mirror of
https://github.com/Memo-2023/mana-monorepo.git
synced 2026-05-14 16:41:08 +02:00
chore(observability): Grafana dashboard for agent-loop metrics
One focused dashboard covering the M1+M2 instrumentation in a single
view. Sections top-to-bottom:
1. Service Health — mana-mcp + mana-ai up/down, 1h deny rate,
compactions/h. The deny rate is the single most important
number during POLICY_MODE=log-only soak: a non-zero
deny/min in log-only means real traffic that enforce mode
would reject.
2. Policy Gate (mana-mcp)
- Decisions / sec by outcome (allow/deny/flagged)
- Deny reasons breakdown — the soak signal for flipping to
enforce. If one reason dominates, address it before the flip.
- Tool invocations / sec by outcome (success / handler-error /
input-invalid)
- Top 10 invoked tools (24h) — usage heatmap for prioritising
which tools deserve the best policy-hint tuning.
- Handler p50/p95/p99 latency per tool.
3. Reminder Channel (mana-ai)
- Rate by producer (token-budget, retry-loop, compacted)
- Rate by severity. The interesting signal is whether
warn/escalate trend DOWN over time — it means the LLM is
actually reacting to the hints. If warn stays flat, the
producer wording probably isn't landing.
4. Context Compactor (mana-ai)
- Triggers/h cumulative
- Turns folded per compaction (p50/p95). Values < 3 flag
MANA_AI_COMPACT_MAX_CTX misconfig — the threshold is firing
on already-short histories.
5. Mission Runner Baseline — tick duration + planner rounds for
correlation (e.g. "did enabling the compactor change mean
tick duration?").
Dashboard provisioning already auto-loads anything in /var/lib/grafana/
dashboards (docker/grafana/provisioning/dashboards/default.yml), so
this is live after the next grafana restart. UID agent-loop.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Till JS
parent
89388fb369
commit
004b3b7fca
1 changed files with 412 additions and 0 deletions
412
docker/grafana/dashboards/agent-loop.json
Normal file
412
docker/grafana/dashboards/agent-loop.json
Normal file
|
|
@ -0,0 +1,412 @@
|
|||
{
|
||||
"annotations": { "list": [] },
|
||||
"editable": true,
|
||||
"fiscalYearStartMonth": 0,
|
||||
"graphTooltip": 1,
|
||||
"id": null,
|
||||
"links": [
|
||||
{
|
||||
"asDropdown": false,
|
||||
"icon": "external link",
|
||||
"includeVars": true,
|
||||
"keepTime": true,
|
||||
"tags": [],
|
||||
"targetBlank": false,
|
||||
"title": "Master Overview",
|
||||
"type": "link",
|
||||
"url": "/d/master-overview/master-overview"
|
||||
}
|
||||
],
|
||||
"panels": [
|
||||
{
|
||||
"collapsed": false,
|
||||
"gridPos": { "h": 1, "w": 24, "x": 0, "y": 0 },
|
||||
"id": 1,
|
||||
"panels": [],
|
||||
"title": "Service Health",
|
||||
"type": "row"
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": {
|
||||
"defaults": {
|
||||
"color": { "mode": "thresholds" },
|
||||
"mappings": [
|
||||
{
|
||||
"options": {
|
||||
"0": { "color": "red", "index": 1, "text": "DOWN" },
|
||||
"1": { "color": "green", "index": 0, "text": "UP" }
|
||||
},
|
||||
"type": "value"
|
||||
}
|
||||
],
|
||||
"thresholds": {
|
||||
"steps": [
|
||||
{ "color": "red", "value": null },
|
||||
{ "color": "green", "value": 1 }
|
||||
]
|
||||
}
|
||||
}
|
||||
},
|
||||
"gridPos": { "h": 4, "w": 6, "x": 0, "y": 1 },
|
||||
"id": 2,
|
||||
"options": {
|
||||
"colorMode": "background",
|
||||
"graphMode": "none",
|
||||
"reduceOptions": { "calcs": ["lastNotNull"] }
|
||||
},
|
||||
"title": "mana-mcp Status",
|
||||
"type": "stat",
|
||||
"targets": [{ "expr": "up{job=\"mana-mcp\"}", "legendFormat": "Status" }]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": {
|
||||
"defaults": {
|
||||
"color": { "mode": "thresholds" },
|
||||
"mappings": [
|
||||
{
|
||||
"options": {
|
||||
"0": { "color": "red", "index": 1, "text": "DOWN" },
|
||||
"1": { "color": "green", "index": 0, "text": "UP" }
|
||||
},
|
||||
"type": "value"
|
||||
}
|
||||
],
|
||||
"thresholds": {
|
||||
"steps": [
|
||||
{ "color": "red", "value": null },
|
||||
{ "color": "green", "value": 1 }
|
||||
]
|
||||
}
|
||||
}
|
||||
},
|
||||
"gridPos": { "h": 4, "w": 6, "x": 6, "y": 1 },
|
||||
"id": 3,
|
||||
"options": {
|
||||
"colorMode": "background",
|
||||
"graphMode": "none",
|
||||
"reduceOptions": { "calcs": ["lastNotNull"] }
|
||||
},
|
||||
"title": "mana-ai Status",
|
||||
"type": "stat",
|
||||
"targets": [{ "expr": "up{job=\"mana-ai\"}", "legendFormat": "Status" }]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": {
|
||||
"defaults": {
|
||||
"unit": "short",
|
||||
"color": { "mode": "thresholds" },
|
||||
"thresholds": {
|
||||
"steps": [
|
||||
{ "color": "green", "value": null },
|
||||
{ "color": "yellow", "value": 1 },
|
||||
{ "color": "red", "value": 10 }
|
||||
]
|
||||
}
|
||||
}
|
||||
},
|
||||
"gridPos": { "h": 4, "w": 6, "x": 12, "y": 1 },
|
||||
"id": 4,
|
||||
"options": {
|
||||
"colorMode": "value",
|
||||
"graphMode": "area",
|
||||
"reduceOptions": { "calcs": ["lastNotNull"] }
|
||||
},
|
||||
"title": "Policy Denies / min (1h avg)",
|
||||
"type": "stat",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "sum(rate(mana_mcp_policy_decisions_total{decision=\"deny\"}[1h])) * 60",
|
||||
"legendFormat": "deny/min"
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": {
|
||||
"defaults": {
|
||||
"unit": "short",
|
||||
"color": { "mode": "thresholds" },
|
||||
"thresholds": {
|
||||
"steps": [
|
||||
{ "color": "green", "value": null },
|
||||
{ "color": "yellow", "value": 1 }
|
||||
]
|
||||
}
|
||||
}
|
||||
},
|
||||
"gridPos": { "h": 4, "w": 6, "x": 18, "y": 1 },
|
||||
"id": 5,
|
||||
"options": {
|
||||
"colorMode": "value",
|
||||
"graphMode": "area",
|
||||
"reduceOptions": { "calcs": ["lastNotNull"] }
|
||||
},
|
||||
"title": "Compactions / h",
|
||||
"type": "stat",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "sum(rate(mana_ai_compactions_triggered_total[1h])) * 3600",
|
||||
"legendFormat": "per hour"
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
{
|
||||
"collapsed": false,
|
||||
"gridPos": { "h": 1, "w": 24, "x": 0, "y": 5 },
|
||||
"id": 10,
|
||||
"panels": [],
|
||||
"title": "Policy Gate (mana-mcp)",
|
||||
"type": "row"
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": {
|
||||
"defaults": { "unit": "short", "color": { "mode": "palette-classic" } }
|
||||
},
|
||||
"gridPos": { "h": 8, "w": 12, "x": 0, "y": 6 },
|
||||
"id": 11,
|
||||
"options": { "legend": { "displayMode": "table", "placement": "bottom", "calcs": ["sum"] } },
|
||||
"title": "Decisions / sec by outcome",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "sum by (decision) (rate(mana_mcp_policy_decisions_total[5m]))",
|
||||
"legendFormat": "{{decision}}"
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": {
|
||||
"defaults": { "unit": "short", "color": { "mode": "palette-classic" } }
|
||||
},
|
||||
"gridPos": { "h": 8, "w": 12, "x": 12, "y": 6 },
|
||||
"id": 12,
|
||||
"options": { "legend": { "displayMode": "table", "placement": "bottom", "calcs": ["sum"] } },
|
||||
"title": "Deny reasons (soak signal for enforce-flip)",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "sum by (reason) (rate(mana_mcp_policy_decisions_total{decision=\"deny\"}[5m]))",
|
||||
"legendFormat": "{{reason}}"
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": {
|
||||
"defaults": { "unit": "short", "color": { "mode": "palette-classic" } }
|
||||
},
|
||||
"gridPos": { "h": 8, "w": 12, "x": 0, "y": 14 },
|
||||
"id": 13,
|
||||
"options": { "legend": { "displayMode": "table", "placement": "bottom", "calcs": ["sum"] } },
|
||||
"title": "Tool invocations / sec by outcome",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "sum by (outcome) (rate(mana_mcp_tool_invocations_total[5m]))",
|
||||
"legendFormat": "{{outcome}}"
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": { "defaults": { "unit": "short" } },
|
||||
"gridPos": { "h": 8, "w": 12, "x": 12, "y": 14 },
|
||||
"id": 14,
|
||||
"options": { "legend": { "displayMode": "table", "placement": "bottom", "calcs": ["sum"] } },
|
||||
"title": "Top 10 invoked tools (24h)",
|
||||
"type": "bargauge",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "topk(10, sum by (tool) (increase(mana_mcp_tool_invocations_total[24h])))",
|
||||
"legendFormat": "{{tool}}"
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": { "defaults": { "unit": "s" } },
|
||||
"gridPos": { "h": 8, "w": 24, "x": 0, "y": 22 },
|
||||
"id": 15,
|
||||
"options": {
|
||||
"legend": { "displayMode": "table", "placement": "bottom", "calcs": ["mean", "max"] }
|
||||
},
|
||||
"title": "Tool handler latency p50 / p95 / p99",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "histogram_quantile(0.5, sum by (le, tool) (rate(mana_mcp_tool_duration_seconds_bucket[5m])))",
|
||||
"legendFormat": "p50 {{tool}}"
|
||||
},
|
||||
{
|
||||
"expr": "histogram_quantile(0.95, sum by (le, tool) (rate(mana_mcp_tool_duration_seconds_bucket[5m])))",
|
||||
"legendFormat": "p95 {{tool}}"
|
||||
},
|
||||
{
|
||||
"expr": "histogram_quantile(0.99, sum by (le, tool) (rate(mana_mcp_tool_duration_seconds_bucket[5m])))",
|
||||
"legendFormat": "p99 {{tool}}"
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
{
|
||||
"collapsed": false,
|
||||
"gridPos": { "h": 1, "w": 24, "x": 0, "y": 30 },
|
||||
"id": 20,
|
||||
"panels": [],
|
||||
"title": "Reminder Channel (mana-ai)",
|
||||
"type": "row"
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": {
|
||||
"defaults": { "unit": "short", "color": { "mode": "palette-classic" } }
|
||||
},
|
||||
"gridPos": { "h": 8, "w": 12, "x": 0, "y": 31 },
|
||||
"id": 21,
|
||||
"options": { "legend": { "displayMode": "table", "placement": "bottom", "calcs": ["sum"] } },
|
||||
"title": "Reminders emitted / sec by producer",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "sum by (producer) (rate(mana_ai_reminders_emitted_total[5m]))",
|
||||
"legendFormat": "{{producer}}"
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": {
|
||||
"defaults": { "unit": "short", "color": { "mode": "palette-classic" } }
|
||||
},
|
||||
"gridPos": { "h": 8, "w": 12, "x": 12, "y": 31 },
|
||||
"id": 22,
|
||||
"options": { "legend": { "displayMode": "table", "placement": "bottom", "calcs": ["sum"] } },
|
||||
"title": "Reminders by severity (warn/escalate should trend down if the LLM reacts)",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "sum by (severity) (rate(mana_ai_reminders_emitted_total[5m]))",
|
||||
"legendFormat": "{{severity}}"
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
{
|
||||
"collapsed": false,
|
||||
"gridPos": { "h": 1, "w": 24, "x": 0, "y": 39 },
|
||||
"id": 30,
|
||||
"panels": [],
|
||||
"title": "Context Compactor (mana-ai)",
|
||||
"type": "row"
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": { "defaults": { "unit": "short" } },
|
||||
"gridPos": { "h": 8, "w": 12, "x": 0, "y": 40 },
|
||||
"id": 31,
|
||||
"options": { "legend": { "displayMode": "list", "placement": "bottom" } },
|
||||
"title": "Compactions triggered (cumulative)",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "sum(increase(mana_ai_compactions_triggered_total[1h]))",
|
||||
"legendFormat": "per 1h"
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": { "defaults": { "unit": "short" } },
|
||||
"gridPos": { "h": 8, "w": 12, "x": 12, "y": 40 },
|
||||
"id": 32,
|
||||
"options": { "legend": { "displayMode": "list", "placement": "bottom" } },
|
||||
"title": "Turns folded per compaction (histogram)",
|
||||
"description": "p50/p95 of messages folded per compaction event. Values < 3 suggest MANA_AI_COMPACT_MAX_CTX is misconfigured (threshold fires on already-short histories).",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "histogram_quantile(0.5, sum by (le) (rate(mana_ai_compacted_turns_bucket[30m])))",
|
||||
"legendFormat": "p50 turns"
|
||||
},
|
||||
{
|
||||
"expr": "histogram_quantile(0.95, sum by (le) (rate(mana_ai_compacted_turns_bucket[30m])))",
|
||||
"legendFormat": "p95 turns"
|
||||
}
|
||||
]
|
||||
},
|
||||
|
||||
{
|
||||
"collapsed": false,
|
||||
"gridPos": { "h": 1, "w": 24, "x": 0, "y": 48 },
|
||||
"id": 40,
|
||||
"panels": [],
|
||||
"title": "Mission Runner Baseline (for correlation)",
|
||||
"type": "row"
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": { "defaults": { "unit": "s" } },
|
||||
"gridPos": { "h": 8, "w": 12, "x": 0, "y": 49 },
|
||||
"id": 41,
|
||||
"options": {
|
||||
"legend": { "displayMode": "table", "placement": "bottom", "calcs": ["mean", "max"] }
|
||||
},
|
||||
"title": "Tick duration p50 / p95",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "histogram_quantile(0.5, sum by (le) (rate(mana_ai_tick_duration_seconds_bucket[5m])))",
|
||||
"legendFormat": "p50"
|
||||
},
|
||||
{
|
||||
"expr": "histogram_quantile(0.95, sum by (le) (rate(mana_ai_tick_duration_seconds_bucket[5m])))",
|
||||
"legendFormat": "p95"
|
||||
}
|
||||
]
|
||||
},
|
||||
{
|
||||
"datasource": { "type": "prometheus", "uid": "${datasource}" },
|
||||
"fieldConfig": { "defaults": { "unit": "short" } },
|
||||
"gridPos": { "h": 8, "w": 12, "x": 12, "y": 49 },
|
||||
"id": 42,
|
||||
"options": { "legend": { "displayMode": "list", "placement": "bottom" } },
|
||||
"title": "Planner rounds per mission (p50/p95)",
|
||||
"type": "timeseries",
|
||||
"targets": [
|
||||
{
|
||||
"expr": "histogram_quantile(0.5, sum by (le) (rate(mana_ai_planner_rounds_bucket[5m])))",
|
||||
"legendFormat": "p50 rounds"
|
||||
},
|
||||
{
|
||||
"expr": "histogram_quantile(0.95, sum by (le) (rate(mana_ai_planner_rounds_bucket[5m])))",
|
||||
"legendFormat": "p95 rounds"
|
||||
}
|
||||
]
|
||||
}
|
||||
],
|
||||
"schemaVersion": 39,
|
||||
"tags": ["mana", "ai", "agent-loop", "policy", "compactor"],
|
||||
"templating": {
|
||||
"list": [
|
||||
{
|
||||
"current": { "selected": false, "text": "VictoriaMetrics", "value": "victoria-metrics" },
|
||||
"hide": 0,
|
||||
"includeAll": false,
|
||||
"name": "datasource",
|
||||
"options": [],
|
||||
"query": "prometheus",
|
||||
"type": "datasource"
|
||||
}
|
||||
]
|
||||
},
|
||||
"time": { "from": "now-6h", "to": "now" },
|
||||
"title": "Agent Loop — Policy, Reminders, Compactor",
|
||||
"uid": "agent-loop",
|
||||
"version": 1
|
||||
}
|
||||
Loading…
Add table
Add a link
Reference in a new issue