Observerbarhet: strukturert logging, metrikker, /metrics-endepunkt (oppgave 12.1)

Legger til observerbarhetslaget i maskinrommet:

- Strukturert JSON-logging via LOG_FORMAT=json (maskinlesbart for
  log-aggregering). Default er human-readable for utvikling.
- Ny metrics-modul med in-memory request latency tracking per rute
  (count, avg, min, max, p50/p95/p99 fra siste 1000 forespørsler).
- Custom axum-middleware erstatter tower_http::TraceLayer — logger
  method, path, status og duration_ms per request, og mater
  metrikk-samleren.
- GET /metrics-endepunkt som returnerer:
  - request_latency: per-rute statistikk
  - queue_depth: pending/running/error/retry fra job_queue
  - ai_cost: aggregert fra ai_usage_log (siste time/24h/30d)
- Default loggnivå endret fra debug til info for mindre støy.

CLAUDE.md

@@ -63,6 +63,7 @@ CLAUDE.md er eneste startdokument. Alt annet ligger under `docs/`:
   - `jobbkø.md` — PostgreSQL-basert køsystem for bakgrunnsjobber
   - `synkronisering.md` — PostgreSQL ↔ SpacetimeDB dataflyt og eierskapsmodell
   - `claude_agent.md` — Claude som chat-deltaker: arkitektur, triggere, sikkerhet
+  - `observerbarhet.md` — Strukturert logging, metrikk-endepunkt (/metrics), AI-kostnad
 - `docs/erfaringer/` — Lærdommer fra v1 (adapter-mønster, Svelte 5, SpacetimeDB, Authentik)
 - `reference/` — Kode fra v1 med gjenbruksverdi (Editor.svelte)
 - `ops/` — Repeterbare vedlikeholdsjobber (ryddejobb, doc-audit, drift-sjekk)

docs/infra/observerbarhet.md

@@ -0,0 +1,104 @@
+# Observerbarhet
+
+Maskinrommets observerbarhetslag: strukturert logging, request-metrikker,
+jobbkø-dybde og AI-kostnadsovervåking.
+
+## Strukturert logging
+
+Maskinrommet bruker `tracing` med `tracing-subscriber`. Format velges via
+miljøvariabel:
+
+| `LOG_FORMAT` | Resultat |
+|---|---|
+| `json` | JSON-linjer (maskinlesbart, egnet for log-aggregering) |
+| *(annet/utelatt)* | Human-readable (standard for utvikling) |
+
+Loggnivå styres via `RUST_LOG` (default: `maskinrommet=info`).
+
+### JSON-format eksempel
+
+```json
+{"timestamp":"2026-03-18T12:00:00Z","level":"INFO","fields":{"method":"GET","path":"/health","status":200,"duration_ms":1.2},"target":"maskinrommet::metrics","message":"request"}
+```
+
+## Metrikk-endepunkt
+
+`GET /metrics` returnerer JSON med tre seksjoner:
+
+### request_latency
+
+In-memory per-rute statistikk (nullstilles ved restart):
+
+```json
+{
+  "uptime_secs": 3600,
+  "total_requests": 1234,
+  "routes": {
+    "/health": {
+      "count": 500,
+      "avg_ms": 1.2,
+      "min_ms": 0.3,
+      "max_ms": 15.0,
+      "p50_ms": 0.8,
+      "p95_ms": 3.5,
+      "p99_ms": 12.0
+    }
+  }
+}
+```
+
+Percentilene beregnes fra de siste 1000 forespørslene per rute.
+
+### queue_depth
+
+Sanntidsstatus fra `job_queue`-tabellen:
+
+```json
+{
+  "pending": 3,
+  "running": 1,
+  "error": 0,
+  "retry": 1,
+  "completed_last_hour": 42
+}
+```
+
+### ai_cost
+
+Aggregerte AI-kostnader fra `ai_usage_log` for tre tidsperioder
+(siste time, 24 timer, 30 dager):
+
+```json
+{
+  "last_hour": {
+    "requests": 5,
+    "input_tokens": 12000,
+    "output_tokens": 3000,
+    "estimated_cost_usd": 0.08
+  }
+}
+```
+
+Kostnadsestimatet er basert på konservative gjennomsnittspriser
+($3/MTok input, $15/MTok output).
+
+## Request-logging middleware
+
+Alle HTTP-forespørsler logges med:
+
+- `method` — HTTP-metode
+- `path` — Forespørsels-sti
+- `status` — HTTP-statuskode
+- `duration_ms` — Total responstid
+
+Dette erstatter `tower_http::TraceLayer` med en mer strukturert
+variant som også mater metrikk-samleren.
+
+## Arkitektur
+
+Metrikker samles i minnet (ingen ekstern avhengighet som Prometheus).
+Designvalg:
+
+- **Enkel:** Ingen nye avhengigheter eller ekstern infrastruktur
+- **Tilstrekkelig:** Queue depth og AI-kostnad fra PG, latency in-memory
+- **Utvidbar:** Kan legge til Prometheus-eksport senere ved behov

maskinrommet/src/main.rs

@@ -12,6 +12,7 @@ mod intentions;
 pub mod jobs;
 pub mod livekit;
 pub mod maintenance;
+pub mod metrics;
 pub mod pruning;
 mod queries;
 pub mod publishing;
@@ -30,11 +31,10 @@ pub mod user_usage;
 mod warmup;
 mod workspace;
 
-use axum::{extract::State, http::StatusCode, routing::{get, post}, Json, Router};
+use axum::{extract::State, http::StatusCode, middleware, routing::{get, post}, Json, Router};
 use serde::Serialize;
 use sqlx::postgres::PgPoolOptions;
 use sqlx::PgPool;
-use tower_http::trace::TraceLayer;
 use tracing_subscriber::{layer::SubscriberExt, util::SubscriberInitExt, EnvFilter};
 
 use auth::{AuthUser, JwksKeys};
@@ -51,6 +51,7 @@ pub struct AppState {
     pub dynamic_page_cache: publishing::DynamicPageCache,
     pub maintenance: maintenance::MaintenanceState,
     pub priority_rules: resources::PriorityRules,
+    pub metrics: metrics::MetricsCollector,
 }
 
 #[derive(Serialize)]
@@ -69,13 +70,21 @@ struct MeResponse {
 
 #[tokio::main]
 async fn main() {
-    tracing_subscriber::registry()
+    // Strukturert logging: LOG_FORMAT=json for maskinlesbart, ellers human-readable.
-        .with(
+    let env_filter = EnvFilter::try_from_default_env()
-            EnvFilter::try_from_default_env()
+        .unwrap_or_else(|_| "maskinrommet=info,tower_http=info".parse().unwrap());
-                .unwrap_or_else(|_| "maskinrommet=debug,tower_http=debug".parse().unwrap()),
+    let log_format = std::env::var("LOG_FORMAT").unwrap_or_default();
-        )
+    if log_format == "json" {
-        .with(tracing_subscriber::fmt::layer())
+        tracing_subscriber::registry()
-        .init();
+            .with(env_filter)
+            .with(tracing_subscriber::fmt::layer().json())
+            .init();
+    } else {
+        tracing_subscriber::registry()
+            .with(env_filter)
+            .with(tracing_subscriber::fmt::layer())
+            .init();
+    }
 
     // Database
     let database_url = std::env::var("DATABASE_URL")
@@ -179,7 +188,8 @@ async fn main() {
 
     let index_cache = publishing::new_index_cache();
     let dynamic_page_cache = publishing::new_dynamic_page_cache();
-    let state = AppState { db, jwks, stdb, cas, index_cache, dynamic_page_cache, maintenance, priority_rules };
+    let metrics = metrics::MetricsCollector::new();
+    let state = AppState { db, jwks, stdb, cas, index_cache, dynamic_page_cache, maintenance, priority_rules, metrics };
 
     // Ruter: /health er offentlig, /me krever gyldig JWT
     let app = Router::new()
@@ -278,7 +288,9 @@ async fn main() {
         .route("/custom-domain/sok", get(custom_domain::serve_custom_domain_search))
         .route("/custom-domain/om", get(custom_domain::serve_custom_domain_about))
         .route("/custom-domain/{article_id}", get(custom_domain::serve_custom_domain_article))
-        .layer(TraceLayer::new_for_http())
+        // Observerbarhet (oppgave 12.1)
+        .route("/metrics", get(metrics::metrics_endpoint))
+        .layer(middleware::from_fn_with_state(state.clone(), metrics::latency_middleware))
         .with_state(state);
 
     let bind = std::env::var("BIND_ADDR").unwrap_or_else(|_| "0.0.0.0:3100".to_string());

maskinrommet/src/metrics.rs

@@ -0,0 +1,326 @@
+// Observerbarhet — in-memory metrikker for maskinrommet.
+//
+// Samler request-latency per rute, eksponerer via /metrics-endepunkt.
+// Queue depth og AI-kostnad hentes direkte fra PG ved forespørsel.
+//
+// Ref: oppgave 12.1
+
+use axum::{
+    body::Body,
+    extract::State,
+    http::{Request, StatusCode},
+    middleware::Next,
+    response::{IntoResponse, Response},
+    Json,
+};
+use serde::Serialize;
+use sqlx::PgPool;
+use std::collections::HashMap;
+use std::sync::{Arc, Mutex};
+use std::time::Instant;
+
+/// In-memory metrikk-samler. Deles via Arc i AppState.
+#[derive(Clone)]
+pub struct MetricsCollector {
+    inner: Arc<Mutex<MetricsInner>>,
+}
+
+struct MetricsInner {
+    /// Latency-data per rute: path → liste av varigheter i mikrosekunder
+    routes: HashMap<String, RouteStats>,
+    /// Totale forespørsler siden oppstart
+    total_requests: u64,
+    /// Oppstartstidspunkt
+    started_at: Instant,
+}
+
+struct RouteStats {
+    count: u64,
+    total_us: u64,
+    min_us: u64,
+    max_us: u64,
+    /// Siste N latencies for p50/p95/p99-beregning (ringbuffer)
+    recent: Vec<u64>,
+}
+
+const RECENT_BUFFER_SIZE: usize = 1000;
+
+impl RouteStats {
+    fn new() -> Self {
+        Self {
+            count: 0,
+            total_us: 0,
+            min_us: u64::MAX,
+            max_us: 0,
+            recent: Vec::with_capacity(RECENT_BUFFER_SIZE),
+        }
+    }
+
+    fn record(&mut self, duration_us: u64) {
+        self.count += 1;
+        self.total_us += duration_us;
+        if duration_us < self.min_us {
+            self.min_us = duration_us;
+        }
+        if duration_us > self.max_us {
+            self.max_us = duration_us;
+        }
+        if self.recent.len() >= RECENT_BUFFER_SIZE {
+            self.recent.remove(0);
+        }
+        self.recent.push(duration_us);
+    }
+
+    fn percentile(&self, p: f64) -> u64 {
+        if self.recent.is_empty() {
+            return 0;
+        }
+        let mut sorted = self.recent.clone();
+        sorted.sort_unstable();
+        let idx = ((p / 100.0) * (sorted.len() as f64 - 1.0)).round() as usize;
+        sorted[idx.min(sorted.len() - 1)]
+    }
+}
+
+impl MetricsCollector {
+    pub fn new() -> Self {
+        Self {
+            inner: Arc::new(Mutex::new(MetricsInner {
+                routes: HashMap::new(),
+                total_requests: 0,
+                started_at: Instant::now(),
+            })),
+        }
+    }
+
+    /// Registrer en forespørsel med varighet.
+    pub fn record(&self, path: &str, duration_us: u64) {
+        let mut inner = self.inner.lock().unwrap();
+        inner.total_requests += 1;
+        inner
+            .routes
+            .entry(path.to_string())
+            .or_insert_with(RouteStats::new)
+            .record(duration_us);
+    }
+
+    /// Hent snapshot av alle rute-metrikker.
+    pub fn snapshot(&self) -> MetricsSnapshot {
+        let inner = self.inner.lock().unwrap();
+        let uptime_secs = inner.started_at.elapsed().as_secs();
+
+        let routes = inner
+            .routes
+            .iter()
+            .map(|(path, stats)| {
+                (
+                    path.clone(),
+                    RouteMetrics {
+                        count: stats.count,
+                        avg_ms: if stats.count > 0 {
+                            (stats.total_us as f64 / stats.count as f64) / 1000.0
+                        } else {
+                            0.0
+                        },
+                        min_ms: if stats.min_us == u64::MAX {
+                            0.0
+                        } else {
+                            stats.min_us as f64 / 1000.0
+                        },
+                        max_ms: stats.max_us as f64 / 1000.0,
+                        p50_ms: stats.percentile(50.0) as f64 / 1000.0,
+                        p95_ms: stats.percentile(95.0) as f64 / 1000.0,
+                        p99_ms: stats.percentile(99.0) as f64 / 1000.0,
+                    },
+                )
+            })
+            .collect();
+
+        MetricsSnapshot {
+            uptime_secs,
+            total_requests: inner.total_requests,
+            routes,
+        }
+    }
+}
+
+#[derive(Serialize)]
+pub struct MetricsSnapshot {
+    pub uptime_secs: u64,
+    pub total_requests: u64,
+    pub routes: HashMap<String, RouteMetrics>,
+}
+
+#[derive(Serialize)]
+pub struct RouteMetrics {
+    pub count: u64,
+    pub avg_ms: f64,
+    pub min_ms: f64,
+    pub max_ms: f64,
+    pub p50_ms: f64,
+    pub p95_ms: f64,
+    pub p99_ms: f64,
+}
+
+// --- Queue depth ---
+
+#[derive(Serialize)]
+pub struct QueueDepth {
+    pub pending: i64,
+    pub running: i64,
+    pub error: i64,
+    pub retry: i64,
+    pub completed_last_hour: i64,
+}
+
+async fn query_queue_depth(db: &PgPool) -> Result<QueueDepth, sqlx::Error> {
+    let row = sqlx::query_as::<_, (i64, i64, i64, i64)>(
+        r#"
+        SELECT
+            COUNT(*) FILTER (WHERE status = 'pending') AS pending,
+            COUNT(*) FILTER (WHERE status = 'running') AS running,
+            COUNT(*) FILTER (WHERE status = 'error') AS error,
+            COUNT(*) FILTER (WHERE status = 'retry') AS retry
+        FROM job_queue
+        "#,
+    )
+    .fetch_one(db)
+    .await?;
+
+    let completed: (i64,) = sqlx::query_as(
+        "SELECT COUNT(*) FROM job_queue WHERE status = 'completed' AND updated_at > now() - interval '1 hour'",
+    )
+    .fetch_one(db)
+    .await?;
+
+    Ok(QueueDepth {
+        pending: row.0,
+        running: row.1,
+        error: row.2,
+        retry: row.3,
+        completed_last_hour: completed.0,
+    })
+}
+
+// --- AI cost ---
+
+#[derive(Serialize)]
+pub struct AiCostSummary {
+    pub last_hour: AiCostPeriod,
+    pub last_24h: AiCostPeriod,
+    pub last_30d: AiCostPeriod,
+}
+
+#[derive(Serialize, Default)]
+pub struct AiCostPeriod {
+    pub requests: i64,
+    pub input_tokens: i64,
+    pub output_tokens: i64,
+    pub estimated_cost_usd: f64,
+}
+
+async fn query_ai_cost(db: &PgPool) -> Result<AiCostSummary, sqlx::Error> {
+    // Hent aggregerte AI-kostnader fra ai_usage_log for tre tidsperioder
+    let row = sqlx::query_as::<_, (i64, i64, i64, i64, i64, i64, i64, i64, i64)>(
+        r#"
+        SELECT
+            COUNT(*) FILTER (WHERE created_at > now() - interval '1 hour'),
+            COALESCE(SUM(input_tokens) FILTER (WHERE created_at > now() - interval '1 hour'), 0),
+            COALESCE(SUM(output_tokens) FILTER (WHERE created_at > now() - interval '1 hour'), 0),
+            COUNT(*) FILTER (WHERE created_at > now() - interval '24 hours'),
+            COALESCE(SUM(input_tokens) FILTER (WHERE created_at > now() - interval '24 hours'), 0),
+            COALESCE(SUM(output_tokens) FILTER (WHERE created_at > now() - interval '24 hours'), 0),
+            COUNT(*) FILTER (WHERE created_at > now() - interval '30 days'),
+            COALESCE(SUM(input_tokens) FILTER (WHERE created_at > now() - interval '30 days'), 0),
+            COALESCE(SUM(output_tokens) FILTER (WHERE created_at > now() - interval '30 days'), 0)
+        FROM ai_usage_log
+        "#,
+    )
+    .fetch_one(db)
+    .await?;
+
+    fn estimate_cost(input_tokens: i64, output_tokens: i64) -> f64 {
+        // Grovt estimat basert på typiske Claude/GPT-priser (USD per 1M tokens)
+        // Input: ~$3/MTok, Output: ~$15/MTok (konservativt gjennomsnitt)
+        let input_cost = input_tokens as f64 * 3.0 / 1_000_000.0;
+        let output_cost = output_tokens as f64 * 15.0 / 1_000_000.0;
+        ((input_cost + output_cost) * 100.0).round() / 100.0
+    }
+
+    Ok(AiCostSummary {
+        last_hour: AiCostPeriod {
+            requests: row.0,
+            input_tokens: row.1,
+            output_tokens: row.2,
+            estimated_cost_usd: estimate_cost(row.1, row.2),
+        },
+        last_24h: AiCostPeriod {
+            requests: row.3,
+            input_tokens: row.4,
+            output_tokens: row.5,
+            estimated_cost_usd: estimate_cost(row.4, row.5),
+        },
+        last_30d: AiCostPeriod {
+            requests: row.6,
+            input_tokens: row.7,
+            output_tokens: row.8,
+            estimated_cost_usd: estimate_cost(row.7, row.8),
+        },
+    })
+}
+
+// --- Middleware ---
+
+/// Axum-middleware som måler request-latency og logger strukturert.
+pub async fn latency_middleware(
+    State(state): State<crate::AppState>,
+    request: Request<Body>,
+    next: Next,
+) -> Response {
+    let path = request.uri().path().to_string();
+    let method = request.method().clone();
+    let start = Instant::now();
+
+    let response = next.run(request).await;
+
+    let duration = start.elapsed();
+    let duration_us = duration.as_micros() as u64;
+    let status = response.status().as_u16();
+
+    state.metrics.record(&path, duration_us);
+
+    tracing::info!(
+        method = %method,
+        path = %path,
+        status = status,
+        duration_ms = duration.as_secs_f64() * 1000.0,
+        "request"
+    );
+
+    response
+}
+
+// --- Endpoint ---
+
+#[derive(Serialize)]
+struct MetricsResponse {
+    request_latency: MetricsSnapshot,
+    queue_depth: Option<QueueDepth>,
+    ai_cost: Option<AiCostSummary>,
+}
+
+/// GET /metrics — samlet observerbarhets-endepunkt.
+pub async fn metrics_endpoint(
+    State(state): State<crate::AppState>,
+) -> Result<impl IntoResponse, StatusCode> {
+    let request_latency = state.metrics.snapshot();
+
+    let queue_depth = query_queue_depth(&state.db).await.ok();
+    let ai_cost = query_ai_cost(&state.db).await.ok();
+
+    Ok(Json(MetricsResponse {
+        request_latency,
+        queue_depth,
+        ai_cost,
+    }))
+}

tasks.md

@@ -266,8 +266,7 @@ kaller dem direkte. Samme verktøy, to brukere.
 
 ## Fase 12: Herding
 
-- [~] 12.1 Observerbarhet: strukturert logging, metrikker (request latency, queue depth, AI cost).
+- [x] 12.1 Observerbarhet: strukturert logging, metrikker (request latency, queue depth, AI cost).
-  > Påbegynt: 2026-03-18T10:55
 - [ ] 12.2 Backup: PG-dump rutine, STDB → PG gjenoppbygging ved krasj.
 - [ ] 12.3 Feilhåndtering: retry med backoff i skrivestien, dead letter queue for feilede PG-skrivinger.
 - [ ] 12.4 Ytelse: profiler STDB-spørringer, optimaliser node_access-oppdatering.