WP-40/41/42: dashboard, security hardening, deployment and operations docs

Dashboard endpoint and page aggregating inventory, vulnerabilities, cluster
health, backups and recent jobs. Security headers (CSP, nosniff, DENY,
referrer policy), 1 MB body limit, configurable login rate limit, audit
steps in CI. Installer script, systemd unit, install/architecture docs.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

.env.example

@@ -9,6 +9,8 @@ BIND=127.0.0.1:8080
 # Created on first start if no user exists
 BOOTSTRAP_ADMIN_EMAIL=admin@example.com
 BOOTSTRAP_ADMIN_PASSWORD=change-me-min-12-chars
+# Login attempts per IP and minute (default 10)
+#LOGIN_RATE_LIMIT=10
 # Set to true behind HTTPS so the refresh cookie is marked Secure
 COOKIE_SECURE=false
 # Directory with the built frontend (served as SPA fallback)

.gitea/workflows/ci.yml

@@ -15,6 +15,9 @@ jobs:
         with: { workspaces: backend }
       - run: cd backend && cargo fmt --check && cargo clippy --workspace --all-targets -- -D warnings
       - run: cd backend && cargo test --workspace
+      - uses: rustsec/audit-check@v2
+        with: { token: ${{ secrets.GITHUB_TOKEN }} }
+        continue-on-error: true
 
   frontend:
     runs-on: ubuntu-latest
@@ -24,6 +27,8 @@ jobs:
         with: { node-version: 22, cache: npm, cache-dependency-path: frontend/package-lock.json }
       - run: cd frontend && npm ci
       - run: cd frontend && npm run lint && npm run typecheck && npm test
+      - run: cd frontend && npm audit --audit-level=high
+        continue-on-error: true
 
   ui:
     runs-on: ubuntu-latest

README.md

@@ -39,6 +39,11 @@ Every feature starts with its tests:
 3. Implement the smallest code that makes them pass, then refactor.
 4. Commit and push.
 
+## Deployment
+
+See [docs/install.md](docs/install.md) (release build, `deploy/install.sh`, systemd unit) and
+[docs/restore.md](docs/restore.md) for restoring backups. `docs/architecture.md` describes the layers.
+
 ## First admin
 
 On start the backend creates an admin user from `BOOTSTRAP_ADMIN_EMAIL` /

ROADMAP.md

@@ -1,6 +1,6 @@
 # SoftVisor Infrastructure Monitoring System – Roadmap
 
-Status: v1.2 (2026-09-02) – Milestones 1 and 2 delivered, test instance on port 3333
+Status: v1.5 (2026-09-02) – all five milestones delivered, test instance on port 3333
 
 This document is derived from `CLAUDE.md`. It breaks the project into work packages (WPs),
 fixes the technical decisions that are not dictated by `CLAUDE.md`, and lists the open
@@ -301,8 +301,8 @@ WP-20, WP-21. Trivy is installed by the deploy script.
 ### Milestone 4 – Backup management (delivered 2026-09-02)
 WP-30, WP-31, WP-32. Restore procedure in `docs/restore.md`.
 
-### Milestone 5 – Hardening and release (planned)
-WP-40, WP-41, WP-42.
+### Milestone 5 – Hardening and release (delivered 2026-09-02)
+WP-40, WP-41, WP-42. Open items: TOTP 2FA, dark mode, running as a non-root service user with a scoped sudoers file, upstream image tag check, backup failure mails.
 
 ---
 
@@ -363,6 +363,6 @@ The server is reachable via `ssh softvisor` (as root). Findings from the inspect
 | WP-30 | M4 | done | 2026-09-02; SMB via smbclient, FTP/FTPS via curl, credentials encrypted |
 | WP-31 | M4 | done | 2026-09-02; sources: PVC hostpath, pg_dumpall, manifests, host dir; openssl encryption |
 | WP-32 | M4 | done | 2026-09-02; upload verify, retention, records; docs/restore.md; failure mail not yet |
-| WP-40 | M5 | todo | |
-| WP-41 | M5 | todo | |
-| WP-42 | M5 | todo | |
+| WP-40 | M5 | done | 2026-09-02; dashboard with tiles for all areas; dark mode not done |
+| WP-41 | M5 | done | 2026-09-02; security headers, CSP, body limit, audits in CI; TOTP 2FA and sudoers scoping not done |
+| WP-42 | M5 | done | 2026-09-02; install.sh, systemd unit, docs/install.md, docs/architecture.md |

backend/crates/api/src/config.rs

@@ -20,6 +20,8 @@ pub struct Config {
     pub bind: SocketAddr,
     pub bootstrap_admin: Option<(String, String)>,
     pub cookie_secure: bool,
+    /// Login attempts per IP and minute.
+    pub login_rate_limit: u32,
     pub frontend_dir: String,
 }
 
@@ -69,6 +71,9 @@ impl Config {
                 .parse()?,
             bootstrap_admin: env("BOOTSTRAP_ADMIN_EMAIL").zip(env("BOOTSTRAP_ADMIN_PASSWORD")),
             cookie_secure: env("COOKIE_SECURE").is_some_and(|v| v == "true" || v == "1"),
+            login_rate_limit: env("LOGIN_RATE_LIMIT")
+                .and_then(|v| v.parse().ok())
+                .unwrap_or(10),
             frontend_dir: env("FRONTEND_DIR").unwrap_or_else(|| "../frontend/dist".into()),
         })
     }

backend/crates/api/src/dashboard.rs

@@ -0,0 +1,172 @@
+//! /api/dashboard: one call with the headline numbers of every area.
+use application::vuln_service::Summary;
+use axum::extract::State;
+use axum::routing::get;
+use axum::{Json, Router};
+use chrono::{DateTime, Duration, Utc};
+use domain::backup::BackupRecord;
+use domain::host::OsInfo;
+use domain::jobs::{JobKind, JobStatus};
+use serde::Serialize;
+use utoipa::ToSchema;
+use uuid::Uuid;
+
+use crate::error::ApiError;
+use crate::extract::AuthUser;
+use crate::AppState;
+
+pub fn router() -> Router<AppState> {
+    Router::new().route("/", get(dashboard))
+}
+
+#[derive(Serialize, ToSchema)]
+pub struct InventoryTile {
+    pub refreshed_at: Option<DateTime<Utc>>,
+    #[schema(value_type = Option<Object>)]
+    pub os: Option<OsInfo>,
+    pub total: usize,
+    pub upgradable: usize,
+    pub security: usize,
+    pub reboot_required: bool,
+}
+
+#[derive(Serialize, ToSchema)]
+pub struct ClusterTile {
+    pub reachable: bool,
+    pub error: Option<String>,
+    pub nodes_ready: usize,
+    pub nodes: usize,
+    pub workloads: usize,
+    pub unhealthy: usize,
+}
+
+#[derive(Serialize, ToSchema)]
+pub struct BackupTile {
+    pub id: Uuid,
+    pub name: String,
+    pub enabled: bool,
+    pub schedule: String,
+    #[schema(value_type = Option<Object>)]
+    pub last_backup: Option<BackupRecord>,
+}
+
+#[derive(Serialize, ToSchema)]
+pub struct JobTile {
+    pub id: Uuid,
+    #[schema(value_type = String)]
+    pub kind: JobKind,
+    #[schema(value_type = String)]
+    pub status: JobStatus,
+    pub started_at: DateTime<Utc>,
+    pub finished_at: Option<DateTime<Utc>>,
+    pub triggered_by: String,
+}
+
+#[derive(Serialize, ToSchema)]
+pub struct DashboardResponse {
+    pub inventory: InventoryTile,
+    #[schema(value_type = Object)]
+    pub vulnerabilities: VulnTile,
+    pub cluster: ClusterTile,
+    pub backups: Vec<BackupTile>,
+    pub recent_jobs: Vec<JobTile>,
+    pub failed_jobs_24h: usize,
+}
+
+#[derive(Serialize)]
+pub struct VulnTile {
+    #[serde(flatten)]
+    pub summary: Summary,
+    pub scanner: String,
+}
+
+#[utoipa::path(get, path = "/api/dashboard", tag = "dashboard", security(("bearer" = [])), responses((status = 200, body = DashboardResponse)))]
+async fn dashboard(
+    State(s): State<AppState>,
+    _: AuthUser,
+) -> Result<Json<DashboardResponse>, ApiError> {
+    let inventory = match s.inventory.current().await? {
+        Some(inv) => InventoryTile {
+            refreshed_at: Some(inv.refreshed_at),
+            total: inv.packages.len(),
+            upgradable: inv.upgradable(),
+            security: inv.security_upgrades(),
+            reboot_required: inv.os.reboot_required,
+            os: Some(inv.os),
+        },
+        None => InventoryTile {
+            refreshed_at: None,
+            os: None,
+            total: 0,
+            upgradable: 0,
+            security: 0,
+            reboot_required: false,
+        },
+    };
+    let scanner = s
+        .vulns
+        .scanner_version()
+        .await
+        .unwrap_or_else(|e| format!("unavailable: {e}"));
+    let vulnerabilities = VulnTile {
+        summary: s.vulns.summary().await?,
+        scanner,
+    };
+    let cluster = match s.cluster.overview().await {
+        Ok(o) => ClusterTile {
+            reachable: true,
+            error: None,
+            nodes_ready: o.nodes.iter().filter(|n| n.ready).count(),
+            nodes: o.nodes.len(),
+            workloads: o.workloads.len(),
+            unhealthy: o.workloads.iter().filter(|w| w.ready < w.desired).count(),
+        },
+        Err(e) => ClusterTile {
+            reachable: false,
+            error: Some(e.to_string()),
+            nodes_ready: 0,
+            nodes: 0,
+            workloads: 0,
+            unhealthy: 0,
+        },
+    };
+    let backups = s
+        .backups
+        .list_strategies()
+        .await?
+        .into_iter()
+        .map(|st| BackupTile {
+            id: st.strategy.id,
+            name: st.strategy.name,
+            enabled: st.strategy.enabled,
+            schedule: st.strategy.schedule,
+            last_backup: st.last_backup,
+        })
+        .collect();
+    let runs = s.jobs.list(50).await?;
+    let since = Utc::now() - Duration::hours(24);
+    let failed_jobs_24h = runs
+        .iter()
+        .filter(|r| r.status == JobStatus::Failed && r.started_at >= since)
+        .count();
+    let recent_jobs = runs
+        .into_iter()
+        .take(8)
+        .map(|r| JobTile {
+            id: r.id,
+            kind: r.kind,
+            status: r.status,
+            started_at: r.started_at,
+            finished_at: r.finished_at,
+            triggered_by: r.triggered_by,
+        })
+        .collect();
+    Ok(Json(DashboardResponse {
+        inventory,
+        vulnerabilities,
+        cluster,
+        backups,
+        recent_jobs,
+        failed_jobs_24h,
+    }))
+}

backend/crates/api/src/lib.rs

@@ -3,11 +3,13 @@ pub mod auth;
 pub mod backups;
 pub mod cluster;
 pub mod config;
+pub mod dashboard;
 pub mod error;
 pub mod extract;
 pub mod jobs;
 pub mod openapi;
 pub mod rate_limit;
+pub mod security;
 pub mod settings;
 pub mod system;
 pub mod test_support;
@@ -164,6 +166,7 @@ impl AppState {
                     inventory: inventory.clone(),
                 }),
             );
+        let login_rate_limit = cfg.login_rate_limit;
         let cluster = Arc::new(ClusterService::new(cluster.clone()));
         let vulns = Arc::new(VulnerabilityService::new(
             scanner,
@@ -190,7 +193,7 @@ impl AppState {
             vulns,
             backups,
             login_limiter: Arc::new(rate_limit::RateLimiter::new(
-                10,
+                login_rate_limit,
                 std::time::Duration::from_secs(60),
             )),
         })
@@ -233,7 +236,10 @@ pub fn build_app(state: AppState) -> Router {
         .nest("/api/cluster", cluster::router())
         .nest("/api/vulnerabilities", vulnerabilities::router())
         .nest("/api/backups", backups::router())
+        .nest("/api/dashboard", dashboard::router())
         .fallback_service(spa)
+        .layer(axum::extract::DefaultBodyLimit::max(1024 * 1024))
+        .layer(axum::middleware::from_fn(security::headers))
         .layer(TraceLayer::new_for_http())
         .with_state(state)
 }

backend/crates/api/src/openapi.rs

@@ -34,6 +34,7 @@ impl Modify for BearerAuth {
         crate::backups::delete_target, crate::backups::test_target, crate::backups::list_strategies, crate::backups::get_strategy,
         crate::backups::create_strategy, crate::backups::update_strategy, crate::backups::delete_strategy,
         crate::backups::run_strategy, crate::backups::records,
+        crate::dashboard::dashboard,
     ),
     modifiers(&BearerAuth)
 )]

backend/crates/api/src/security.rs

@@ -0,0 +1,34 @@
+//! Security headers applied to every response.
+use axum::http::{header, HeaderValue, Request};
+use axum::middleware::Next;
+use axum::response::Response;
+
+pub const CSP: &str = "default-src 'self'; script-src 'self'; style-src 'self' 'unsafe-inline'; img-src 'self' data:; \
+                       font-src 'self'; connect-src 'self'; frame-ancestors 'none'; base-uri 'self'; form-action 'self'";
+
+pub async fn headers(req: Request<axum::body::Body>, next: Next) -> Response {
+    let is_asset = req.uri().path().starts_with("/assets/");
+    let mut res = next.run(req).await;
+    let h = res.headers_mut();
+    h.insert(
+        "x-content-type-options",
+        HeaderValue::from_static("nosniff"),
+    );
+    h.insert("x-frame-options", HeaderValue::from_static("DENY"));
+    h.insert("referrer-policy", HeaderValue::from_static("same-origin"));
+    h.insert("content-security-policy", HeaderValue::from_static(CSP));
+    h.insert(
+        "permissions-policy",
+        HeaderValue::from_static("camera=(), microphone=(), geolocation=()"),
+    );
+    if !h.contains_key(header::CACHE_CONTROL) {
+        let value = if is_asset {
+            "public, max-age=31536000, immutable"
+        } else {
+            "no-store"
+        };
+        h.insert(header::CACHE_CONTROL, HeaderValue::from_static(value));
+    }
+    h.remove(header::SERVER);
+    res
+}

backend/crates/api/src/test_support.rs

@@ -23,6 +23,7 @@ pub fn test_config() -> Config {
         bind: "127.0.0.1:0".parse().unwrap(),
         bootstrap_admin: None,
         cookie_secure: false,
+        login_rate_limit: 10,
         frontend_dir: "/nonexistent".into(),
     }
 }

backend/crates/api/tests/dashboard.rs

@@ -0,0 +1,77 @@
+//! WP-40: GET /api/dashboard aggregates the state of all areas.
+mod common;
+
+use axum::http::StatusCode;
+use common::{get, post, test_app_with_admin};
+use serde_json::json;
+
+const ADMIN: &str = "admin@example.com";
+const PW: &str = "admin-password-123";
+
+async fn wait(app: &axum::Router, token: &str, id: &str) {
+    for _ in 0..100 {
+        let r = get(app, &format!("/api/jobs/{id}"), Some(token)).await;
+        if r.json["status"] != "running" {
+            return;
+        }
+        tokio::time::sleep(std::time::Duration::from_millis(30)).await;
+    }
+}
+
+#[tokio::test]
+async fn dashboard_reflects_inventory_vulnerabilities_cluster_backups_and_jobs() {
+    let app = test_app_with_admin().await;
+    let token = common::login(&app, ADMIN, PW).await.access;
+
+    let d = get(&app, "/api/dashboard", Some(&token)).await;
+    assert_eq!(d.status, StatusCode::OK, "{}", d.json);
+    assert!(d.json["inventory"]["refreshed_at"].is_null());
+    assert_eq!(d.json["vulnerabilities"]["total"]["critical"], 0);
+    assert_eq!(d.json["cluster"]["reachable"], true);
+    assert_eq!(d.json["cluster"]["workloads"], 5);
+    assert_eq!(d.json["cluster"]["unhealthy"], 0);
+    assert_eq!(d.json["backups"].as_array().unwrap().len(), 0);
+    assert_eq!(d.json["recent_jobs"].as_array().unwrap().len(), 0);
+
+    for kind in ["package_refresh", "vulnerability_scan"] {
+        let run = post(&app, "/api/jobs/run", json!({"kind": kind}), Some(&token)).await;
+        wait(&app, &token, run.json["id"].as_str().unwrap()).await;
+    }
+    let t = post(&app, "/api/backups/targets", json!({"name": "NAS", "kind": "smb", "host": "nas", "share": "b", "username": "u", "password": "p"}), Some(&token)).await;
+    let s = post(&app, "/api/backups/strategies", json!({"name": "DB", "source": {"type": "host_path", "path": "/tmp"}, "schedule": "0 0 2 * * *", "target_id": t.json["id"], "retention": 2}), Some(&token)).await;
+    assert_eq!(s.status, StatusCode::CREATED, "{}", s.json);
+
+    let d = get(&app, "/api/dashboard", Some(&token)).await;
+    assert!(d.json["inventory"]["refreshed_at"].is_string());
+    assert_eq!(
+        d.json["inventory"]["os"]["name"],
+        "Debian GNU/Linux 12 (bookworm)"
+    );
+    assert_eq!(d.json["inventory"]["upgradable"], 3);
+    assert_eq!(d.json["inventory"]["security"], 2);
+    assert_eq!(d.json["inventory"]["reboot_required"], true);
+    assert!(
+        d.json["vulnerabilities"]["total"]["critical"]
+            .as_u64()
+            .unwrap()
+            >= 2
+    );
+    assert!(d.json["vulnerabilities"]["last_scan"].is_string());
+    let b = &d.json["backups"][0];
+    assert_eq!(b["name"], "DB");
+    assert!(b["last_backup"].is_null());
+    assert_eq!(b["enabled"], true);
+    let jobs = d.json["recent_jobs"].as_array().unwrap();
+    assert_eq!(jobs.len(), 2);
+    assert_eq!(jobs[0]["kind"], "vulnerability_scan", "newest first");
+    assert!(
+        jobs[0].get("log").is_none(),
+        "no logs in the dashboard payload"
+    );
+    assert_eq!(d.json["failed_jobs_24h"], 0);
+
+    assert_eq!(
+        get(&app, "/api/dashboard", None).await.status,
+        StatusCode::UNAUTHORIZED
+    );
+}

backend/crates/api/tests/security.rs

@@ -0,0 +1,72 @@
+//! WP-41: security headers and cookie hardening.
+mod common;
+
+use axum::http::StatusCode;
+use common::{get, post, test_app_with_admin};
+use serde_json::json;
+
+#[tokio::test]
+async fn responses_carry_security_headers() {
+    let app = test_app_with_admin().await;
+    let res = get(&app, "/healthz", None).await;
+    let h = |k: &str| {
+        res.headers
+            .get(k)
+            .and_then(|v| v.to_str().ok())
+            .unwrap_or("")
+            .to_string()
+    };
+    assert_eq!(h("x-content-type-options"), "nosniff");
+    assert_eq!(h("x-frame-options"), "DENY");
+    assert_eq!(h("referrer-policy"), "same-origin");
+    let csp = h("content-security-policy");
+    assert!(csp.contains("default-src 'self'"), "{csp}");
+    assert!(csp.contains("frame-ancestors 'none'"), "{csp}");
+    assert_eq!(h("cache-control"), "no-store");
+    assert!(res.headers.get("server").is_none());
+
+    // API errors are JSON, not HTML, and still carry the headers
+    let res = get(&app, "/api/users", None).await;
+    assert_eq!(res.status, StatusCode::UNAUTHORIZED);
+    assert_eq!(res.json["error"], "unauthorized");
+    assert_eq!(
+        res.headers.get("x-content-type-options").unwrap(),
+        "nosniff"
+    );
+}
+
+#[tokio::test]
+async fn refresh_cookie_is_strict_and_scoped() {
+    let app = test_app_with_admin().await;
+    let res = post(
+        &app,
+        "/api/auth/login",
+        json!({"email": "admin@example.com", "password": "admin-password-123"}),
+        None,
+    )
+    .await;
+    let cookie = res.headers.get("set-cookie").unwrap().to_str().unwrap();
+    assert!(
+        cookie.contains("HttpOnly")
+            && cookie.contains("SameSite=Strict")
+            && cookie.contains("Path=/api/auth")
+    );
+    // cross-site style request without the cookie cannot refresh
+    assert_eq!(
+        post(&app, "/api/auth/refresh", json!({}), None)
+            .await
+            .status,
+        StatusCode::UNAUTHORIZED
+    );
+}
+
+#[tokio::test]
+async fn oversized_json_bodies_are_rejected() {
+    let app = test_app_with_admin().await;
+    let token = common::login(&app, "admin@example.com", "admin-password-123")
+        .await
+        .access;
+    let big = "x".repeat(2 * 1024 * 1024);
+    let res = post(&app, "/api/users", json!({"email": "a@b.de", "display_name": big, "password": "user-password-123", "role": "user"}), Some(&token)).await;
+    assert_eq!(res.status, StatusCode::PAYLOAD_TOO_LARGE);
+}

backend/crates/infrastructure/src/trivy.rs

@@ -372,8 +372,9 @@ mod tests {
     #[tokio::test]
     async fn image_scan_prefers_local_containerd_when_configured() {
         use std::sync::Mutex;
+        type Call = (Vec<String>, Vec<(String, String)>);
         #[derive(Default)]
-        struct Env(Mutex<Vec<(Vec<String>, Vec<(String, String)>)>>);
+        struct Env(Mutex<Vec<Call>>);
         #[async_trait]
         impl CommandRunner for Env {
             async fn run(&self, p: &str, a: &[&str]) -> Result<crate::host::Output, DomainError> {

deploy/install.sh

@@ -0,0 +1,52 @@
+#!/usr/bin/env bash
+# Server-side installer/updater. Run as root on the Debian host with the release
+# archive (monitoring-server binary + dist/) in the current directory:
+#   ./install.sh [--port 8080]
+set -euo pipefail
+DIR=/opt/monitoring
+PORT=${PORT:-8080}
+[[ "${1:-}" == "--port" ]] && PORT=$2
+
+echo "== dependencies"
+apt-get install -y -q smbclient curl gzip tar openssl >/dev/null
+if ! command -v trivy >/dev/null; then
+  apt-get install -y -q wget apt-transport-https gnupg >/dev/null
+  wget -qO- https://aquasecurity.github.io/trivy-repo/deb/public.key | gpg --dearmor -o /usr/share/keyrings/trivy.gpg
+  echo "deb [signed-by=/usr/share/keyrings/trivy.gpg] https://aquasecurity.github.io/trivy-repo/deb generic main" > /etc/apt/sources.list.d/trivy.list
+  apt-get update -q >/dev/null && apt-get install -y -q trivy >/dev/null
+fi
+
+echo "== files"
+mkdir -p "$DIR/data"
+systemctl stop monitoring.service 2>/dev/null || true
+install -m 0755 monitoring-server "$DIR/monitoring-server"
+rm -rf "$DIR/dist" && cp -r dist "$DIR/dist"
+install -m 0644 monitoring.service /etc/systemd/system/monitoring.service
+
+if [[ ! -f "$DIR/.env" ]]; then
+  echo "== first run: creating $DIR/.env"
+  ADMIN_PW=$(openssl rand -base64 18 | tr -d '/+=' | head -c 20)
+  umask 077
+  cat > "$DIR/.env" <<ENV
+DATABASE_URL=sqlite://$DIR/data/monitoring.db?mode=rwc
+JWT_SECRET=$(openssl rand -hex 32)
+MASTER_KEY=$(openssl rand -hex 32)
+BIND=0.0.0.0:$PORT
+BOOTSTRAP_ADMIN_EMAIL=admin@softvisor.de
+BOOTSTRAP_ADMIN_PASSWORD=$ADMIN_PW
+COOKIE_SECURE=false
+FRONTEND_DIR=$DIR/dist
+WORK_DIR=$DIR/data/work
+RUST_LOG=info,sqlx=warn
+ENV
+  echo "Bootstrap admin: admin@softvisor.de / $ADMIN_PW  (change it after the first login)"
+fi
+grep -q '^MASTER_KEY=' "$DIR/.env" || echo "MASTER_KEY=$(openssl rand -hex 32)" >> "$DIR/.env"
+
+echo "== service"
+systemctl daemon-reload
+systemctl enable -q monitoring.service
+systemctl restart monitoring.service
+sleep 2
+systemctl is-active monitoring.service
+curl -s "http://127.0.0.1:$PORT/healthz" && echo

deploy/monitoring.service

@@ -0,0 +1,19 @@
+[Unit]
+Description=SoftVisor Infrastructure Monitoring
+After=network-online.target snap.microk8s.daemon-kubelite.service
+Wants=network-online.target
+
+[Service]
+WorkingDirectory=/opt/monitoring
+EnvironmentFile=/opt/monitoring/.env
+ExecStart=/opt/monitoring/monitoring-server
+Restart=on-failure
+RestartSec=5
+# The service manages apt and the cluster, so it runs as root; keep the rest tight.
+NoNewPrivileges=false
+ProtectSystem=false
+PrivateTmp=true
+UMask=0077
+
+[Install]
+WantedBy=multi-user.target

docs/architecture.md

@@ -0,0 +1,27 @@
+# Architecture
+
+Single Rust binary (axum) serving the API and the built Vue SPA; SQLite for state; host tools
+(`apt-get`, `dpkg-query`, `snap`, `microk8s kubectl`, `trivy`, `smbclient`, `curl`, `tar`,
+`openssl`) are invoked as subprocesses behind ports so every use case is testable with fakes.
+
+```
+backend/crates/
+  domain/          entities, validation, ports (traits)        no I/O
+  application/     use cases: auth, users, settings, jobs,     depends on domain
+                   scheduler, inventory, upgrade, cluster,
+                   vulnerabilities, backups
+  infrastructure/  SQLite repos, Argon2/JWT/AES-GCM, lettre,   implements the ports
+                   Debian inspector/updater, kube-rs gateway,
+                   Trivy, smbclient/curl storage, collectors
+  api/             axum routes, auth extractors, OpenAPI,      wires everything
+                   security headers, config, main
+frontend/          Vue 3 + TypeScript + Tailwind, Pinia stores, Playwright e2e
+```
+
+Cross-cutting: a `JobRunner` executes long-running work (refresh, upgrade, scan, backup) as
+persisted job runs with live logs; a `Scheduler` ticks every 30 s and starts due jobs from cron
+expressions (settings) and backup strategies. Secrets at rest are AES-256-GCM encrypted with
+`MASTER_KEY`. Authentication: Argon2id passwords, 15-minute JWT access tokens, rotating refresh
+tokens in an HttpOnly, SameSite=Strict cookie scoped to `/api/auth`, reuse detection revokes the
+token family. `FAKE_HOST=true` swaps all host/cluster/scanner/storage adapters for fakes so the
+app runs on a developer machine and in the UI tests.

docs/install.md

@@ -0,0 +1,56 @@
+# Installation and operation
+
+## Build a release
+
+On a machine with Docker (cross-compiles a static x86_64 binary) and Node:
+
+```bash
+cd frontend && npm ci && npm run build && cd ..
+docker run --rm -v "$PWD/backend":/home/rust/src -v cargo-registry-musl:/root/.cargo/registry \
+  messense/rust-musl-cross:x86_64-musl cargo build --release -p api
+mkdir -p release && cp backend/target/x86_64-unknown-linux-musl/release/monitoring-server deploy/install.sh deploy/monitoring.service release/
+cp -r frontend/dist release/dist
+tar -C release -czf monitoring-release.tar.gz .
+```
+
+## Install or update on the Debian host
+
+```bash
+scp monitoring-release.tar.gz root@server:/tmp/
+ssh root@server 'mkdir -p /tmp/rel && tar -C /tmp/rel -xzf /tmp/monitoring-release.tar.gz && cd /tmp/rel && ./install.sh --port 8080'
+```
+
+The installer installs `smbclient`, `curl`, `openssl` and Trivy, copies the files to
+`/opt/monitoring`, creates `/opt/monitoring/.env` with random secrets and a bootstrap admin on
+the first run, and (re)starts the `monitoring.service` systemd unit. Re-running it keeps the
+existing `.env` and database.
+
+For the quick test deployment used during development see `deploy/deploy-test.sh`.
+
+## Configuration
+
+All settings live in `/opt/monitoring/.env` (see `.env.example`). Relevant keys:
+
+| Key | Purpose |
+|-----|---------|
+| `JWT_SECRET` | signs access tokens; changing it logs everyone out |
+| `MASTER_KEY` | encrypts stored SMTP/SMB/FTP passwords and backup passphrases. **Back it up**: without it stored secrets cannot be read |
+| `BIND` | listen address, e.g. `0.0.0.0:8080` |
+| `COOKIE_SECURE` | set `true` behind HTTPS |
+| `KUBECONFIG` / `KUBECTL` | defaults to the microk8s client config and `microk8s kubectl` |
+| `CONTAINERD_ADDRESS` | image scans read local images from this socket first |
+| `LOGIN_RATE_LIMIT` | login attempts per IP and minute |
+
+Put the app behind a TLS-terminating reverse proxy (e.g. the cluster's ingress or nginx on the host)
+for production use and set `COOKIE_SECURE=true`.
+
+## Operations
+
+- Logs: `journalctl -u monitoring.service -f`
+- Database: SQLite at `/opt/monitoring/data/monitoring.db`. Back it up together with `.env` by
+  creating a backup strategy of type "Directory on the host" for `/opt/monitoring` (the `.env`
+  contains the `MASTER_KEY`, so encrypt that strategy or store it on a trusted target).
+- Scheduled jobs: package refresh hourly, vulnerability scan daily at 03:00, backups per strategy;
+  all adjustable in Settings. Failed or interrupted runs appear on the Jobs page and on the dashboard.
+- Restore procedures: `docs/restore.md`.
+- The service runs as root because it drives `apt-get`, `microk8s kubectl` and reads volume data.

frontend/e2e/auth.spec.ts

@@ -25,7 +25,7 @@ test('admin logs in, manages users, logs out; user has no admin access', async (
   await expect(page.getByRole('heading', { name: 'Dashboard' })).toBeVisible()
 
   // create a user
-  await page.getByRole('link', { name: 'Users' }).click()
+  await page.getByRole('navigation').getByRole('link', { name: 'Users' }).click()
   await page.getByRole('button', { name: 'New user' }).click()
   const email = `e2e-${Date.now()}@example.com`
   await page.getByLabel('Email').fill(email)
@@ -48,7 +48,7 @@ test('admin logs in, manages users, logs out; user has no admin access', async (
   // the new user can log in but not manage users
   await login(page, email, 'user-password-123')
   await expect(page.getByRole('heading', { name: 'Dashboard' })).toBeVisible()
-  await expect(page.getByRole('link', { name: 'Users' })).toHaveCount(0)
+  await expect(page.getByRole('navigation').getByRole('link', { name: 'Users' })).toHaveCount(0)
   await page.goto('/users')
   await expect(page.getByText('You do not have permission')).toBeVisible()
 })

frontend/e2e/backups.spec.ts

@@ -5,7 +5,7 @@ test('admin creates a target and a strategy, runs it and sees the backup', async
   await page.getByLabel('Email').fill('admin@example.com')
   await page.getByLabel('Password').fill('admin-password-123')
   await page.getByRole('button', { name: 'Sign in' }).click()
-  await page.getByRole('link', { name: 'Backups' }).click()
+  await page.getByRole('navigation').getByRole('link', { name: 'Backups' }).click()
 
   await page.getByRole('button', { name: 'New target' }).click()
   const name = `NAS ${Date.now()}`

frontend/e2e/cluster.spec.ts

@@ -7,7 +7,7 @@ test('cluster page shows node, workloads and lets an admin restart a workload',
   await page.getByLabel('Email').fill('admin@example.com')
   await page.getByLabel('Password').fill('admin-password-123')
   await page.getByRole('button', { name: 'Sign in' }).click()
-  await page.getByRole('link', { name: 'Kubernetes' }).click()
+  await page.getByRole('navigation').getByRole('link', { name: 'Kubernetes' }).click()
 
   await expect(page.getByTestId('node-version')).toContainText('v1.32')
   const row = page.getByRole('row', { name: /^gitea gitea-postgresql statefulset/ })

frontend/e2e/dashboard.spec.ts

@@ -0,0 +1,14 @@
+import { test, expect } from '@playwright/test'
+
+test('dashboard shows the overview tiles after login', async ({ page }) => {
+  await page.goto('/login')
+  await page.getByLabel('Email').fill('admin@example.com')
+  await page.getByLabel('Password').fill('admin-password-123')
+  await page.getByRole('button', { name: 'Sign in' }).click()
+  await expect(page.getByRole('heading', { name: 'Dashboard' })).toBeVisible()
+  await expect(page.getByTestId('tile-workloads')).toContainText('5')
+  await expect(page.getByTestId('tile-critical')).toBeVisible()
+  await expect(page.getByRole('heading', { name: 'Backups' })).toBeVisible()
+  await page.getByTestId('tile-workloads').click()
+  await expect(page).toHaveURL(/\/cluster/)
+})

frontend/e2e/settings-jobs.spec.ts

@@ -10,7 +10,7 @@ async function loginAdmin(page: Page) {
 
 test('admin saves SMTP settings and a schedule', async ({ page }) => {
   await loginAdmin(page)
-  await page.getByRole('link', { name: 'Settings' }).click()
+  await page.getByRole('navigation').getByRole('link', { name: 'Settings' }).click()
   await page.getByLabel('SMTP host').fill('mail.example.com')
   await page.getByLabel('Port').fill('587')
   await page.getByLabel('Username').fill('bot')
@@ -31,7 +31,7 @@ test('admin saves SMTP settings and a schedule', async ({ page }) => {
 
 test('admin runs a job manually and sees the log', async ({ page }) => {
   await loginAdmin(page)
-  await page.getByRole('link', { name: 'Jobs' }).click()
+  await page.getByRole('navigation').getByRole('link', { name: 'Jobs' }).click()
   await page.getByLabel('Job').selectOption('package_refresh')
   await page.getByRole('button', { name: 'Run now' }).click()
   const row = page.getByRole('row', { name: /package_refresh/ }).first()

frontend/e2e/updates.spec.ts

@@ -5,7 +5,7 @@ test('updates page shows OS info and packages after a refresh', async ({ page })
   await page.getByLabel('Email').fill('admin@example.com')
   await page.getByLabel('Password').fill('admin-password-123')
   await page.getByRole('button', { name: 'Sign in' }).click()
-  await page.getByRole('link', { name: 'Updates' }).click()
+  await page.getByRole('navigation').getByRole('link', { name: 'Updates' }).click()
 
   await page.getByRole('button', { name: 'Refresh inventory' }).click()
   await expect(page.getByTestId('os-name')).toContainText('Debian', { timeout: 15_000 })

frontend/e2e/upgrade.spec.ts

@@ -5,7 +5,7 @@ test('admin updates a selected package and sees the live log', async ({ page })
   await page.getByLabel('Email').fill('admin@example.com')
   await page.getByLabel('Password').fill('admin-password-123')
   await page.getByRole('button', { name: 'Sign in' }).click()
-  await page.getByRole('link', { name: 'Updates' }).click()
+  await page.getByRole('navigation').getByRole('link', { name: 'Updates' }).click()
   await page.getByRole('button', { name: 'Refresh inventory' }).click()
   await expect(page.getByTestId('os-name')).toContainText('Debian', { timeout: 15_000 })
 

frontend/e2e/vulnerabilities.spec.ts

@@ -5,7 +5,7 @@ test('admin runs a scan, filters findings and acknowledges one', async ({ page }
   await page.getByLabel('Email').fill('admin@example.com')
   await page.getByLabel('Password').fill('admin-password-123')
   await page.getByRole('button', { name: 'Sign in' }).click()
-  await page.getByRole('link', { name: 'Vulnerabilities' }).click()
+  await page.getByRole('navigation').getByRole('link', { name: 'Vulnerabilities' }).click()
 
   await page.getByRole('button', { name: 'Scan now' }).click()
   await expect(page.getByTestId('count-critical')).not.toHaveText('0', { timeout: 20_000 })
@@ -29,7 +29,7 @@ test('notification threshold can be changed in settings', async ({ page }) => {
   await page.getByLabel('Email').fill('admin@example.com')
   await page.getByLabel('Password').fill('admin-password-123')
   await page.getByRole('button', { name: 'Sign in' }).click()
-  await page.getByRole('link', { name: 'Settings' }).click()
+  await page.getByRole('navigation').getByRole('link', { name: 'Settings' }).click()
   await page.getByLabel('Notify from severity').selectOption('medium')
   await page.getByRole('button', { name: 'Save notifications' }).click()
   await expect(page.getByRole('status')).toContainText('Notification settings saved')

frontend/playwright.config.ts

@@ -21,6 +21,7 @@ export default defineConfig({
         JWT_SECRET: 'e2e-test-secret-do-not-use-in-production',
         MASTER_KEY: '000102030405060708090a0b0c0d0e0f101112131415161718191a1b1c1d1e1f',
         FAKE_HOST: 'true',
+        LOGIN_RATE_LIMIT: '1000',
         BOOTSTRAP_ADMIN_EMAIL: 'admin@example.com',
         BOOTSTRAP_ADMIN_PASSWORD: 'admin-password-123',
         BIND: '127.0.0.1:8080',

frontend/src/api/types.ts

@@ -231,3 +231,32 @@ export interface StrategyStatus extends BackupStrategyView {
   target_name: string
   last_backup: BackupRecord | null
 }
+
+export interface Dashboard {
+  inventory: {
+    refreshed_at: string | null
+    os: OsInfo | null
+    total: number
+    upgradable: number
+    security: number
+    reboot_required: boolean
+  }
+  vulnerabilities: VulnSummary
+  cluster: {
+    reachable: boolean
+    error: string | null
+    nodes_ready: number
+    nodes: number
+    workloads: number
+    unhealthy: number
+  }
+  backups: {
+    id: string
+    name: string
+    enabled: boolean
+    schedule: string
+    last_backup: BackupRecord | null
+  }[]
+  recent_jobs: Omit<JobRun, 'log'>[]
+  failed_jobs_24h: number
+}

frontend/src/pages/DashboardPage.test.ts

@@ -0,0 +1,101 @@
+import { mount, flushPromises } from '@vue/test-utils'
+import { createPinia, setActivePinia } from 'pinia'
+import DashboardPage from './DashboardPage.vue'
+import { api } from '../api/client'
+import type { Dashboard } from '../api/types'
+
+vi.mock('../api/client', () => ({
+  api: { get: vi.fn(), post: vi.fn(), patch: vi.fn(), put: vi.fn(), delete: vi.fn() },
+  ApiError: class extends Error {},
+}))
+vi.mock('vue-router', () => ({
+  RouterLink: { template: '<a><slot /></a>' },
+  useRouter: () => ({ push: vi.fn() }),
+}))
+
+const data: Dashboard = {
+  inventory: {
+    refreshed_at: '2026-09-02T10:00:00Z',
+    os: {
+      hostname: 'srv',
+      name: 'Debian GNU/Linux 12 (bookworm)',
+      version: '12',
+      kernel: '6.1',
+      uptime_secs: 100,
+      reboot_required: true,
+    },
+    total: 430,
+    upgradable: 3,
+    security: 2,
+    reboot_required: true,
+  },
+  vulnerabilities: {
+    total: { critical: 5, high: 10, medium: 2, low: 1, unknown: 0 },
+    os: { critical: 1, high: 2, medium: 0, low: 0, unknown: 0 },
+    images: { critical: 4, high: 8, medium: 2, low: 1, unknown: 0 },
+    last_scan: '2026-09-02T09:00:00Z',
+    scanner: '0.74.0',
+  },
+  cluster: { reachable: true, error: null, nodes_ready: 1, nodes: 1, workloads: 16, unhealthy: 1 },
+  backups: [
+    { id: 'b1', name: 'Gitea DB', enabled: true, schedule: '0 0 2 * * *', last_backup: null },
+    {
+      id: 'b2',
+      name: 'Repos',
+      enabled: true,
+      schedule: '0 0 3 * * *',
+      last_backup: {
+        id: 'r',
+        strategy_id: 'b2',
+        filename: 'repos_x.tar.gz',
+        size_bytes: 1000,
+        sha256: 'x',
+        created_at: '2026-09-02T03:00:00Z',
+      },
+    },
+  ],
+  recent_jobs: [
+    {
+      id: 'j1',
+      kind: 'vulnerability_scan',
+      status: 'failed',
+      started_at: '2026-09-02T09:00:00Z',
+      finished_at: '2026-09-02T09:05:00Z',
+      triggered_by: 'scheduler',
+    },
+  ],
+  failed_jobs_24h: 1,
+}
+
+beforeEach(() => {
+  setActivePinia(createPinia())
+  vi.mocked(api.get).mockResolvedValue(data)
+})
+
+describe('DashboardPage', () => {
+  it('renders the key numbers of every area', async () => {
+    const w = mount(DashboardPage)
+    await flushPromises()
+    expect(w.text()).toContain('Debian GNU/Linux 12')
+    expect(w.find('[data-testid=tile-upgradable]').text()).toContain('3')
+    expect(w.text()).toContain('Reboot required')
+    expect(w.find('[data-testid=tile-critical]').text()).toContain('5')
+    expect(w.find('[data-testid=tile-workloads]').text()).toContain('16')
+    expect(w.text()).toContain('1 unhealthy')
+    expect(w.text()).toContain('Gitea DB')
+    expect(w.text()).toContain('never')
+    expect(w.text()).toContain('repos_x.tar.gz')
+    expect(w.find('[data-testid=tile-failed-jobs]').text()).toContain('1')
+  })
+
+  it('shows the cluster as unreachable with the error', async () => {
+    vi.mocked(api.get).mockResolvedValue({
+      ...data,
+      cluster: { ...data.cluster, reachable: false, error: 'connection refused' },
+    })
+    const w = mount(DashboardPage)
+    await flushPromises()
+    expect(w.text()).toContain('unreachable')
+    expect(w.text()).toContain('connection refused')
+  })
+})

frontend/src/pages/DashboardPage.vue

@@ -1,6 +1,156 @@
+<script setup lang="ts">
+import { onMounted, ref } from 'vue'
+import { api, ApiError } from '../api/client'
+import type { Dashboard } from '../api/types'
+import { useToastStore } from '../stores/toast'
+
+const toast = useToastStore()
+const d = ref<Dashboard | null>(null)
+
+onMounted(async () => {
+  try {
+    d.value = await api.get<Dashboard>('/api/dashboard')
+  } catch (e) {
+    toast.error(e instanceof ApiError ? e.message : 'Request failed')
+  }
+})
+
+const fmt = (s: string | null | undefined) => (s ? new Date(s).toLocaleString() : 'never')
+const tile = 'block rounded-lg border border-gray-200 bg-white p-4 hover:border-blue-300'
+</script>
+
 <template>
   <h1 class="text-2xl font-semibold">Dashboard</h1>
-  <p class="mt-2 text-gray-600">
-    Server overview, update status, vulnerabilities and backups will appear here.
-  </p>
+  <p v-if="!d" class="mt-2 text-gray-600">Loading overview…</p>
+
+  <template v-if="d">
+    <div class="mt-6 grid gap-4 md:grid-cols-4">
+      <RouterLink to="/updates" :class="tile" class="md:col-span-2">
+        <div class="text-xs uppercase text-gray-500">Operating system</div>
+        <div class="mt-1 text-lg font-medium">
+          {{ d.inventory.os?.name ?? 'not inventoried yet' }}
+        </div>
+        <div class="text-sm text-gray-600">
+          <template v-if="d.inventory.os"
+            >{{ d.inventory.os.hostname }} · kernel {{ d.inventory.os.kernel }} ·
+            {{ d.inventory.total }} packages</template
+          >
+          <template v-else>Run a package refresh on the Updates page.</template>
+        </div>
+        <span
+          v-if="d.inventory.reboot_required"
+          class="mt-2 inline-block rounded-md bg-amber-100 px-2 py-0.5 text-xs font-medium text-amber-800"
+          >Reboot required</span
+        >
+      </RouterLink>
+      <RouterLink to="/updates" :class="tile" data-testid="tile-upgradable">
+        <div class="text-xs uppercase text-gray-500">Upgradable packages</div>
+        <div
+          class="mt-1 text-2xl font-semibold"
+          :class="d.inventory.security ? 'text-red-700' : ''"
+        >
+          {{ d.inventory.upgradable }}
+        </div>
+        <div class="text-sm text-gray-600">
+          {{ d.inventory.security }} security · refreshed {{ fmt(d.inventory.refreshed_at) }}
+        </div>
+      </RouterLink>
+      <RouterLink to="/vulnerabilities" :class="tile" data-testid="tile-critical">
+        <div class="text-xs uppercase text-gray-500">Vulnerabilities</div>
+        <div class="mt-1 text-2xl font-semibold text-red-700">
+          {{ d.vulnerabilities.total.critical }}
+        </div>
+        <div class="text-sm text-gray-600">
+          critical · {{ d.vulnerabilities.total.high }} high · scan
+          {{ fmt(d.vulnerabilities.last_scan) }}
+        </div>
+      </RouterLink>
+      <RouterLink to="/cluster" :class="tile" data-testid="tile-workloads">
+        <div class="text-xs uppercase text-gray-500">Kubernetes</div>
+        <template v-if="d.cluster.reachable">
+          <div class="mt-1 text-2xl font-semibold">{{ d.cluster.workloads }}</div>
+          <div class="text-sm" :class="d.cluster.unhealthy ? 'text-red-700' : 'text-gray-600'">
+            workloads · {{ d.cluster.unhealthy }} unhealthy · {{ d.cluster.nodes_ready }}/{{
+              d.cluster.nodes
+            }}
+            nodes ready
+          </div>
+        </template>
+        <template v-else>
+          <div class="mt-1 text-lg font-medium text-red-700">unreachable</div>
+          <div class="text-sm text-gray-600">{{ d.cluster.error }}</div>
+        </template>
+      </RouterLink>
+      <RouterLink to="/jobs" :class="tile" data-testid="tile-failed-jobs">
+        <div class="text-xs uppercase text-gray-500">Failed jobs (24h)</div>
+        <div class="mt-1 text-2xl font-semibold" :class="d.failed_jobs_24h ? 'text-red-700' : ''">
+          {{ d.failed_jobs_24h }}
+        </div>
+        <div class="text-sm text-gray-600">{{ d.recent_jobs.length }} recent runs</div>
+      </RouterLink>
+    </div>
+
+    <div class="mt-8 grid gap-6 lg:grid-cols-2">
+      <section class="rounded-lg border border-gray-200 bg-white p-4">
+        <div class="flex items-center justify-between">
+          <h2 class="font-medium">Backups</h2>
+          <RouterLink to="/backups" class="text-sm text-blue-600 hover:underline"
+            >Manage</RouterLink
+          >
+        </div>
+        <table class="mt-3 w-full text-left text-sm">
+          <tbody>
+            <tr
+              v-for="b in d.backups"
+              :key="b.id"
+              class="border-t border-gray-100"
+              :class="b.enabled ? '' : 'text-gray-400'"
+            >
+              <td class="py-1.5 font-medium">{{ b.name }}</td>
+              <td class="font-mono text-xs">{{ b.schedule }}</td>
+              <td class="text-xs">
+                <template v-if="b.last_backup"
+                  >{{ b.last_backup.filename }}<br /><span class="text-gray-500">{{
+                    fmt(b.last_backup.created_at)
+                  }}</span></template
+                >
+                <template v-else>never</template>
+              </td>
+            </tr>
+            <tr v-if="d.backups.length === 0">
+              <td class="py-3 text-gray-500">No backup strategies configured.</td>
+            </tr>
+          </tbody>
+        </table>
+      </section>
+      <section class="rounded-lg border border-gray-200 bg-white p-4">
+        <div class="flex items-center justify-between">
+          <h2 class="font-medium">Recent jobs</h2>
+          <RouterLink to="/jobs" class="text-sm text-blue-600 hover:underline">All jobs</RouterLink>
+        </div>
+        <table class="mt-3 w-full text-left text-sm">
+          <tbody>
+            <tr v-for="j in d.recent_jobs" :key="j.id" class="border-t border-gray-100">
+              <td class="py-1.5 font-mono text-xs">{{ j.kind }}</td>
+              <td
+                :class="
+                  j.status === 'failed'
+                    ? 'font-medium text-red-700'
+                    : j.status === 'running'
+                      ? 'text-blue-600'
+                      : 'text-green-700'
+                "
+              >
+                {{ j.status }}
+              </td>
+              <td class="text-xs text-gray-500">{{ fmt(j.started_at) }} · {{ j.triggered_by }}</td>
+            </tr>
+            <tr v-if="d.recent_jobs.length === 0">
+              <td class="py-3 text-gray-500">No jobs have run yet.</td>
+            </tr>
+          </tbody>
+        </table>
+      </section>
+    </div>
+  </template>
 </template>