
+                "h": 6,
+                "w": 5,
+                "x": 19,
+                "y": 1
+            },
+            "id": 12,
+            "options": {
+                "colorMode": "value",
+                "graphMode": "none",
+                "justifyMode": "auto",
+                "orientation": "auto",
+                "reduceOptions": {
+                    "calcs": [
+                        "lastNotNull"
+                    ],
+                    "fields": "",
+                    "values": false
+                },
+                "textMode": "auto",
+                "wideLayout": true
+            },
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "sum(reactive_tasks_queue_depth{status=\"pending\"})",
+                    "legendFormat": "Pending Tasks",
+                    "range": true,
+                    "refId": "A"
+                }
+            ],
+            "title": "Pending Queue",
+            "type": "stat",
+            "fieldConfig": {
+                "defaults": {
+                    "color": {
+                        "mode": "palette-classic"
+                    },
+                    "thresholds": {
+                        "mode": "absolute",
+                        "steps": [
+                            {
+                                "color": "green",
+                                "value": null
+                            },
+                            {
+                                "color": "orange",
+                                "value": 100
+                            },
+                            {
+                                "color": "red",
+                                "value": 1000
+                            }
+                        ]
+                    }
+                }
+            }
+        },
+        {
+            "collapsed": false,
+            "gridPos": {
+                "h": 1,
+                "w": 24,
+                "x": 0,
+                "y": 7
+            },
+            "id": 13,
+            "panels": [],
+            "title": "Performance Analysis",
+            "type": "row"
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "gridPos": {
+                "h": 8,
+                "w": 12,
+                "x": 0,
+                "y": 8
+            },
+            "id": 3,
+            "options": {
+                "calculate": false,
+                "cellGap": 1,
+                "color": {
+                    "mode": "scheme",
+                    "scheme": "Spectral",
+                    "exponent": 0.5,
+                    "steps": 64
+                },
+                "exemplars": {
+                    "color": "rgba(255,0,255,0.7)"
+                },
+                "filterValues": {
+                    "le": 1e-9
+                },
+                "legend": {
+                    "show": true
+                },
+                "tooltip": {
+                    "show": true,
+                    "yHistogram": false
+                },
+                "yAxis": {
+                    "axisPlacement": "left",
+                    "reverse": false,
+                    "unit": "s"
+                }
+            },
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "sum(rate(reactive_tasks_duration_seconds_bucket[$__rate_interval])) by (le)",
+                    "format": "heatmap",
+                    "legendFormat": "{{le}}",
+                    "range": true,
+                    "refId": "A"
+                }
+            ],
+            "title": "Task Duration Heatmap",
+            "type": "heatmap"
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "gridPos": {
+                "h": 8,
+                "w": 12,
+                "x": 12,
+                "y": 8
+            },
+            "id": 4,
+            "options": {
+                "legend": {
+                    "calcs": [
+                        "mean",
+                        "max"
+                    ],
+                    "displayMode": "table",
+                    "placement": "right",
+                    "showLegend": true
+                },
+                "tooltip": {
+                    "mode": "multi",
+                    "sort": "desc"
+                }
+            },
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "sum(rate(reactive_tasks_duration_seconds_count[1m])) by (task_name)",
+                    "legendFormat": "{{task_name}}",
+                    "range": true,
+                    "refId": "A"
+                }
+            ],
+            "title": "Throughput by Task",
+            "type": "timeseries",
+            "fieldConfig": {
+                "defaults": {
+                    "unit": "ops"
+                }
+            }
+        },
+        {
+            "collapsed": false,
+            "gridPos": {
+                "h": 1,
+                "w": 24,
+                "x": 0,
+                "y": 16
+            },
+            "id": 14,
+            "panels": [],
+            "title": "Reliability & Errors",
+            "type": "row"
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "gridPos": {
+                "h": 8,
+                "w": 12,
+                "x": 0,
+                "y": 17
+            },
+            "id": 5,
+            "options": {
+                "legend": {
+                    "calcs": [],
+                    "displayMode": "list",
+                    "placement": "bottom",
+                    "showLegend": true
+                },
+                "tooltip": {
+                    "mode": "single",
+                    "sort": "none"
+                }
+            },
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "sum(rate(reactive_tasks_duration_seconds_count{status=\"failed\"}[1m])) by (task_name)",
+                    "legendFormat": "{{task_name}} Failures",
+                    "range": true,
+                    "refId": "A"
+                }
+            ],
+            "title": "Error Rate (Failures/sec)",
+            "type": "timeseries",
+            "fieldConfig": {
+                "defaults": {
+                    "unit": "ops",
+                    "color": {
+                        "mode": "fixed",
+                        "fixedColor": "red"
+                    }
+                }
+            }
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "gridPos": {
+                "h": 8,
+                "w": 12,
+                "x": 12,
+                "y": 17
+            },
+            "id": 6,
+            "options": {
+                "legend": {
+                    "calcs": [
+                        "lastNotNull"
+                    ],
+                    "displayMode": "list",
+                    "placement": "right",
+                    "showLegend": true
+                },
+                "tooltip": {
+                    "mode": "single",
+                    "sort": "none"
+                }
+            },
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "increase(reactive_tasks_retries_total[5m])",
+                    "legendFormat": "{{task_name}} Retries",
+                    "range": true,
+                    "refId": "A"
+                }
+            ],
+            "title": "Recent Retries (Last 5m)",
+            "type": "barchart"
+        },
+        {
+            "collapsed": false,
+            "gridPos": {
+                "h": 1,
+                "w": 24,
+                "x": 0,
+                "y": 25
+            },
+            "id": 15,
+            "panels": [],
+            "title": "Queues & Lag Deep Dive",
+            "type": "row"
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "type": "barchart",
+            "title": "Queue Composition by Task",
+            "gridPos": {
+                "x": 0,
+                "y": 26,
+                "w": 12,
+                "h": 8
+            },
+            "options": {
+                "stacking": "normal",
+                "legend": {
+                    "displayMode": "table",
+                    "placement": "right",
+                    "calcs": [
+                        "lastNotNull",
+                        "max"
+                    ]
+                }
+            },
+            "fieldConfig": {
+                "defaults": {
+                    "decimals": 0
+                },
+                "overrides": [
+                    {
+                        "matcher": {
+                            "id": "byRegexp",
+                            "options": ".* - failed"
+                        },
+                        "properties": [
+                            {
+                                "id": "color",
+                                "value": {
+                                    "fixedColor": "red",
+                                    "mode": "fixed"
+                                }
+                            }
+                        ]
+                    },
+                    {
+                        "matcher": {
+                            "id": "byRegexp",
+                            "options": ".* - completed"
+                        },
+                        "properties": [
+                            {
+                                "id": "color",
+                                "value": {
+                                    "fixedColor": "green",
+                                    "mode": "fixed"
+                                }
+                            }
+                        ]
+                    },
+                    {
+                        "matcher": {
+                            "id": "byRegexp",
+                            "options": ".* - processing"
+                        },
+                        "properties": [
+                            {
+                                "id": "color",
+                                "value": {
+                                    "fixedColor": "blue",
+                                    "mode": "fixed"
+                                }
+                            }
+                        ]
+                    },
+                    {
+                        "matcher": {
+                            "id": "byRegexp",
+                            "options": ".* - pending"
+                        },
+                        "properties": [
+                            {
+                                "id": "color",
+                                "value": {
+                                    "fixedColor": "orange",
+                                    "mode": "fixed"
+                                }
+                            }
+                        ]
+                    }
+                ]
+            },
+            "targets": [
+                {
+                    "refId": "A",
+                    "expr": "sum(reactive_tasks_queue_depth) by (task_name, status)",
+                    "legendFormat": "{{task_name}} - {{status}}"
+                }
+            ]
+        },
+        {
+            "datasource": {
+                "type": "prometheus",
+                "uid": "prometheus"
+            },
+            "gridPos": {
+                "h": 8,
+                "w": 12,
+                "x": 12,
+                "y": 26
+            },
+            "id": 8,
+            "type": "table",
+            "title": "Top Lagging Tasks",
+            "targets": [
+                {
+                    "datasource": {
+                        "type": "prometheus",
+                        "uid": "prometheus"
+                    },
+                    "editorMode": "code",
+                    "expr": "topk(10, reactive_tasks_global_lag_seconds)",
+                    "legendFormat": "{{task_name}}",
+                    "range": true,
+                    "format": "table",
+                    "instant": true,
+                    "refId": "A"
+                }
+            ],
+            "transformations": [
+                {
+                    "id": "organize",
+                    "options": {
+                        "excludeByName": {
+                            "Time": true
+                        },
+                        "renameByName": {
+                            "Value": "Lag (seconds)"
+                        }
+                    }
+                }
+            ],
+            "fieldConfig": {
+                "defaults": {
+                    "unit": "s",
+                    "custom": {
+                        "align": "auto",
+                        "displayMode": "color-background-solid"
+                    },
+                    "thresholds": {
+                        "mode": "absolute",
+                        "steps": [
+                            {
+                                "color": "green",
+                                "value": null
+                            },
+                            {
+                                "color": "yellow",
+                                "value": 10
+                            },
+                            {
+                                "color": "red",
+                                "value": 60
+                            }
+                        ]
+                    }
+                }
+            }
+        }
+    ],
+    "refresh": "5s",
+    "schemaVersion": 38,
+    "style": "dark",
+    "tags": [
+        "mongodash",
+        "reactive-tasks"
+    ],
+    "templating": {
+        "list": []
+    },
+    "time": {
+        "from": "now-15m",
+        "to": "now"
+    },
+    "timepicker": {},
+    "timezone": "",
+    "title": "Reactive Tasks - System Overview",
+    "uid": "reactive-tasks-overview",
+    "version": 2,
+    "weekStart": ""
+}