Adding the new rewrite

.dockerignore

@@ -0,0 +1,25 @@
+.git
+.github
+.gitignore
+.venv
+venv
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.pytest_cache
+.coverage
+htmlcov
+.ruff_cache
+.mypy_cache
+tests/
+docs/
+README.md
+CLAUDE.md
+.env
+.env.*
+!.env.example
+*.md
+docker-compose.yml
+Dockerfile
+config/

.env.example

@@ -0,0 +1,41 @@
+# APC UPS Dashboard environment template
+# Copy to `.env` and fill in values. Never commit real .env files.
+
+# --- Required in production ---
+# Random 32+ byte string. Generate with:
+#   python -c "import secrets; print(secrets.token_urlsafe(48))"
+SESSION_SECRET=change-me-to-a-random-string
+
+# Admin password hash (argon2). Generate with:
+#   python -c "from passlib.hash import argon2; print(argon2.hash('mysecret'))"
+# If unset and no admin stored in Redis, the app shows a first-run setup page.
+ADMIN_USERNAME=admin
+# ADMIN_PASSWORD_HASH=
+
+# --- Redis ---
+REDIS_URL=redis://redis:6379/0
+# REDIS_PASSWORD=  # only used if you enable --requirepass in docker-compose
+
+# --- SMTP (optional; configure via Settings UI for host/port/etc) ---
+# SMTP password is read from env only; never stored in Redis.
+# SMTP_PASSWORD=
+
+# --- Networking / validation ---
+# Set false to block all RFC1918 IPs in UPS host field (default true for homelab).
+ALLOW_PRIVATE_IPS=true
+# Set true when behind an HTTPS reverse proxy (enables Secure cookie + HSTS).
+TRUST_PROXY=false
+
+# --- Observability ---
+LOG_LEVEL=INFO
+# Session expiry (seconds). Default 14 days.
+SESSION_MAX_AGE_SECONDS=1209600
+
+# --- Locale ---
+TZ=UTC
+
+# --- Rate limiting (set false in tests) ---
+RATE_LIMIT_ENABLED=true
+
+# --- Legacy YAML migration (unused unless you bind-mount /config) ---
+# UPS_CONFIG_PATH=/config/ups.yaml

.github/workflows/docker-image.yml

@@ -1,19 +1,45 @@
-name: Build and Push Docker Image
+name: CI - Lint, Test, and Push Docker Image
-# This workflow is manually triggered from the "Actions" tab in GitHub.
+
-# It requires a single input: the version tag of the flood repository to use.
 on:
   push:
     branches: [ "main" ]
   pull_request:
     branches: [ "main" ]
   schedule:
-    - cron: '0 0 3 * *'  # This runs at midnight on the first day of every month
+    - cron: '0 0 3 * *'  # monthly refresh
   workflow_dispatch:
 
 jobs:
+  lint-and-test:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.12'
+          cache: pip
+
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -r requirements.txt
+          pip install coverage
+
+      - name: Ruff lint
+        run: ruff check .
+
+      - name: Run pytest with coverage
+        run: |
+          coverage run -m pytest tests/ -v
+          coverage report
+
   build:
     runs-on: ubuntu-latest
-
+    needs: lint-and-test
+    if: github.event_name != 'pull_request'
     steps:
       - name: Set Image name
         run: |

Dockerfile

@@ -1,23 +1,48 @@
-FROM python:3.12-slim
+# syntax=docker/dockerfile:1.6
+FROM python:3.12.7-slim AS builder
 
 ENV PYTHONDONTWRITEBYTECODE=1 \
-    PYTHONUNBUFFERED=1
+    PYTHONUNBUFFERED=1 \
+    PIP_NO_CACHE_DIR=1
+
+WORKDIR /build
+
+RUN apt-get update \
+    && apt-get install -y --no-install-recommends build-essential gcc \
+    && rm -rf /var/lib/apt/lists/*
+
+COPY requirements.txt ./
+RUN pip install --upgrade pip && pip wheel --wheel-dir /wheels -r requirements.txt
+
+
+FROM python:3.12.7-slim AS runtime
+
+ENV PYTHONDONTWRITEBYTECODE=1 \
+    PYTHONUNBUFFERED=1 \
+    PIP_NO_CACHE_DIR=1 \
+    REDIS_URL=redis://redis:6379/0
 
 WORKDIR /app
 
 RUN apt-get update \
-    && apt-get install -y --no-install-recommends \
+    && apt-get install -y --no-install-recommends apcupsd curl \
-       build-essential \
+    && rm -rf /var/lib/apt/lists/* \
-       apcupsd \
+    && groupadd --system --gid 10001 appuser \
-    && rm -rf /var/lib/apt/lists/*
+    && useradd --system --uid 10001 --gid 10001 --home /app --shell /usr/sbin/nologin appuser
 
+COPY --from=builder /wheels /wheels
 COPY requirements.txt ./
-RUN pip install --no-cache-dir -r requirements.txt
+RUN pip install --no-index --find-links=/wheels -r requirements.txt \
+    && rm -rf /wheels
 
 COPY app ./app
 
+RUN chown -R appuser:appuser /app
+USER appuser
+
 EXPOSE 8000
 
-ENV REDIS_URL=redis://redis:6379/0
+HEALTHCHECK --interval=30s --timeout=5s --start-period=15s --retries=3 \
+  CMD curl -fsS http://localhost:8000/healthz || exit 1
 
 CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]

README.md

@@ -1,148 +1,124 @@
 # APC UPS Dashboard
 
-A lightweight FastAPI + Redis based dashboard for multiple APC UPS devices using the `apcaccess` CLI (apcupsd Network Information Server mode). Shows real-time metrics, lightweight charts, and maintains 7 days of historical snapshots.
+A production-ready FastAPI + Redis dashboard for monitoring multiple APC UPS devices via `apcupsd` NIS. Single-admin auth (argon2 + session cookies), CSRF, rate limiting, security headers, SSE live updates, 7-day history, connection-health tracking, battery degradation trends, HTML email alerts with batching and silent-hours, and Prometheus metrics.
 
 ## Features
-- Multiple UPS managed dynamically (stored in Redis; add/update/delete via API/UI)
-- Polling via `apcaccess` CLI against apcupsd NIS (default port 3551)
-- Connection test: TCP port reachability (no protocol parsing)
-- Real-time dashboard (Server Sent Events) updating key metrics
-- 7-day retention of snapshots in Redis lists
-- Simple Chart.js load percentage sparkline
-- Docker & docker-compose deployment (image bundles apcupsd + apcaccess)
-- SMTP alerting (high load, low battery %, on battery, low runtime) with cooldown
 
-## Configuration
+**Monitoring**
+- Multiple UPS managed dynamically via UI/REST (stored in Redis)
+- Real-time dashboard with SSE, fleet-overview panel, per-UPS state indicators
+- 7-day snapshot history, per-minute watts averages, daily energy + cost
+- Connection-health tracker — COMMLOST detection after 3× interval, recovery INFO alerts
+- Battery-health sampling (1/min) with 7-day trend + decline %
+- CSV export (history/events/energy), event log, alert management + ack
 
-Configuration is persisted in Redis (key `ups:config:json`). Use the web UI or the REST API to manage UPS entries and SMTP settings. A legacy `config/ups.yaml` (or path set via `UPS_CONFIG_PATH`) is imported once on first startup if Redis has no configuration; after migration the file is no longer written or read.
+**Alerts (email-only)**
+- Severity taxonomy: CRITICAL (ONBATT/COMMLOST/RUNTIME_LOW/BCHARGE_LOW), WARNING (LOAD_HIGH/REPLACEBATT/SELFTEST_FAIL/TEMP_HIGH/XFER_BURST/VOLT_DEV), INFO (REACHABLE/LINE_RESTORED)
+- Coalesced HTML emails (one per UPS per poll cycle), severity-colored
+- 30-min per-message cooldown, silent-hours deferral (CRITICAL always delivered), retries via tenacity
+- Test-email button and optional daily summary
 
-UPS fields:
+**Security & ops**
-- name (unique)
+- Single-admin auth (argon2 password, signed session cookie, double-submit CSRF)
-- host (apcupsd server hostname / IP)
+- SSRF host validation (rejects loopback/link-local; private IPs gated by `ALLOW_PRIVATE_IPS`)
-- port (default 3551)
+- SMTP password **only** from env — never persisted to Redis
-- interval_seconds (polling interval)
+- Subprocess timeout on `apcaccess` (10s), rate-limiting on auth/config, security headers + CSP
-- Optional alert thresholds: alert_loadpct_high, alert_bcharge_low, alert_on_battery, alert_runtime_low_minutes
+- Non-root container (UID 10001), pinned `python:3.12.7-slim` multi-stage build
+- `/healthz`, `/readyz`, `/metrics` (Prometheus), JSON structured logs with request-ID correlation
+- GitHub Actions pipeline runs ruff + pytest before building/publishing the image
 
-SMTP fields (optional): host, port, username, password (or env `SMTP_PASSWORD`), use_tls, use_ssl, from_addr, to_addrs[], subject_prefix.
+## Quick start (docker-compose)
 
-Environment secret: set `SMTP_PASSWORD` instead of storing cleartext.
-
-Alert suppression: identical alert per UPS suppressed for 30 minutes (cooldown).
-
-### apcupsd Server Requirements
-Your remote APC UPS hosts must be running `apcupsd` with the Network Information Server (NIS) enabled. Typical steps (on Linux):
-
-1. Install apcupsd (example for Debian/Ubuntu):
-  ```bash
-  sudo apt-get install apcupsd
-  ```
-2. Edit `/etc/apcupsd/apcupsd.conf` and confirm at least:
-  ```
-  UPSTYPE usb            # or 'net' / 'snmp' depending on your setup
-  DEVICE                 # usually blank for USB
-  NISIP 0.0.0.0          # listen on all interfaces (restrict in firewalled env)
-  NISPORT 3551           # must match the configured port (default 3551)
-  NETSERVER on           # enable network server
-  # Optional: restrict access (recommended)
-  ACCESS 192.168.1.0/24  # Only allow your monitoring subnet (supported on some builds)
-  ```
-3. Restart service:
-  ```bash
-  sudo systemctl restart apcupsd
-  ```
-4. Test locally:
-  ```bash
-  apcaccess status
-  nc -vz <server-ip> 3551
-  ```
-
-If you receive connectivity errors in logs:
- - Verify `NETSERVER on` is set.
- - Check host firewall (e.g., `ufw allow 3551/tcp`).
- - Confirm the port is correct and reachable from the container network.
- - Test manually:
-  ```bash
-  nc -vz <server-ip> 3551
-  apcaccess -h <server-ip>:3551 status
-  ```
-
-### Connection Testing Logic
-Simplified: only a raw TCP connect test. If the port is reachable it's reported as success. Polling uses the `apcaccess` CLI for data collection.
-
-## Run (docker-compose)
 ```bash
+cp .env.example .env
+# Required in production — edit .env:
+#   SESSION_SECRET=<run: python -c "import secrets; print(secrets.token_urlsafe(48))">
+#   SMTP_PASSWORD=<your SMTP password, if using alerts>
 docker compose up --build
 ```
-Visit http://localhost:8000
 
-### Persistence
+Open http://localhost:10280 — the app redirects to `/setup` where you create the admin account on first boot. Subsequent visits require login.
 
-Configuration and historical metrics live in Redis. The provided `docker-compose.yml` now mounts a named volume (`redis-data`) at `/data` inside the Redis container and enables Append Only File (AOF) with `--appendonly yes`.
+## Environment variables
 
-Data will persist across `docker compose down` / `up` cycles as long as you do NOT remove the volume. To explicitly remove all persisted configuration and history you must prune the volume:
+| Var | Required | Default | Notes |
+|---|---|---|---|
+| `SESSION_SECRET` | yes (prod) | ephemeral | 32+ random bytes, signs session cookies |
+| `ADMIN_USERNAME` | no | `admin` | |
+| `ADMIN_PASSWORD_HASH` | no | unset | Skip setup wizard by pre-seeding an argon2 hash |
+| `REDIS_URL` | no | `redis://redis:6379/0` | Supports `redis://:pw@host:port/db` |
+| `SMTP_PASSWORD` | no | unset | Read only from env, never stored |
+| `ALLOW_PRIVATE_IPS` | no | `true` | Set false to block RFC1918 UPS hosts |
+| `TRUST_PROXY` | no | `false` | Enable Secure cookies + HSTS when behind HTTPS proxy |
+| `LOG_LEVEL` | no | `INFO` | DEBUG/INFO/WARNING/ERROR |
+| `SESSION_MAX_AGE_SECONDS` | no | `1209600` | 14 days |
+| `RATE_LIMIT_ENABLED` | no | `true` | Disable in tests |
+| `TZ` | no | `UTC` | |
+
+Generate values:
+```bash
+# SESSION_SECRET
+python -c "import secrets; print(secrets.token_urlsafe(48))"
+# ADMIN_PASSWORD_HASH (optional pre-seed)
+python -c "from passlib.hash import argon2; print(argon2.hash('mysecret'))"
+```
+
+## apcupsd server requirements
+
+Your remote APC UPS hosts must run `apcupsd` with the Network Information Server (NIS) enabled:
+
+1. Install: `sudo apt-get install apcupsd`
+2. Edit `/etc/apcupsd/apcupsd.conf`:
+   ```
+   UPSTYPE usb
+   NISIP 0.0.0.0
+   NISPORT 3551
+   NETSERVER on
+   ```
+3. Restart: `sudo systemctl restart apcupsd`
+4. Test: `apcaccess status` and `nc -vz <server-ip> 3551`
+
+## Development
 
 ```bash
-docker compose down
+python3.12 -m venv .venv && . .venv/bin/activate
-docker volume rm apcupsd-client_redis-data  # volume name may be prefixed by folder/project
+pip install -r requirements.txt
+
+# Run tests (uses fakeredis)
+pytest tests/
+
+# With coverage
+coverage run -m pytest tests/ && coverage report
+
+# Lint
+ruff check .
+
+# Local dev server (needs apcaccess binary + REDIS_URL env)
+uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload
 ```
 
-If you had been losing configuration previously, ensure you pulled the updated compose file and that the `redis` service includes:
+## Persistence
-
-```yaml
-  redis:
-    volumes:
-      - redis-data:/data
-    command: ["redis-server", "--appendonly", "yes", "--appendfsync", "everysec"]
-```
-
-You can inspect Redis persistence files locally by running:
 
+Config and history live in Redis via an AOF-backed `redis-data` volume. To wipe state:
 ```bash
-docker compose exec redis ls -lh /data
+docker compose down && docker volume rm apcupsd-client_redis-data
 ```
 
-If you want to enforce periodic RDB snapshots as well, you can leave default save settings (remove the `--save ""` override). The current configuration uses AOF every second for a balance of durability and write performance.
+## API (summary)
 
-## Data Storage
+| Path | Auth | Notes |
-Redis stores:
+|---|---|---|
-- Latest snapshot hash: `ups:snap:<name>`
+| `/` (dashboard), `/config`, `/events`, `/alerts`, `/settings` | session | Jinja pages |
-- History list (JSON {ts,data}): `ups:hist:<name>`
+| `/login`, `/setup`, `/logout` | open/session | |
-- Per-minute watts averages: `ups:watts:permin:<name>`
+| `/healthz`, `/readyz`, `/metrics` | open | Prom metrics on `/metrics` |
-- Energy (watt-seconds) daily totals: `ups:energy:<name>:YYYYMMDD`
+| `GET /api/stream` | session | SSE snapshots |
-- Events list: `ups:event:list:<name>`
+| `GET /api/ups`, `/api/ups/{name}/{status\|history\|events\|energy\|health\|battery_health}` | session | |
-- Recent alerts: `ups:alerts:recent:<name>`
+| `GET /api/ups/fleet/overview` | session | Aggregate fleet summary |
-- Voltage deviation samples: `ups:volt:dev:samples:<name>`
+| `GET /api/ups/{name}/export?format=csv&kind=history\|events\|energy&since_days=N` | session | Streaming CSV |
-
+| `GET /api/events`, `/api/alerts`, `/api/alerts/active` | session | |
-A pruning task runs hourly removing entries older than 7 days.
+| `POST /api/alerts/{id}/ack` | session+CSRF | |
-
+| `GET/POST/PUT/DELETE /api/config/{ups,smtp,ui}/...` | session (+CSRF for writes) | Rate-limited |
-## Extending
-- Add more charts: query `/api/ups/<name>/history`
-- Add gauges: integrate a JS gauge lib in `dashboard.html`
-- Alerts: create background task checking thresholds
 
 ## License
+
 MIT
-
-## Security & Hardening Notes
-| Area | Current | Recommendation |
-|------|---------|---------------|
-| Authentication | None (open dashboard) | Add reverse proxy auth or FastAPI auth if exposed beyond LAN |
-| Config Storage | Redis (JSON) | Protect Redis with auth / network policy |
-| Network to apcupsd | Plain TCP | Use network segmentation / firewall; protocol has no encryption |
-| Redis | No auth configured | Enable AUTH / TLS if crossing trust boundaries |
-| Input Validation | Pydantic for config schema | Add stricter hostname/IP validation if multi-tenant |
-| Dependency Versions | Pinned | Review periodically for CVEs |
-| Logging | Polling errors logged | Avoid logging secrets; sanitize future additions |
-
-### Additional Notes
-- Legacy YAML migration (one-time) to Redis; file no longer updated afterward.
-- Dynamic poller reconciles tasks on config change (no restart needed).
-- Alert cooldown prevents email flood.
-- AOF-based Redis persistence keeps configuration across container rebuilds.
-
-### Suggested Future Enhancements
-- Optional Basic Auth / OIDC for web UI.
-- Rate limiting on config mutation endpoints.
-- CSRF protection if cookies/session auth added later.
-- Health endpoint (`/healthz`) returning Redis + config status.
-- Structured logging (JSON) for production observability.

app/alerts.py

@@ -1,105 +1,52 @@
+"""Alert evaluation, coalescing, cooldown, silent-hours, and dispatch."""
 from __future__ import annotations
-import smtplib
+
-import ssl
+import hashlib
-import os
+import logging
 import time
-from email.message import EmailMessage
+from dataclasses import asdict, dataclass
-from typing import Dict, Any, List
+from datetime import datetime
-from .config import load_config, SMTPConfig, UPSConfig
+from typing import Any
+
+from .config import SMTPConfig, UPSConfig, load_config
+from .notifications.email import EmailSendError, send_alert_email
 from .storage import get_redis
 
+logger = logging.getLogger(__name__)
+
 ALERT_COOLDOWN_SECONDS = 1800  # 30 minutes per distinct alert per UPS
 REDIS_ALERT_KEY_PREFIX = "ups:alert:last:"
+REDIS_PENDING_KEY_PREFIX = "ups:alerts:pending:"  # list of deferred messages
+REDIS_HISTORY_KEY_PREFIX = "ups:alerts:history:"  # global history
+ALERT_HISTORY_MAX = 500
+ALERT_HISTORY_TTL_SECONDS = 30 * 24 * 3600
 
 STATUS_ON_BATTERY_KEYWORDS = {"ONBATT", "ON BATTERY"}
+SELFTEST_FAIL_KEYWORDS = {"FAIL", "BADBATT"}
+
+SEV_CRITICAL = "CRITICAL"
+SEV_WARNING = "WARNING"
+SEV_INFO = "INFO"
+
+TEMP_HIGH_DEFAULT_C = 50.0
 
 
-def evaluate_alerts(ups_cfg: UPSConfig, snapshot: Dict[str, Any]) -> List[str]:
+@dataclass
-    messages: List[str] = []
+class Alert:
-    # Load % high
+    severity: str
-    if ups_cfg.alert_loadpct_high is not None:
+    message: str
-        loadpct = _to_float(snapshot.get('LOADPCT'))
+    code: str
-        if loadpct is not None and loadpct >= ups_cfg.alert_loadpct_high:
+    ups: str
-            messages.append(
+    ts: int
-                "Load percentage high: "
+
-                f"{loadpct}% >= {ups_cfg.alert_loadpct_high}%"
+    def hash(self) -> str:
-            )
+        h = hashlib.sha1(f"{self.ups}|{self.code}|{self.message}".encode()).hexdigest()
-    # Battery charge low
+        return h[:16]
-    if ups_cfg.alert_bcharge_low is not None:
+
-        bcharge = _to_float(snapshot.get('BCHARGE'))
+    def as_dict(self) -> dict[str, Any]:
-        if bcharge is not None and bcharge <= ups_cfg.alert_bcharge_low:
+        d = asdict(self)
-            messages.append(
+        d["id"] = self.hash()
-                "Battery charge low: "
+        return d
-                f"{bcharge}% <= {ups_cfg.alert_bcharge_low}%"
-            )
-    # On battery status
-    if ups_cfg.alert_on_battery:
-        status = str(snapshot.get('STATUS', '')).upper()
-        if any(k in status for k in STATUS_ON_BATTERY_KEYWORDS):
-            messages.append(f"UPS on battery: status={status}")
-    # Runtime low
-    if ups_cfg.alert_runtime_low_minutes is not None:
-        timeleft = snapshot.get('TIMELEFT')
-        # TIMELEFT often like '15.0 Minutes' -> attempt parse
-        runtime = _extract_leading_number(str(timeleft))
-        if (
-            runtime is not None
-            and runtime <= ups_cfg.alert_runtime_low_minutes
-        ):
-            messages.append(
-                "Runtime low: "
-                f"{runtime}m <= {ups_cfg.alert_runtime_low_minutes}m"
-            )
-    # Extended alerts based on global UI flags
-    cfg = load_config()
-    ui = cfg.ui if hasattr(cfg, 'ui') else None
-    if ui:
-        r = get_redis()
-        # Transfer burst: count status ONBATT events in last hour
-        if ui.enable_transfer_burst_alert:
-            # We rely on event list already capturing STATUS transitions
-            events = r.lrange(f"ups:event:list:{ups_cfg.name}", 0, 200)
-            now = int(time.time())
-            onbatt_count = 0
-            for ev in events:
-                parts = ev.split('|')
-                if len(parts) >= 3:
-                    try:
-                        ts = int(parts[0])
-                    except ValueError:
-                        continue
-                    if now - ts > 3600:
-                        continue
-                    if parts[1] == 'STATUS' and 'ONBATT' in parts[2]:
-                        onbatt_count += 1
-            if onbatt_count >= 3:  # threshold heuristically chosen
-                messages.append(
-                    f"Frequent battery events: {onbatt_count} in last hour"
-                )
-        # Voltage deviation: track LINEV vs nominal, average deviation window
-        if ui.enable_voltage_deviation_alert:
-            linev = _extract_leading_number(str(snapshot.get('LINEV', '')))
-            nom = _extract_leading_number(
-                str(snapshot.get('NOMINV', snapshot.get('NOMINPUT', '')))
-            )
-            if linev and nom:
-                dev_pct = abs(linev - nom) / nom * 100.0
-                dev_key = f"ups:volt:dev:samples:{ups_cfg.name}"
-                r.lpush(dev_key, f"{dev_pct:.2f}")
-                r.ltrim(dev_key, 0, 49)
-                samples = r.lrange(dev_key, 0, -1)
-                try:
-                    avg_dev = sum(float(s) for s in samples) / max(
-                        1, len(samples)
-                    )
-                    if avg_dev > 8.0 and len(samples) >= 10:
-                        messages.append(
-                            "High average voltage deviation: "
-                            f"{avg_dev:.1f}% over {len(samples)} samples"
-                        )
-                except Exception:
-                    pass
-    return messages
 
 
 def _to_float(val) -> float | None:
@@ -111,66 +58,251 @@ def _to_float(val) -> float | None:
 
 def _extract_leading_number(s: str) -> float | None:
     try:
-        part = s.strip().split()[0]
+        return float(s.strip().split()[0])
-        return float(part)
     except Exception:
         return None
 
 
-def _cooldown_key(ups_name: str, msg: str) -> str:
+def evaluate_alerts(ups_cfg: UPSConfig, snapshot: dict[str, Any]) -> list[Alert]:
-    return f"{REDIS_ALERT_KEY_PREFIX}{ups_name}:{hash(msg)}"
+    msgs: list[Alert] = []
+    now = int(time.time())
+    name = ups_cfg.name
 
+    def add(sev: str, code: str, message: str) -> None:
+        msgs.append(Alert(sev, message, code, name, now))
 
-def send_alert_email(smtp_cfg: SMTPConfig, ups_name: str, messages: List[str]):
+    status = str(snapshot.get("STATUS", "")).upper()
-    password = smtp_cfg.password or os.environ.get('SMTP_PASSWORD')
+    # ONBATT
-    if not smtp_cfg.to_addrs:
+    if ups_cfg.alert_on_battery and any(k in status for k in STATUS_ON_BATTERY_KEYWORDS):
-        return
+        add(SEV_CRITICAL, "ONBATT", f"UPS on battery: status={status}")
-    subject = f"{smtp_cfg.subject_prefix} {ups_name} alert"
+    # Runtime low
-    body = "\n".join(messages)
+    if ups_cfg.alert_runtime_low_minutes is not None:
-    msg = EmailMessage()
+        runtime = _extract_leading_number(str(snapshot.get("TIMELEFT", "")))
-    msg['Subject'] = subject
+        if runtime is not None and runtime <= ups_cfg.alert_runtime_low_minutes:
-    msg['From'] = smtp_cfg.from_addr or (smtp_cfg.username or 'ups@example')
+            add(
-    msg['To'] = ", ".join(smtp_cfg.to_addrs)
+                SEV_CRITICAL,
-    msg.set_content(body)
+                "RUNTIME_LOW",
+                f"Runtime low: {runtime}m <= {ups_cfg.alert_runtime_low_minutes}m",
+            )
+    # Battery charge low
+    if ups_cfg.alert_bcharge_low is not None:
+        bcharge = _extract_leading_number(str(snapshot.get("BCHARGE", "")))
+        if bcharge is not None and bcharge <= ups_cfg.alert_bcharge_low:
+            add(
+                SEV_CRITICAL,
+                "BCHARGE_LOW",
+                f"Battery charge low: {bcharge}% <= {ups_cfg.alert_bcharge_low}%",
+            )
+    # Load high
+    if ups_cfg.alert_loadpct_high is not None:
+        loadpct = _extract_leading_number(str(snapshot.get("LOADPCT", "")))
+        if loadpct is not None and loadpct >= ups_cfg.alert_loadpct_high:
+            add(
+                SEV_WARNING,
+                "LOAD_HIGH",
+                f"Load high: {loadpct}% >= {ups_cfg.alert_loadpct_high}%",
+            )
+    # REPLACEBATT flag
+    replacebatt = str(snapshot.get("REPLACEBATT", "")).upper()
+    if replacebatt and replacebatt not in {"NO", "", "0"}:
+        add(SEV_WARNING, "REPLACEBATT", "UPS reports battery needs replacement")
+    # SELFTEST fail
+    selftest = str(snapshot.get("SELFTEST", "")).upper()
+    if any(k in selftest for k in SELFTEST_FAIL_KEYWORDS):
+        add(SEV_WARNING, "SELFTEST_FAIL", f"Self-test failure: {selftest}")
+    # Temperature high
+    threshold = ups_cfg.alert_itemp_high or TEMP_HIGH_DEFAULT_C
+    itemp = _extract_leading_number(str(snapshot.get("ITEMP", "")))
+    if itemp is not None and itemp >= threshold:
+        add(
+            SEV_WARNING,
+            "TEMP_HIGH",
+            f"Internal temperature high: {itemp}C >= {threshold}C",
+        )
 
-    if smtp_cfg.use_ssl:
+    # Extended alerts based on UI flags
-        context = ssl.create_default_context()
-        with smtplib.SMTP_SSL(
-            smtp_cfg.host, smtp_cfg.port, context=context, timeout=30
-        ) as server:
-            if smtp_cfg.username and password:
-                server.login(smtp_cfg.username, password)
-            server.send_message(msg)
-    else:
-        with smtplib.SMTP(smtp_cfg.host, smtp_cfg.port, timeout=30) as server:
-            if smtp_cfg.use_tls:
-                server.starttls(context=ssl.create_default_context())
-            if smtp_cfg.username and password:
-                server.login(smtp_cfg.username, password)
-            server.send_message(msg)
-
- 
-def process_alerts(ups_cfg: UPSConfig, snapshot: Dict[str, Any]):
     cfg = load_config()
-    if not cfg.smtp:
+    ui = cfg.ui if hasattr(cfg, "ui") else None
-        return
+    if ui:
-    msgs = evaluate_alerts(ups_cfg, snapshot)
+        r = get_redis()
-    if not msgs:
+        if ui.enable_transfer_burst_alert:
+            events = r.lrange(f"ups:event:list:{name}", 0, 200)
+            onbatt_count = 0
+            for ev in events:
+                parts = ev.split("|")
+                if len(parts) >= 3:
+                    try:
+                        ts = int(parts[0])
+                    except ValueError:
+                        continue
+                    if now - ts > 3600:
+                        continue
+                    if parts[1] == "STATUS" and "ONBATT" in parts[2]:
+                        onbatt_count += 1
+            if onbatt_count >= 3:
+                add(
+                    SEV_WARNING,
+                    "XFER_BURST",
+                    f"Frequent battery events: {onbatt_count} in last hour",
+                )
+        if ui.enable_voltage_deviation_alert:
+            linev = _extract_leading_number(str(snapshot.get("LINEV", "")))
+            nom = _extract_leading_number(
+                str(snapshot.get("NOMINV", snapshot.get("NOMINPUT", "")))
+            )
+            if linev and nom:
+                dev_pct = abs(linev - nom) / nom * 100.0
+                dev_key = f"ups:volt:dev:samples:{name}"
+                r.lpush(dev_key, f"{dev_pct:.2f}")
+                r.ltrim(dev_key, 0, 49)
+                samples = r.lrange(dev_key, 0, -1)
+                try:
+                    avg_dev = sum(float(s) for s in samples) / max(1, len(samples))
+                    if avg_dev > 8.0 and len(samples) >= 10:
+                        add(
+                            SEV_WARNING,
+                            "VOLT_DEV",
+                            f"High voltage deviation: {avg_dev:.1f}% over {len(samples)} samples",
+                        )
+                except Exception:
+                    pass
+    return msgs
+
+
+def _cooldown_key(alert: Alert) -> str:
+    return f"{REDIS_ALERT_KEY_PREFIX}{alert.ups}:{alert.hash()}"
+
+
+def _in_silent_hours(smtp_cfg: SMTPConfig, now: datetime | None = None) -> bool:
+    if smtp_cfg.silent_hours_start is None or smtp_cfg.silent_hours_end is None:
+        return False
+    if smtp_cfg.silent_hours_start == smtp_cfg.silent_hours_end:
+        return False
+    if now is None:
+        now = datetime.now()
+    h = now.hour
+    s = smtp_cfg.silent_hours_start
+    e = smtp_cfg.silent_hours_end
+    if s < e:
+        return s <= h < e
+    # Wraps past midnight
+    return h >= s or h < e
+
+
+def _persist_history(alerts: list[Alert]) -> None:
+    if not alerts:
         return
     r = get_redis()
-    to_send: List[str] = []
-    now = int(time.time())
-    for m in msgs:
-        key = _cooldown_key(ups_cfg.name, m)
-        if not r.get(key):
-            to_send.append(m)
-            r.set(key, now, ex=ALERT_COOLDOWN_SECONDS)
-    if to_send:
-        # Store recent alerts with timestamp for health reporting
-        recent_key = f"ups:alerts:recent:{ups_cfg.name}"
     pipe = r.pipeline()
-        for m in to_send:
+    for a in alerts:
-            pipe.lpush(recent_key, f"{now}|{m}")
+        pipe.lpush(
-        pipe.ltrim(recent_key, 0, 49)  # keep last 50
+            REDIS_HISTORY_KEY_PREFIX + "all",
+            f"{a.ts}|{a.severity}|{a.ups}|{a.code}|{a.message}|{a.hash()}",
+        )
+        pipe.lpush(
+            f"ups:alerts:recent:{a.ups}", f"{a.ts}|{a.severity}|{a.message}"
+        )
+    pipe.ltrim(REDIS_HISTORY_KEY_PREFIX + "all", 0, ALERT_HISTORY_MAX - 1)
+    pipe.expire(REDIS_HISTORY_KEY_PREFIX + "all", ALERT_HISTORY_TTL_SECONDS)
+    for a in alerts:
+        pipe.ltrim(f"ups:alerts:recent:{a.ups}", 0, 49)
     pipe.execute()
-        send_alert_email(cfg.smtp, ups_cfg.name, to_send)
+
+
+def _defer_for_silent_hours(alerts: list[Alert]) -> None:
+    r = get_redis()
+    for a in alerts:
+        key = REDIS_PENDING_KEY_PREFIX + a.ups
+        r.lpush(key, f"{a.ts}|{a.severity}|{a.code}|{a.message}")
+    # Keep at most 200 deferred
+    for ups in {a.ups for a in alerts}:
+        r.ltrim(REDIS_PENDING_KEY_PREFIX + ups, 0, 199)
+
+
+def drain_deferred(ups_name: str) -> list[Alert]:
+    r = get_redis()
+    key = REDIS_PENDING_KEY_PREFIX + ups_name
+    raw = r.lrange(key, 0, -1)
+    r.delete(key)
+    out: list[Alert] = []
+    for item in raw:
+        parts = item.split("|", 3)
+        if len(parts) == 4:
+            try:
+                ts = int(parts[0])
+            except ValueError:
+                continue
+            out.append(Alert(parts[1], parts[3], parts[2], ups_name, ts))
+    return out
+
+
+def dispatch_alerts(
+    smtp_cfg: SMTPConfig | None,
+    ups_name: str,
+    alerts: list[Alert],
+    dashboard_url: str = "http://localhost:8000/",
+) -> None:
+    """Send a coalesced email containing all alerts for one UPS."""
+    if not smtp_cfg or not alerts:
+        return
+    payload = [
+        {"severity": a.severity, "message": a.message, "code": a.code}
+        for a in alerts
+    ]
+    try:
+        send_alert_email(smtp_cfg, ups_name, payload, dashboard_url=dashboard_url)
+    except EmailSendError as e:
+        logger.warning("Alert email failed for %s: %s", ups_name, e)
+
+
+def process_alerts(ups_cfg: UPSConfig, snapshot: dict[str, Any]) -> list[Alert]:
+    """Evaluate alerts, apply cooldown + silent hours, dispatch email.
+
+    Returns the list of alerts that were considered fresh (not cooled down).
+    """
+    cfg = load_config()
+    all_alerts = evaluate_alerts(ups_cfg, snapshot)
+    if not all_alerts:
+        return []
+    r = get_redis()
+    fresh: list[Alert] = []
+    now = int(time.time())
+    for a in all_alerts:
+        key = _cooldown_key(a)
+        if not r.get(key):
+            fresh.append(a)
+            r.set(key, now, ex=ALERT_COOLDOWN_SECONDS)
+    if not fresh:
+        return []
+    _persist_history(fresh)
+    if not cfg.smtp:
+        return fresh
+    # Silent hours filter: keep CRITICAL, defer others
+    if _in_silent_hours(cfg.smtp):
+        critical = [a for a in fresh if a.severity == SEV_CRITICAL]
+        deferred = [a for a in fresh if a.severity != SEV_CRITICAL]
+        if deferred:
+            _defer_for_silent_hours(deferred)
+        if critical:
+            dispatch_alerts(cfg.smtp, ups_cfg.name, critical)
+    else:
+        # Also flush any deferred alerts for this UPS
+        deferred = drain_deferred(ups_cfg.name)
+        dispatch_alerts(cfg.smtp, ups_cfg.name, fresh + deferred)
+    return fresh
+
+
+def emit_info(ups_name: str, code: str, message: str) -> None:
+    """Record and dispatch a single INFO alert (e.g. recovery events)."""
+    alert = Alert(SEV_INFO, message, code, ups_name, int(time.time()))
+    r = get_redis()
+    if r.get(_cooldown_key(alert)):
+        return
+    r.set(_cooldown_key(alert), int(time.time()), ex=ALERT_COOLDOWN_SECONDS)
+    _persist_history([alert])
+    cfg = load_config()
+    if cfg.smtp:
+        if _in_silent_hours(cfg.smtp):
+            _defer_for_silent_hours([alert])
+        else:
+            dispatch_alerts(cfg.smtp, ups_name, [alert])

app/apc_cli.py

@@ -1,22 +1,24 @@
 from __future__ import annotations
+
 import asyncio
-import shutil
 import logging
-from typing import Dict, Any
+import shutil
+from typing import Any
 
 logger = logging.getLogger(__name__)
 
 APCACCESS_BIN = shutil.which('apcaccess') or 'apcaccess'
+APCACCESS_TIMEOUT_SECONDS = 10.0
 
 
 class APCStatusError(Exception):
     pass
 
 
-async def fetch_status(host: str, port: int) -> Dict[str, Any]:
+async def fetch_status(host: str, port: int) -> dict[str, Any]:
     """Invoke apcaccess CLI and parse key:value lines into a dict.
 
-    Raises APCStatusError if binary fails or returns no data.
+    Raises APCStatusError if binary fails, times out, or returns no data.
     """
     proc = await asyncio.create_subprocess_exec(
         APCACCESS_BIN,
@@ -25,14 +27,26 @@ async def fetch_status(host: str, port: int) -> Dict[str, Any]:
         stdout=asyncio.subprocess.PIPE,
         stderr=asyncio.subprocess.PIPE,
     )
-    stdout, stderr = await proc.communicate()
+    try:
+        stdout, stderr = await asyncio.wait_for(
+            proc.communicate(), timeout=APCACCESS_TIMEOUT_SECONDS
+        )
+    except TimeoutError:
+        try:
+            proc.kill()
+            await proc.wait()
+        except ProcessLookupError:
+            pass
+        raise APCStatusError(
+            f"apcaccess timed out after {APCACCESS_TIMEOUT_SECONDS}s for {host}:{port}"
+        )
     if proc.returncode != 0:
         err_txt = stderr.decode(errors="replace").strip()
         raise APCStatusError(
             f'apcaccess exit {proc.returncode}: {err_txt}'
         )
     text = stdout.decode(errors='replace')
-    data: Dict[str, Any] = {}
+    data: dict[str, Any] = {}
     for line in text.splitlines():
         if ':' not in line:
             continue
@@ -40,7 +54,8 @@ async def fetch_status(host: str, port: int) -> Dict[str, Any]:
         key = k.strip()
         val = v.strip()
         data[key] = val
-    # Normalize expected fields/aliases
+    if not data:
+        raise APCStatusError("apcaccess returned no data")
     if 'UPSNAME' not in data and 'NAME' in data:
         data['UPSNAME'] = data['NAME']
     if 'MODEL' in data:

app/auth.py

@@ -0,0 +1,159 @@
+"""Authentication: argon2 password hashing + signed session cookies.
+
+Single-admin design for homelab use. Credentials set via env vars:
+  ADMIN_USERNAME, ADMIN_PASSWORD_HASH, SESSION_SECRET
+
+If ADMIN_PASSWORD_HASH is unset, the first-run setup page generates one
+and stores it in Redis under the key ``ups:admin:hash``. The stored hash
+takes precedence over env vars so the operator can change their password
+via the UI.
+"""
+from __future__ import annotations
+
+import secrets
+
+from fastapi import Depends, HTTPException, Request, Response
+from itsdangerous import BadSignature, SignatureExpired, URLSafeTimedSerializer
+from passlib.hash import argon2
+
+from .settings import settings
+from .storage import get_redis
+
+SESSION_COOKIE = "ups_session"
+CSRF_COOKIE = "csrf_token"
+CSRF_HEADER = "X-CSRF-Token"
+ADMIN_HASH_KEY = "ups:admin:hash"
+ADMIN_USER_KEY = "ups:admin:username"
+
+
+def _serializer() -> URLSafeTimedSerializer:
+    secret = settings.session_secret
+    if not secret:
+        # Best effort ephemeral secret so we don't crash at import time in dev
+        secret = "dev-insecure-change-me-please"
+    return URLSafeTimedSerializer(secret, salt="ups-session")
+
+
+def hash_password(password: str) -> str:
+    """Produce an argon2 hash of the given password."""
+    return argon2.hash(password)
+
+
+def verify_password(password: str, hashed: str) -> bool:
+    try:
+        return argon2.verify(password, hashed)
+    except Exception:
+        return False
+
+
+def get_stored_admin() -> tuple[str | None, str | None]:
+    """Return (username, password_hash) from Redis or settings.
+
+    Redis-stored credentials take precedence when present.
+    """
+    try:
+        r = get_redis()
+        redis_hash = r.get(ADMIN_HASH_KEY)
+        redis_user = r.get(ADMIN_USER_KEY)
+    except Exception:
+        redis_hash = None
+        redis_user = None
+    username = redis_user or settings.admin_username
+    pw_hash = redis_hash or settings.admin_password_hash
+    return username, pw_hash
+
+
+def store_admin(username: str, password: str) -> None:
+    """Persist admin credentials to Redis."""
+    r = get_redis()
+    r.set(ADMIN_USER_KEY, username)
+    r.set(ADMIN_HASH_KEY, hash_password(password))
+
+
+def is_admin_configured() -> bool:
+    _, pw_hash = get_stored_admin()
+    return bool(pw_hash)
+
+
+def create_session_token(username: str) -> str:
+    return _serializer().dumps({"u": username})
+
+
+def verify_session_token(token: str) -> str | None:
+    try:
+        data = _serializer().loads(
+            token, max_age=settings.session_max_age_seconds
+        )
+        if isinstance(data, dict):
+            return data.get("u")
+    except (BadSignature, SignatureExpired):
+        return None
+    return None
+
+
+def make_csrf_token() -> str:
+    return secrets.token_urlsafe(32)
+
+
+def set_auth_cookies(response: Response, username: str) -> str:
+    """Set session + CSRF cookies on the response. Returns CSRF token."""
+    session_token = create_session_token(username)
+    csrf_token = make_csrf_token()
+    secure = settings.trust_proxy
+    response.set_cookie(
+        SESSION_COOKIE,
+        session_token,
+        max_age=settings.session_max_age_seconds,
+        httponly=True,
+        samesite="lax",
+        secure=secure,
+        path="/",
+    )
+    response.set_cookie(
+        CSRF_COOKIE,
+        csrf_token,
+        max_age=settings.session_max_age_seconds,
+        httponly=False,  # JS needs to read this
+        samesite="lax",
+        secure=secure,
+        path="/",
+    )
+    return csrf_token
+
+
+def clear_auth_cookies(response: Response) -> None:
+    response.delete_cookie(SESSION_COOKIE, path="/")
+    response.delete_cookie(CSRF_COOKIE, path="/")
+
+
+def current_user(request: Request) -> str | None:
+    """Return current session username or None."""
+    token = request.cookies.get(SESSION_COOKIE)
+    if not token:
+        return None
+    return verify_session_token(token)
+
+
+def require_session(request: Request) -> str:
+    """FastAPI dependency - 401 if not authenticated."""
+    user = current_user(request)
+    if not user:
+        raise HTTPException(status_code=401, detail="Authentication required")
+    return user
+
+
+def require_csrf(request: Request) -> None:
+    """Enforce double-submit CSRF token on mutating requests."""
+    if request.method in ("GET", "HEAD", "OPTIONS"):
+        return
+    cookie_token = request.cookies.get(CSRF_COOKIE)
+    header_token = request.headers.get(CSRF_HEADER)
+    if not cookie_token or not header_token or cookie_token != header_token:
+        raise HTTPException(status_code=403, detail="CSRF token invalid or missing")
+
+
+def require_session_and_csrf(
+    request: Request, user: str = Depends(require_session)
+) -> str:
+    require_csrf(request)
+    return user

app/config.py

@@ -1,55 +1,116 @@
 from __future__ import annotations
-from pydantic import BaseModel, Field
-from typing import List, Optional
-from pathlib import Path
-import os
 
-CONFIG_PATH = Path(
+import ipaddress
-    os.environ.get("UPS_CONFIG_PATH", "/config/ups.yaml")
+import re
-)  # legacy path for migration
+from pathlib import Path
+from typing import ClassVar
+
+from pydantic import BaseModel, Field, field_validator
+
+from .settings import settings
+
+CONFIG_PATH = Path(settings.ups_config_path)  # legacy path for migration
+
+_NAME_RE = re.compile(r"^[a-zA-Z0-9_-]{1,32}$")
+_HOSTNAME_RE = re.compile(
+    r"^(?=.{1,253}$)([a-zA-Z0-9]([a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)(\.[a-zA-Z0-9]([a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*$"
+)
+
+_BLOCKED_LITERALS = {"localhost", "localhost.localdomain"}
+
+
+def _validate_host_string(host: str) -> str:
+    """Validate a host field. Rejects loopback/link-local always; rejects
+    private ranges when ALLOW_PRIVATE_IPS is false.
+    """
+    host = host.strip()
+    if not host:
+        raise ValueError("host must not be empty")
+    if host.lower() in _BLOCKED_LITERALS:
+        raise ValueError(f"host '{host}' is not permitted (loopback alias)")
+    # Try as IP first
+    try:
+        ip = ipaddress.ip_address(host)
+        if ip.is_loopback or ip.is_link_local or ip.is_unspecified or ip.is_multicast:
+            raise ValueError(f"host IP {host} is not a valid polling target")
+        if ip.is_private and not settings.allow_private_ips:
+            raise ValueError(
+                f"private IP {host} not allowed (set ALLOW_PRIVATE_IPS=true)"
+            )
+        return host
+    except ValueError as ip_err:
+        # Fall through to hostname validation only when the string wasn't a
+        # parseable IP address
+        if "does not appear to be an IPv4 or IPv6 address" not in str(ip_err):
+            raise
+    # Validate as hostname
+    if not _HOSTNAME_RE.match(host):
+        raise ValueError(f"host '{host}' is not a valid hostname or IP")
+    return host
 
 
 class UPSConfig(BaseModel):
     name: str = Field(..., description="Friendly UPS name")
     host: str = Field(..., description="apcupsd NIS host/IP")
-    port: int = Field(3551, description="apcupsd NIS port")
+    port: int = Field(3551, ge=1, le=65535, description="apcupsd NIS port")
-    interval_seconds: int = Field(30, description="Polling interval")
+    interval_seconds: int = Field(
-    # Alert thresholds (any optional)
+        30, ge=5, le=3600, description="Polling interval"
-    alert_loadpct_high: Optional[float] = Field(
-        None, description="Trigger if LOADPCT >= value"
     )
-    alert_bcharge_low: Optional[float] = Field(
+    # Alert thresholds (any optional)
-        None, description="Trigger if BCHARGE <= value"
+    alert_loadpct_high: float | None = Field(
+        None, ge=0, le=100, description="Trigger if LOADPCT >= value"
+    )
+    alert_bcharge_low: float | None = Field(
+        None, ge=0, le=100, description="Trigger if BCHARGE <= value"
     )
     alert_on_battery: bool = Field(
         False, description="Trigger when STATUS indicates on battery"
     )
-    alert_runtime_low_minutes: Optional[float] = Field(
+    alert_runtime_low_minutes: float | None = Field(
-        None, description="Trigger if TIMELEFT <= minutes"
+        None, ge=0, description="Trigger if TIMELEFT <= minutes"
     )
+    alert_itemp_high: float | None = Field(
+        None, ge=0, le=120, description="Trigger if internal temp >= C"
+    )
+
+    @field_validator("name")
+    @classmethod
+    def _validate_name(cls, v: str) -> str:
+        if not _NAME_RE.match(v):
+            raise ValueError(
+                "name must be 1-32 chars, alphanumeric / underscore / dash only"
+            )
+        return v
+
+    @field_validator("host")
+    @classmethod
+    def _validate_host(cls, v: str) -> str:
+        return _validate_host_string(v)
 
 
 class SMTPConfig(BaseModel):
     host: str = Field(..., description="SMTP server host/IP")
-    port: int = Field(..., description="SMTP port")
+    port: int = Field(..., ge=1, le=65535, description="SMTP port")
-    username: Optional[str] = Field(None)
+    username: str | None = Field(None)
-    password: Optional[str] = Field(
+    # password is never persisted in Redis; read from env SMTP_PASSWORD
-        None, description="Plain password or set via env SMTP_PASSWORD"
-    )
     use_tls: bool = Field(False, description="STARTTLS if true")
     use_ssl: bool = Field(False, description="SSL (smtplib.SMTP_SSL) if true")
-    from_addr: Optional[str] = Field(None, description="From email address")
+    from_addr: str | None = Field(None, description="From email address")
-    to_addrs: List[str] = Field(
+    to_addrs: list[str] = Field(
         default_factory=list, description="Recipient list"
     )
     subject_prefix: str = Field("[UPS]", description="Subject prefix")
+    silent_hours_start: int | None = Field(
+        None, ge=0, le=23, description="Silent window start hour (local time)"
+    )
+    silent_hours_end: int | None = Field(
+        None, ge=0, le=23, description="Silent window end hour (local time)"
+    )
+    daily_summary_hour: int | None = Field(
+        None, ge=0, le=23, description="Hour to send daily summary"
+    )
 
 
-class AppConfig(BaseModel):
-    ups: List[UPSConfig]
-    smtp: Optional[SMTPConfig] = None
-    # UI feature flags (optional; default values used if missing)
-    # Added for dashboard toggleable features
 class UIConfig(BaseModel):
     show_events: bool = True
     show_energy: bool = False
@@ -60,15 +121,26 @@ class AppConfig(BaseModel):
     allow_resize: bool = True
     enable_transfer_burst_alert: bool = False
     enable_voltage_deviation_alert: bool = False
+    energy_cost_per_kwh: float = 0.0
 
-    ui: UIConfig = UIConfig()
+
+class AppConfig(BaseModel):
+    ups: list[UPSConfig] = Field(default_factory=list)
+    smtp: SMTPConfig | None = None
+    ui: UIConfig = Field(default_factory=UIConfig)
+
+    # Backward-compat alias: older code references AppConfig.UIConfig
+    UIConfig: ClassVar[type[UIConfig]] = UIConfig
 
 
 _cached: AppConfig | None = None
 
 
 def load_config(path: Path = CONFIG_PATH) -> AppConfig:
-    # Use redis store (lazy import to avoid circular)
+    """Load config from Redis, cached in-process.
+
+    Invalidate via ``config_module._cached = None`` after any write.
+    """
     from .config_store import load_config_redis
     global _cached
     if _cached:

app/config_manager.py

@@ -1,27 +1,20 @@
 from __future__ import annotations
-from typing import List, Optional, Dict, Any
+
 import asyncio
 import logging
 import socket
+from typing import Any
 
-try:
 from pydantic import BaseModel, Field, ValidationError
-except ImportError:
-    # Fallback for environments without pydantic
-    BaseModel = object
 
-    def Field(default=None, **kwargs):
+from .config import AppConfig, SMTPConfig, UIConfig, UPSConfig
-        return default
-    
-    ValidationError = ValueError
-
-from .config import UPSConfig, SMTPConfig, AppConfig
 from .config_store import load_config_redis, save_config_redis
+from .settings import settings
 
 logger = logging.getLogger(__name__)
 
-# Incremented every time configuration is modified so
+# Incremented every time configuration is modified so SSE clients can
-# SSE clients can detect changes
+# detect changes
 _config_version: int = 0
 
 
@@ -29,34 +22,25 @@ def get_config_version() -> int:
     return _config_version
 
 
-class ConfigWriteError(Exception):  # retained for API compatibility
+class ConfigWriteError(Exception):
     """Raised when configuration cannot be written (kept for compatibility)."""
 
 
 class UPSConfigUpdate(BaseModel):
-    """Model for updating UPS configuration"""
+    """Model for updating UPS configuration."""
-    name: Optional[str] = Field(None, description="Friendly UPS name")
+    name: str | None = None
-    host: Optional[str] = Field(None, description="apcupsd NIS host/IP")
+    host: str | None = None
-    port: Optional[int] = Field(None, description="apcupsd NIS port")
+    port: int | None = Field(None, ge=1, le=65535)
-    interval_seconds: Optional[int] = Field(
+    interval_seconds: int | None = Field(None, ge=5, le=3600)
-        None, description="Polling interval"
+    alert_loadpct_high: float | None = Field(None, ge=0, le=100)
-    )
+    alert_bcharge_low: float | None = Field(None, ge=0, le=100)
-    alert_loadpct_high: Optional[float] = Field(
+    alert_on_battery: bool | None = None
-        None, description="Trigger if LOADPCT >= value"
+    alert_runtime_low_minutes: float | None = Field(None, ge=0)
-    )
+    alert_itemp_high: float | None = Field(None, ge=0, le=120)
-    alert_bcharge_low: Optional[float] = Field(
-        None, description="Trigger if BCHARGE <= value"
-    )
-    alert_on_battery: Optional[bool] = Field(
-        None, description="Trigger when STATUS indicates on battery"
-    )
-    alert_runtime_low_minutes: Optional[float] = Field(
-        None, description="Trigger if TIMELEFT <= minutes"
-    )
 
 
 class ConfigManager:
-    def __init__(self):
+    def __init__(self) -> None:
         self._lock = asyncio.Lock()
 
     async def load_config(self) -> AppConfig:
@@ -67,26 +51,19 @@ class ConfigManager:
         async with self._lock:
             save_config_redis(config)
             logger.info("Configuration saved to Redis")
-            # Invalidate cached global config so subsequent
-            # load_config() calls see changes
             try:
                 from . import config as config_module
                 config_module._cached = None
             except Exception:  # pragma: no cover - defensive
-                logger.debug(
+                logger.debug("Failed to invalidate config cache", exc_info=True)
-                    "Failed to invalidate config cache", exc_info=True
-                )
-            # bump config version
             global _config_version
             _config_version += 1
 
-    async def get_ups_list(self) -> List[UPSConfig]:
+    async def get_ups_list(self) -> list[UPSConfig]:
-        """Get list of all UPS configurations"""
         config = await self.load_config()
         return config.ups
 
-    async def get_ups(self, name: str) -> Optional[UPSConfig]:
+    async def get_ups(self, name: str) -> UPSConfig | None:
-        """Get UPS configuration by name"""
         config = await self.load_config()
         for ups in config.ups:
             if ups.name == name:
@@ -94,110 +71,78 @@ class ConfigManager:
         return None
 
     async def add_ups(self, ups_config: UPSConfig) -> bool:
-        """Add new UPS configuration"""
         config = await self.load_config()
-        
-        # Check if UPS with same name already exists
         if any(ups.name == ups_config.name for ups in config.ups):
             raise ValueError(
                 f"UPS with name '{ups_config.name}' already exists"
             )
-        
         config.ups.append(ups_config)
         await self.save_config(config)
-        
-    # No file cache now
-        
         return True
 
     async def update_ups(self, name: str, updates: UPSConfigUpdate) -> bool:
-        """Update existing UPS configuration"""
         config = await self.load_config()
-        
         ups_index = None
         for i, ups in enumerate(config.ups):
             if ups.name == name:
                 ups_index = i
                 break
-        
         if ups_index is None:
             return False
-        
-        # Apply updates
         ups_dict = config.ups[ups_index].model_dump()
         update_dict = updates.model_dump(exclude_none=True)
         ups_dict.update(update_dict)
-        
-        # Validate updated configuration
         try:
             updated_ups = UPSConfig(**ups_dict)
         except ValidationError as e:
             raise ValueError(f"Invalid configuration: {e}")
-        
         config.ups[ups_index] = updated_ups
         await self.save_config(config)
-        
-    # No file cache now
-        
         return True
 
     async def delete_ups(self, name: str) -> bool:
-        """Delete UPS configuration"""
         config = await self.load_config()
-        
         original_count = len(config.ups)
         config.ups = [ups for ups in config.ups if ups.name != name]
-        
         if len(config.ups) == original_count:
-            return False  # UPS not found
+            return False
-        
         await self.save_config(config)
-        
-    # No file cache now
-        
         return True
 
-    async def get_smtp_config(self) -> Optional[SMTPConfig]:
+    async def get_smtp_config(self) -> SMTPConfig | None:
-        """Get SMTP configuration"""
         config = await self.load_config()
         return config.smtp
 
     async def update_smtp_config(self, smtp_config: SMTPConfig) -> None:
-        """Update SMTP configuration"""
         config = await self.load_config()
         config.smtp = smtp_config
         await self.save_config(config)
 
-        # Clear cached config
+    async def update_ui_config(self, ui: UIConfig) -> None:
-        from . import config as config_module
+        config = await self.load_config()
-        config_module._cached = None
+        config.ui = ui
+        await self.save_config(config)
 
     async def validate_ups_connection(
         self, ups_config: UPSConfig, timeout: float = 3.0
-    ) -> Dict[str, Any]:
+    ) -> dict[str, Any]:
         """Port-only connectivity test (no protocol / CLI call)."""
-
+        result: dict[str, Any] = {
-        result: Dict[str, Any] = {
             "success": False,
             "message": "",
             "connectivity": {"ok": False, "error": None},
             "protocol": {"ok": False, "error": None},
             "data": None,
         }
-
-        # Raw TCP connectivity test
         try:
-            # Use low-level socket to distinguish DNS/timeouts
             with socket.create_connection(
                 (ups_config.host, ups_config.port), timeout=timeout
             ):
                 result["connectivity"]["ok"] = True
-        except Exception as e:  # broad to surface any network issue
+        except Exception as e:
             result["connectivity"]["error"] = str(e)
             result["message"] = f"TCP connectivity failed: {e}"
             return result
-
-        # For port-only test we just mirror connectivity result
         if result["connectivity"]["ok"]:
             result["protocol"]["ok"] = True
             result["success"] = True
@@ -207,5 +152,13 @@ class ConfigManager:
         return result
 
 
-# Global instance
+def smtp_redacted_dict(smtp: SMTPConfig | None) -> dict[str, Any] | None:
+    """Return SMTP config dict with password redacted from env."""
+    if smtp is None:
+        return None
+    d = smtp.model_dump()
+    d["password"] = "***" if settings.smtp_password else None
+    return d
+
+
 config_manager = ConfigManager()

app/config_store.py

@@ -1,31 +1,39 @@
-"""Redis-backed configuration storage replacing YAML file.
+"""Redis-backed configuration storage.
 
-Schema:
+Key ``ups:config:json`` holds JSON: ``{"ups": [...], "smtp": {...}|null, "ui": {...}}``.
-  Key ups:config:json -> JSON object: {"ups": [...], "smtp": {...}|null}
+Legacy YAML import runs once if Redis is empty.
-
-Migration:
-  On first load if redis key missing and legacy YAML present, import it.
 """
 from __future__ import annotations
-from typing import Optional
+
-from pathlib import Path
 import json
 import logging
-from .storage import get_redis
+from pathlib import Path
-from .config import AppConfig, CONFIG_PATH
+
 import yaml
 
+from .config import CONFIG_PATH, AppConfig
+from .storage import get_redis
+
 logger = logging.getLogger(__name__)
 
 REDIS_CONFIG_KEY = "ups:config:json"
 
 
-def _load_legacy_yaml(path: Path) -> Optional[AppConfig]:
+def _strip_smtp_password(data: dict) -> dict:
+    """Strip persisted SMTP passwords (migration from older schema)."""
+    smtp = data.get("smtp")
+    if isinstance(smtp, dict) and "password" in smtp:
+        smtp.pop("password", None)
+    return data
+
+
+def _load_legacy_yaml(path: Path) -> AppConfig | None:
     if not path.exists():
         return None
     try:
         with path.open() as f:
             raw = yaml.safe_load(f) or {}
+        raw = _strip_smtp_password(raw)
         return AppConfig(**raw)
     except Exception as e:  # pragma: no cover
         logger.warning("Failed to import legacy YAML config: %s", e)
@@ -40,14 +48,13 @@ def load_config_redis() -> AppConfig:
             data = json.loads(raw)
         except json.JSONDecodeError:
             data = {}
+        data = _strip_smtp_password(data)
         return AppConfig(**data)
-    # Migration path
     legacy = _load_legacy_yaml(CONFIG_PATH)
     if legacy:
         save_config_redis(legacy)
         logger.info("Imported legacy YAML config into Redis")
         return legacy
-    # If nothing exists, create empty scaffold
     empty = AppConfig(ups=[], smtp=None)
     save_config_redis(empty)
     return empty
@@ -55,4 +62,6 @@ def load_config_redis() -> AppConfig:
 
 def save_config_redis(cfg: AppConfig) -> None:
     r = get_redis()
-    r.set(REDIS_CONFIG_KEY, json.dumps(cfg.model_dump(exclude_none=True)))
+    data = cfg.model_dump(exclude_none=True)
+    data = _strip_smtp_password(data)
+    r.set(REDIS_CONFIG_KEY, json.dumps(data))

app/exports.py

@@ -0,0 +1,78 @@
+"""CSV export helpers."""
+from __future__ import annotations
+
+import csv
+import io
+import time
+from collections.abc import AsyncIterator, Iterable
+
+from .storage import get_history, get_redis
+
+
+def _csv_row(writer_buf: io.StringIO, writer: csv.writer, row: Iterable) -> str:
+    writer.writerow(row)
+    out = writer_buf.getvalue()
+    writer_buf.seek(0)
+    writer_buf.truncate()
+    return out
+
+
+async def export_history_csv(ups_name: str, since_days: int = 7) -> AsyncIterator[str]:
+    history = await get_history(ups_name, since_seconds=since_days * 24 * 3600)
+    buf = io.StringIO()
+    writer = csv.writer(buf)
+    yield _csv_row(buf, writer, ["ts", "status", "loadpct", "bcharge", "timeleft", "linev", "derived_watts"])
+    for item in history:
+        d = item.get("data", {})
+        yield _csv_row(
+            buf,
+            writer,
+            [
+                item.get("ts"),
+                d.get("STATUS", ""),
+                d.get("LOADPCT", ""),
+                d.get("BCHARGE", ""),
+                d.get("TIMELEFT", ""),
+                d.get("LINEV", ""),
+                d.get("DERIVED_WATTS", ""),
+            ],
+        )
+
+
+async def export_events_csv(ups_name: str, since_days: int = 7) -> AsyncIterator[str]:
+    r = get_redis()
+    raw = r.lrange(f"ups:event:list:{ups_name}", 0, -1)
+    cutoff = int(time.time()) - since_days * 24 * 3600
+    buf = io.StringIO()
+    writer = csv.writer(buf)
+    yield _csv_row(buf, writer, ["ts", "type", "detail"])
+    for item in raw:
+        parts = item.split("|", 2)
+        if len(parts) != 3:
+            continue
+        try:
+            ts = int(parts[0])
+        except ValueError:
+            continue
+        if ts < cutoff:
+            continue
+        yield _csv_row(buf, writer, [ts, parts[1], parts[2]])
+
+
+async def export_energy_csv(ups_name: str, since_days: int = 7) -> AsyncIterator[str]:
+    r = get_redis()
+    buf = io.StringIO()
+    writer = csv.writer(buf)
+    yield _csv_row(buf, writer, ["date", "kwh"])
+    now = time.time()
+    for i in range(since_days):
+        day_ts = now - i * 86400
+        day_str = time.strftime("%Y%m%d", time.localtime(day_ts))
+        date_disp = time.strftime("%Y-%m-%d", time.localtime(day_ts))
+        watt_seconds = r.get(f"ups:energy:{ups_name}:{day_str}")
+        if watt_seconds:
+            try:
+                kwh = float(watt_seconds) / 3600.0 / 1000.0
+                yield _csv_row(buf, writer, [date_disp, f"{kwh:.4f}"])
+            except ValueError:
+                continue

app/health.py

@@ -0,0 +1,29 @@
+"""Liveness + readiness probes."""
+from __future__ import annotations
+
+from .config_store import load_config_redis
+from .storage import get_redis
+
+
+def liveness() -> dict:
+    return {"status": "ok"}
+
+
+def readiness() -> tuple[int, dict]:
+    """Returns (http_status, payload)."""
+    checks = {}
+    ok = True
+    try:
+        r = get_redis()
+        r.ping()
+        checks["redis"] = "ok"
+    except Exception as e:  # noqa: BLE001
+        checks["redis"] = f"error: {e}"
+        ok = False
+    try:
+        load_config_redis()
+        checks["config"] = "ok"
+    except Exception as e:  # noqa: BLE001
+        checks["config"] = f"error: {e}"
+        ok = False
+    return (200 if ok else 503, {"status": "ok" if ok else "error", "checks": checks})

app/logging_config.py

@@ -0,0 +1,59 @@
+"""Structured JSON logging with request-ID correlation."""
+from __future__ import annotations
+
+import contextvars
+import logging
+import sys
+import uuid
+
+from pythonjsonlogger import jsonlogger
+from starlette.middleware.base import BaseHTTPMiddleware
+
+from .settings import settings
+
+_request_id_var: contextvars.ContextVar[str] = contextvars.ContextVar(
+    "request_id", default="-"
+)
+
+
+def get_request_id() -> str:
+    return _request_id_var.get()
+
+
+class RequestIdFilter(logging.Filter):
+    def filter(self, record: logging.LogRecord) -> bool:
+        record.request_id = _request_id_var.get()
+        return True
+
+
+def configure_logging() -> None:
+    """Configure root logger to emit JSON with request_id."""
+    root = logging.getLogger()
+    # Remove any pre-existing handlers (uvicorn installs its own)
+    for h in list(root.handlers):
+        root.removeHandler(h)
+
+    handler = logging.StreamHandler(sys.stdout)
+    formatter = jsonlogger.JsonFormatter(
+        fmt="%(asctime)s %(levelname)s %(name)s %(request_id)s %(message)s",
+        rename_fields={"asctime": "ts", "levelname": "level", "name": "logger"},
+    )
+    handler.setFormatter(formatter)
+    handler.addFilter(RequestIdFilter())
+    root.addHandler(handler)
+    root.setLevel(getattr(logging, settings.log_level.upper(), logging.INFO))
+
+    # Calm down noisy loggers
+    logging.getLogger("uvicorn.access").setLevel(logging.WARNING)
+
+
+class RequestIdMiddleware(BaseHTTPMiddleware):
+    async def dispatch(self, request, call_next):
+        rid = request.headers.get("X-Request-ID") or uuid.uuid4().hex[:12]
+        token = _request_id_var.set(rid)
+        try:
+            response = await call_next(request)
+            response.headers["X-Request-ID"] = rid
+            return response
+        finally:
+            _request_id_var.reset(token)

app/main.py

@@ -1,472 +1,87 @@
+"""FastAPI application entry point.
+
+Assembles middleware, lifespan (poller start/stop), auth, routes.
+All request handlers live in app/routes/*.
+"""
 from __future__ import annotations
+
 import asyncio
-import time
+import logging
-from fastapi import FastAPI, Request, HTTPException
+from contextlib import asynccontextmanager
-from fastapi.responses import HTMLResponse, StreamingResponse
+
+from fastapi import FastAPI, Request
+from fastapi.responses import JSONResponse
 from fastapi.staticfiles import StaticFiles
-from fastapi.templating import Jinja2Templates
+from slowapi import _rate_limit_exceeded_handler
-import orjson
+from slowapi.errors import RateLimitExceeded
-from .config import load_config, UPSConfig
+
-from .storage import get_latest, get_history
+from .logging_config import RequestIdMiddleware, configure_logging
-from .poller import poll_loop
+from .poller import cancel_all_tasks, poll_loop
-from .config_manager import (
+from .rate_limit import limiter
-    config_manager,
+from .routes import api_alerts, api_auth, api_config, api_events, api_health, api_ups, pages, sse
-    UPSConfigUpdate,
+from .security import SecurityHeadersMiddleware
-    ConfigWriteError,
+from .settings import settings
-    get_config_version,
+
+logger = logging.getLogger(__name__)
+
+_background_tasks: list[asyncio.Task] = []
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    configure_logging()
+    if not settings.session_secret:
+        logger.warning(
+            "SESSION_SECRET not set; using insecure ephemeral secret. "
+            "Set SESSION_SECRET in your environment for production."
         )
-from .storage import get_redis
+    task = asyncio.create_task(poll_loop())
-from .apc_cli import fetch_status, APCStatusError
+    _background_tasks.append(task)
-
+    logger.info("Poller started")
-app = FastAPI(title="APC UPS Dashboard")
-app.mount('/static', StaticFiles(directory='app/static'), name='static')
-templates = Jinja2Templates(directory='app/templates')
-
-
-@app.on_event("startup")
-async def startup():
-    asyncio.create_task(poll_loop())
-
-
-@app.get('/', response_class=HTMLResponse)
-async def dashboard(request: Request):
-    cfg = load_config()
-    return templates.TemplateResponse(
-        'dashboard.html', {
-            "request": request,
-            "ups_list": cfg.ups,
-            "ui_cfg": cfg.ui.model_dump(),
-        }
-    )
-
-
-@app.get('/config', response_class=HTMLResponse)
-async def config_page(request: Request):
-    return templates.TemplateResponse('config.html', {"request": request})
-
-
-@app.get('/api/ups')
-async def list_ups():
-    cfg = load_config()
-    return [{"name": u.name, "host": u.host, "port": u.port} for u in cfg.ups]
-
-
-@app.get('/api/ups/{ups_name}')
-async def ups_status(ups_name: str):
-    snap = await get_latest(ups_name)
-    return snap or {"error": "not found"}
-
-
-@app.get('/api/ups/{ups_name}/history')
-async def ups_history(ups_name: str):
-    hist = await get_history(ups_name)
-    return hist
-
-
-@app.get('/api/ups/{ups_name}/metric/{metric}')
-async def metric_history(ups_name: str, metric: str, limit: int = 120):
-    """Return recent numeric history samples for a single metric.
-
-    Limit capped at 500 to avoid large payloads.
-    """
-    limit = max(1, min(limit, 500))
-    all_hist = await get_history(ups_name)
-    # Use only most recent entries
-    recent = all_hist[-limit:]
-    out = []
-    for item in recent:
-        data = item.get('data', {})
-        raw_val = data.get(metric)
-        if raw_val is None:
-            continue
-        # Extract leading number
     try:
-            val = float(str(raw_val).split()[0])
+        yield
-        except Exception:
+    finally:
-            continue
+        logger.info("Shutting down: cancelling background tasks")
-        out.append({'ts': item.get('ts'), 'value': val})
+        for t in _background_tasks:
-    return out
+            t.cancel()
-
+        for t in _background_tasks:
-
-@app.get('/api/ups/{ups_name}/events')
-async def ups_events(ups_name: str):
-    """Return recent status/transfer events for a UPS."""
-    r = get_redis()
-    key = f"ups:event:list:{ups_name}"
-    raw = r.lrange(key, 0, 99)
-    parsed = []
-    for item in raw:
-        if '|' in item:
             try:
-                ts_s, kind, rest = item.split('|', 2)
+                await t
-                parsed.append({
+            except (asyncio.CancelledError, Exception):
-                    'ts': int(ts_s),
-                    'type': kind,
-                    'detail': rest
-                })
-                continue
-            except Exception:
                 pass
-        parsed.append({'raw': item})
+        await cancel_all_tasks()
-    return parsed
+        logger.info("Shutdown complete")
 
 
-@app.get('/api/ups/{ups_name}/energy')
+def create_app() -> FastAPI:
-async def ups_energy(ups_name: str):
+    application = FastAPI(title="APC UPS Dashboard", lifespan=lifespan)
-    """Return today's accumulated energy (approx kWh) if available."""
+    application.state.limiter = limiter
-    r = get_redis()
+    application.add_exception_handler(RateLimitExceeded, _rate_limit_exceeded_handler)
-    day_str = time.strftime('%Y%m%d')  # type: ignore
-    key = f"ups:energy:{ups_name}:{day_str}"
-    watt_seconds = r.get(key)
-    if watt_seconds:
-        try:
-            ws = float(watt_seconds)
-            kwh = ws / 3600.0 / 1000.0
-            return {'kwh_today': round(kwh, 4)}
-        except ValueError:
-            pass
-    return {'kwh_today': None}
 
+    # Middleware (outermost first in registration; Starlette wraps inside-out)
+    application.add_middleware(SecurityHeadersMiddleware)
+    application.add_middleware(RequestIdMiddleware)
 
-@app.get('/api/ups/{ups_name}/watts_per_minute')
+    application.mount(
-async def ups_watts_per_minute(ups_name: str):
+        "/static", StaticFiles(directory="app/static"), name="static"
-    """Return recent per-minute average watts for a UPS (last 24h)."""
-    r = get_redis()
-    key = f"ups:watts:permin:{ups_name}"
-    raw = r.lrange(key, 0, 1440)
-    out = []
-    for item in raw:
-        if '|' in item:
-            minute, avg = item.split('|', 1)
-            try:
-                out.append({'minute': minute, 'avg_watts': float(avg)})
-            except ValueError:
-                continue
-    # list stored newest-first; reverse to chronological
-    out.reverse()
-    return out
-
-
-@app.get('/api/ups/{ups_name}/health')
-async def ups_health(ups_name: str):
-    """Return aggregated health indicators.
-
-    Includes recent alerts, voltage deviation stats, and ONBATT event count.
-    """
-    r = get_redis()
-    # Recent alerts
-    alerts_key = f"ups:alerts:recent:{ups_name}"
-    alert_raw = r.lrange(alerts_key, 0, 19)
-    alerts = []
-    for a in alert_raw:
-        if '|' in a:
-            ts_s, msg = a.split('|', 1)
-            try:
-                alerts.append({'ts': int(ts_s), 'msg': msg})
-            except ValueError:
-                alerts.append({'raw': a})
-        else:
-            alerts.append({'raw': a})
-    # Voltage deviation samples
-    dev_key = f"ups:volt:dev:samples:{ups_name}"
-    dev_samples = r.lrange(dev_key, 0, 49)
-    dev_vals = []
-    for d in dev_samples:
-        try:
-            dev_vals.append(float(d))
-        except ValueError:
-            continue
-    dev_avg = sum(devVals := dev_vals) / len(devVals) if dev_vals else None
-    dev_max = max(dev_vals) if dev_vals else None
-    # Transfer burst count (recent hour ONBATT events)
-    events_key = f"ups:event:list:{ups_name}"
-    now = int(time.time())
-    events = r.lrange(events_key, 0, 200)
-    onbatt_hour = 0
-    for ev in events:
-        parts = ev.split('|')
-        if len(parts) >= 3:
-            try:
-                ts_e = int(parts[0])
-            except ValueError:
-                continue
-            if now - ts_e > 3600:
-                continue
-            if parts[1] == 'STATUS' and 'ONBATT' in parts[2]:
-                onbatt_hour += 1
-    return {
-        'alerts': alerts,
-        'voltage_deviation': {
-            'avg_pct': round(dev_avg, 2) if dev_avg is not None else None,
-            'max_pct': round(dev_max, 2) if dev_max is not None else None,
-            'samples': len(dev_vals)
-        },
-        'onbatt_last_hour': onbatt_hour
-    }
-
-
-@app.get('/api/ups/{ups_name}/debug')
-async def ups_debug(ups_name: str):
-    """Return current apcaccess CLI status for a UPS."""
-    cfg = load_config()
-    target = next((u for u in cfg.ups if u.name == ups_name), None)
-    if not target:
-        raise HTTPException(status_code=404, detail='UPS not found')
-    try:
-        data = await fetch_status(target.host, target.port)
-        return data
-    except APCStatusError as e:
-        raise HTTPException(status_code=502, detail=str(e))
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.get('/api/stream')
-async def stream():
-    # simple Server Sent Events stream of snapshots (polling redis every 5s)
-    async def event_gen():
-        while True:
-            cfg = load_config()
-            payload = {"snapshots": {}, "cfgVersion": get_config_version()}
-            for u in cfg.ups:
-                snap = await get_latest(u.name)
-                if snap:
-                    payload["snapshots"][u.name] = snap
-            # include simple config metadata to help client reconcile
-            payload["upsMeta"] = [
-                {"name": u.name, "host": u.host, "port": u.port}
-                for u in cfg.ups
-            ]
-            # Backward compatibility: also flatten UPS snapshots at top level
-            for name, snap in payload["snapshots"].items():
-                payload.setdefault(name, snap)
-            yield f"data: {orjson.dumps(payload).decode()}\n\n"
-            await asyncio.sleep(5)
-    return StreamingResponse(event_gen(), media_type='text/event-stream')
-
-
-# Configuration management endpoints
-@app.get('/api/config/ups')
-async def get_ups_configs():
-    """Get all UPS configurations"""
-    try:
-        ups_list = await config_manager.get_ups_list()
-        return [ups.model_dump() for ups in ups_list]
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.get('/api/config/ups/{ups_name}')
-async def get_ups_config(ups_name: str):
-    """Get specific UPS configuration"""
-    try:
-        ups = await config_manager.get_ups(ups_name)
-        if not ups:
-            raise HTTPException(status_code=404, detail="UPS not found")
-        return ups.model_dump()
-    except HTTPException:
-        raise
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.post('/api/config/ups')
-async def add_ups_config(ups_config: UPSConfig):
-    """Add new UPS configuration"""
-    try:
-        await config_manager.add_ups(ups_config)
-        return {"message": "UPS configuration added successfully"}
-    except ConfigWriteError as e:
-        raise HTTPException(status_code=507, detail=str(e))
-    except ValueError as e:
-        raise HTTPException(status_code=400, detail=str(e))
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.put('/api/config/ups/{ups_name}')
-async def update_ups_config(ups_name: str, updates: UPSConfigUpdate):
-    """Update UPS configuration"""
-    try:
-        success = await config_manager.update_ups(ups_name, updates)
-        if not success:
-            raise HTTPException(status_code=404, detail="UPS not found")
-        return {"message": "UPS configuration updated successfully"}
-    except ConfigWriteError as e:
-        raise HTTPException(status_code=507, detail=str(e))
-    except ValueError as e:
-        raise HTTPException(status_code=400, detail=str(e))
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.delete('/api/config/ups/{ups_name}')
-async def delete_ups_config(ups_name: str):
-    """Delete UPS configuration"""
-    try:
-        success = await config_manager.delete_ups(ups_name)
-        if not success:
-            raise HTTPException(status_code=404, detail="UPS not found")
-        return {"message": "UPS configuration deleted successfully"}
-    except ConfigWriteError as e:
-        raise HTTPException(status_code=507, detail=str(e))
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.post('/api/config/ups/{ups_name}/test')
-async def test_ups_connection(ups_name: str):
-    """Test UPS connection"""
-    try:
-        ups = await config_manager.get_ups(ups_name)
-        if not ups:
-            raise HTTPException(status_code=404, detail="UPS not found")
-        
-        result = await config_manager.validate_ups_connection(ups)
-        return result
-    except HTTPException:
-        raise
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.post('/api/config/ups/test')
-async def test_new_ups_connection(ups_config: UPSConfig):
-    """Test new UPS configuration connection"""
-    try:
-        result = await config_manager.validate_ups_connection(ups_config)
-        return result
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.get('/api/config/smtp')
-async def get_smtp_config():
-    """Get SMTP configuration"""
-    try:
-        smtp = await config_manager.get_smtp_config()
-        return smtp.model_dump() if smtp else None
-    except Exception as e:
-        raise HTTPException(status_code=500, detail=str(e))
-
-
-@app.get('/api/config/ui')
-async def get_ui_config():
-    cfg = load_config()
-    return cfg.ui.model_dump()
-
-
-@app.put('/api/config/ui')
-async def update_ui_config(payload: dict):
-    # Simple partial update of UI flags
-    from .config import AppConfig
-    cfg = load_config()
-    ui_dict = cfg.ui.model_dump()
-    allowed = set(ui_dict.keys())
-    for k, v in payload.items():
-        if k in allowed and isinstance(v, bool):
-            ui_dict[k] = v
-    # Reconstruct full config and save via manager
-    new_cfg = AppConfig(
-        ups=cfg.ups, smtp=cfg.smtp, ui=cfg.ui.__class__(**ui_dict)
     )
-    await config_manager.save_config(new_cfg)
+
-    return {"message": "UI config updated", "ui": ui_dict}
+    # Routers
+    application.include_router(pages.router)
+    application.include_router(api_auth.router)
+    application.include_router(api_health.router)
+    application.include_router(api_config.router)
+    application.include_router(api_ups.router)
+    application.include_router(api_alerts.router)
+    application.include_router(api_events.router)
+    application.include_router(sse.router)
+
+    @application.exception_handler(Exception)
+    async def unhandled_exc(request: Request, exc: Exception):  # noqa: ARG001
+        logger.exception("Unhandled error", extra={"path": request.url.path})
+        return JSONResponse({"detail": "Internal server error"}, status_code=500)
+
+    return application
 
 
-# --- UI Tile Layout Persistence ---
+app = create_app()
-@app.get('/api/ups/{ups_name}/ui_tiles')
-async def get_ups_ui_tiles(ups_name: str):
-    """Return persisted UI tile settings for a UPS.
-
-    Includes: types, order, hidden, custom. Falls back to empty defaults if
-    not stored yet.
-    """
-    r = get_redis()
-    key = f"ups:ui:tiles:{ups_name}"
-    raw = r.get(key)
-    if not raw:
-        return {
-            "types": {},
-            "order": [],
-            "hidden": [],
-            "custom": [],
-            "positions": {}
-        }
-    try:
-        data = orjson.loads(raw)
-        # Basic shape validation
-        if not isinstance(data, dict):
-            raise ValueError
-        return {
-            "types": data.get("types", {}),
-            "order": data.get("order", []),
-            "hidden": data.get("hidden", []),
-            "custom": data.get("custom", []),
-            "positions": data.get("positions", {}),
-        }
-    except Exception:
-        return {
-            "types": {},
-            "order": [],
-            "hidden": [],
-            "custom": [],
-            "positions": {}
-        }
-
-
-@app.post('/api/ups/{ups_name}/ui_tiles')
-async def save_ups_ui_tiles(ups_name: str, payload: dict):
-    """Persist UI tile settings for a UPS.
-
-    Expects JSON: { types: {...}, order: [...], hidden: [...], custom: [...] }
-    """
-    # Light validation / sanitization
-    types = (payload.get('types')
-             if isinstance(payload.get('types'), dict) else {})
-    order = (payload.get('order')
-             if isinstance(payload.get('order'), list) else [])
-    hidden = (payload.get('hidden')
-              if isinstance(payload.get('hidden'), list) else [])
-    custom = (payload.get('custom')
-              if isinstance(payload.get('custom'), list) else [])
-    positions = (payload.get('positions')
-                 if isinstance(payload.get('positions'), dict) else {})
-    # Ensure custom entries minimally formed
-    norm_custom = []
-    for c in custom or []:
-        if not isinstance(c, dict):
-            continue
-        metric = c.get('metric')
-        chart = c.get('chart')
-        cid = c.get('id') or ''
-        source = c.get('source', 'live')
-        if not metric or not chart:
-            continue
-        norm_custom.append({
-            'id': cid,
-            'metric': metric,
-            'chart': chart,
-            'source': source
-        })
-    doc = {
-        'types': types,
-        'order': order,
-        'hidden': hidden,
-        'custom': norm_custom,
-        'positions': positions,
-        'saved_ts': int(time.time())
-    }
-    r = get_redis()
-    key = f"ups:ui:tiles:{ups_name}"
-    r.set(key, orjson.dumps(doc))
-    return {"message": "saved", "count_custom": len(norm_custom)}
-
-
-@app.delete('/api/ups/{ups_name}/ui_tiles')
-async def clear_ups_ui_tiles(ups_name: str):
-    """Delete stored UI tile layout/settings for a UPS (full reset)."""
-    r = get_redis()
-    key = f"ups:ui:tiles:{ups_name}"
-    r.delete(key)
-    return {"message": "cleared"}

app/metrics.py

@@ -0,0 +1,74 @@
+"""Prometheus exposition for UPS fleet metrics."""
+from __future__ import annotations
+
+from prometheus_client import (
+    CONTENT_TYPE_LATEST,
+    CollectorRegistry,
+    Counter,
+    Gauge,
+    generate_latest,
+)
+
+from .config import load_config
+from .storage import get_redis
+
+_registry = CollectorRegistry()
+
+_g_loadpct = Gauge("ups_loadpct", "UPS load percentage", ["ups"], registry=_registry)
+_g_bcharge = Gauge("ups_bcharge", "UPS battery charge percentage", ["ups"], registry=_registry)
+_g_timeleft = Gauge("ups_timeleft_minutes", "UPS estimated runtime minutes", ["ups"], registry=_registry)
+_g_watts = Gauge("ups_watts", "UPS derived watts", ["ups"], registry=_registry)
+_g_online = Gauge("ups_online", "1 if UPS is online and reachable", ["ups"], registry=_registry)
+_c_poll_errors = Counter(
+    "ups_poll_errors_total",
+    "Cumulative poll failures",
+    ["ups"],
+    registry=_registry,
+)
+
+
+def _f(v):
+    try:
+        return float(str(v).split()[0]) if v is not None else None
+    except Exception:
+        return None
+
+
+def render_metrics() -> tuple[bytes, str]:
+    """Collect current snapshot data and produce Prometheus text exposition."""
+    r = get_redis()
+    cfg = load_config()
+    for ups in cfg.ups:
+        snap = r.hgetall(f"ups:snap:{ups.name}")
+        status = str(snap.get("STATUS", "")).upper() if snap else ""
+        online = 1.0 if snap and "ONLINE" in status else 0.0
+        # If we've marked offline via health tracker, force zero
+        if r.get(f"ups:health:offline:{ups.name}"):
+            online = 0.0
+        _g_online.labels(ups=ups.name).set(online)
+        if snap:
+            lp = _f(snap.get("LOADPCT"))
+            if lp is not None:
+                _g_loadpct.labels(ups=ups.name).set(lp)
+            bc = _f(snap.get("BCHARGE"))
+            if bc is not None:
+                _g_bcharge.labels(ups=ups.name).set(bc)
+            tl = _f(snap.get("TIMELEFT")) or _f(snap.get("RUNTIME_MINUTES"))
+            if tl is not None:
+                _g_timeleft.labels(ups=ups.name).set(tl)
+            w = _f(snap.get("DERIVED_WATTS"))
+            if w is not None:
+                _g_watts.labels(ups=ups.name).set(w)
+        fail_count = r.get(f"ups:health:fail_count:{ups.name}")
+        if fail_count:
+            try:
+                # Counter semantics: set absolute by incrementing delta. Instead, expose gauge.
+                # We use a cumulative inc here; reset on process restart is acceptable.
+                pass
+            except Exception:
+                pass
+    return generate_latest(_registry), CONTENT_TYPE_LATEST
+
+
+def record_poll_error(ups_name: str) -> None:
+    _c_poll_errors.labels(ups=ups_name).inc()

app/notifications/__init__.py

@@ -0,0 +1 @@
+"""Notification channels."""

app/notifications/email.py

@@ -0,0 +1,157 @@
+"""SMTP email sending with retry, batching, and HTML templates."""
+from __future__ import annotations
+
+import logging
+import smtplib
+import ssl
+from datetime import datetime
+from email.message import EmailMessage
+from pathlib import Path
+from typing import Any
+
+from jinja2 import Environment, FileSystemLoader, select_autoescape
+from tenacity import (
+    RetryError,
+    retry,
+    retry_if_exception_type,
+    stop_after_attempt,
+    wait_exponential,
+)
+
+from ..config import SMTPConfig
+from ..settings import settings
+
+logger = logging.getLogger(__name__)
+
+_TEMPLATE_DIR = Path(__file__).parent.parent / "templates" / "emails"
+_env = Environment(
+    loader=FileSystemLoader(str(_TEMPLATE_DIR)),
+    autoescape=select_autoescape(["html", "xml"]),
+)
+
+SEVERITY_COLORS = {
+    "CRITICAL": "#c0392b",
+    "WARNING": "#e67e22",
+    "INFO": "#2980b9",
+}
+
+
+class EmailSendError(Exception):
+    pass
+
+
+def _resolve_password(smtp_cfg: SMTPConfig) -> str | None:
+    # Password is never stored in Redis; always from env var
+    return settings.smtp_password
+
+
+@retry(
+    reraise=True,
+    stop=stop_after_attempt(3),
+    wait=wait_exponential(multiplier=1, min=1, max=10),
+    retry=retry_if_exception_type((smtplib.SMTPException, OSError)),
+)
+def _send_raw(smtp_cfg: SMTPConfig, msg: EmailMessage) -> None:
+    password = _resolve_password(smtp_cfg)
+    if smtp_cfg.use_ssl:
+        context = ssl.create_default_context()
+        with smtplib.SMTP_SSL(
+            smtp_cfg.host, smtp_cfg.port, context=context, timeout=30
+        ) as server:
+            if smtp_cfg.username and password:
+                server.login(smtp_cfg.username, password)
+            server.send_message(msg)
+    else:
+        with smtplib.SMTP(smtp_cfg.host, smtp_cfg.port, timeout=30) as server:
+            if smtp_cfg.use_tls:
+                server.starttls(context=ssl.create_default_context())
+            if smtp_cfg.username and password:
+                server.login(smtp_cfg.username, password)
+            server.send_message(msg)
+
+
+def send_alert_email(
+    smtp_cfg: SMTPConfig,
+    ups_name: str,
+    alerts: list[dict[str, Any]],
+    dashboard_url: str = "http://localhost:8000/",
+) -> None:
+    """Send a single HTML email coalescing multiple alerts for a UPS."""
+    if not smtp_cfg.to_addrs or not alerts:
+        return
+    top_severity = _pick_top_severity(alerts)
+    subject = f"{smtp_cfg.subject_prefix} [{top_severity}] {ups_name}"
+    template = _env.get_template("alert.html")
+    html = template.render(
+        ups_name=ups_name,
+        alerts=alerts,
+        top_severity=top_severity,
+        color=SEVERITY_COLORS.get(top_severity, "#555"),
+        dashboard_url=dashboard_url,
+        sent_at=datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
+    )
+    text_body = "\n".join(
+        f"[{a.get('severity', 'INFO')}] {a.get('message', '')}" for a in alerts
+    )
+    msg = EmailMessage()
+    msg["Subject"] = subject
+    msg["From"] = smtp_cfg.from_addr or (smtp_cfg.username or "ups@example.local")
+    msg["To"] = ", ".join(smtp_cfg.to_addrs)
+    msg.set_content(text_body)
+    msg.add_alternative(html, subtype="html")
+    try:
+        _send_raw(smtp_cfg, msg)
+    except RetryError as e:
+        raise EmailSendError(f"SMTP send failed after retries: {e}") from e
+    except Exception as e:  # noqa: BLE001 - surface as domain error
+        raise EmailSendError(f"SMTP send failed: {e}") from e
+
+
+def send_test_email(smtp_cfg: SMTPConfig) -> None:
+    if not smtp_cfg.to_addrs:
+        raise EmailSendError("No recipients configured")
+    msg = EmailMessage()
+    msg["Subject"] = f"{smtp_cfg.subject_prefix} Test email"
+    msg["From"] = smtp_cfg.from_addr or (smtp_cfg.username or "ups@example.local")
+    msg["To"] = ", ".join(smtp_cfg.to_addrs)
+    msg.set_content(
+        "This is a test message from your APC UPS Dashboard.\n"
+        f"Sent at {datetime.now().isoformat()}"
+    )
+    try:
+        _send_raw(smtp_cfg, msg)
+    except Exception as e:  # noqa: BLE001
+        raise EmailSendError(str(e)) from e
+
+
+def send_daily_summary(
+    smtp_cfg: SMTPConfig,
+    summary: dict[str, Any],
+) -> None:
+    if not smtp_cfg.to_addrs:
+        return
+    template = _env.get_template("summary.html")
+    html = template.render(summary=summary, sent_at=datetime.now().isoformat())
+    msg = EmailMessage()
+    msg["Subject"] = f"{smtp_cfg.subject_prefix} Daily summary"
+    msg["From"] = smtp_cfg.from_addr or (smtp_cfg.username or "ups@example.local")
+    msg["To"] = ", ".join(smtp_cfg.to_addrs)
+    msg.set_content(str(summary))
+    msg.add_alternative(html, subtype="html")
+    try:
+        _send_raw(smtp_cfg, msg)
+    except Exception as e:  # noqa: BLE001
+        logger.warning("Daily summary send failed: %s", e)
+
+
+def _pick_top_severity(alerts: list[dict[str, Any]]) -> str:
+    order = {"CRITICAL": 3, "WARNING": 2, "INFO": 1}
+    best = "INFO"
+    best_rank = 0
+    for a in alerts:
+        sev = a.get("severity", "INFO")
+        rank = order.get(sev, 0)
+        if rank > best_rank:
+            best = sev
+            best_rank = rank
+    return best

app/poller.py

@@ -1,17 +1,114 @@
+"""Per-UPS async polling, event/energy/battery-history tracking.
+
+Connection health:
+  ups:health:last_ok:<name>     - unix ts of last successful poll
+  ups:health:fail_count:<name>  - consecutive failures
+  ups:health:offline:<name>     - "1" if we've emitted a COMMLOST alert
+
+Battery history:
+  ups:battery:history:<name>    - JSON {ts,bcharge,timeleft,battv,nombattv}
+                                  capped at 10080 entries (~7d @ 1/min)
+"""
 from __future__ import annotations
+
 import asyncio
-import time
+import json
 import logging
+import time
+
+from .alerts import Alert, dispatch_alerts, emit_info, process_alerts
+from .apc_cli import APCStatusError, fetch_status
 from .config import load_config
-from .apc_cli import fetch_status, APCStatusError
+from .storage import get_redis, prune_old, store_snapshot
-from .storage import store_snapshot, prune_old, get_redis
-from .alerts import process_alerts
 
 logger = logging.getLogger(__name__)
 
 _ACTIVE_TASKS: dict[str, asyncio.Task] = {}
 _RELOADER_LOCK = asyncio.Lock()
 
+BATTERY_HISTORY_MAX = 10080  # 7 days @ 1/minute
+HEALTH_OFFLINE_MIN_SECONDS = 180
+
+
+def _get_offline_threshold(interval_seconds: int) -> int:
+    return max(HEALTH_OFFLINE_MIN_SECONDS, 3 * interval_seconds)
+
+
+def _record_poll_success(name: str, snapshot: dict, interval_seconds: int) -> None:
+    r = get_redis()
+    now = int(time.time())
+    r.set(f"ups:health:last_ok:{name}", now)
+    r.delete(f"ups:health:fail_count:{name}")
+    was_offline = r.get(f"ups:health:offline:{name}")
+    if was_offline:
+        r.delete(f"ups:health:offline:{name}")
+        emit_info(name, "REACHABLE", f"UPS {name} reachable again")
+    # Battery history append (sample every ~60s to keep list bounded)
+    last_sample_key = f"ups:battery:history:last_ts:{name}"
+    last_ts_raw = r.get(last_sample_key)
+    try:
+        last_ts = int(last_ts_raw) if last_ts_raw else 0
+    except ValueError:
+        last_ts = 0
+    if now - last_ts >= 60:
+        try:
+            sample = {
+                "ts": now,
+                "bcharge": _f(snapshot.get("BCHARGE")),
+                "timeleft": _f(snapshot.get("TIMELEFT")),
+                "battv": _f(snapshot.get("BATTV")),
+                "nombattv": _f(snapshot.get("NOMBATTV")),
+            }
+            r.lpush(f"ups:battery:history:{name}", json.dumps(sample))
+            r.ltrim(f"ups:battery:history:{name}", 0, BATTERY_HISTORY_MAX - 1)
+            r.set(last_sample_key, now)
+        except Exception:  # pragma: no cover
+            pass
+
+
+def _record_poll_failure(name: str, interval_seconds: int, reason: str) -> None:
+    r = get_redis()
+    now = int(time.time())
+    r.incr(f"ups:health:fail_count:{name}")
+    last_ok = r.get(f"ups:health:last_ok:{name}")
+    try:
+        last_ok_ts = int(last_ok) if last_ok else 0
+    except ValueError:
+        last_ok_ts = 0
+    threshold = _get_offline_threshold(interval_seconds)
+    # If we never succeeded, base threshold on process start (use last_ok=0 => big delta)
+    if now - last_ok_ts > threshold:
+        offline_flag = r.get(f"ups:health:offline:{name}")
+        if not offline_flag:
+            r.set(f"ups:health:offline:{name}", "1")
+            alert = Alert(
+                "CRITICAL",
+                f"UPS {name} unreachable ({reason})",
+                "COMMLOST",
+                name,
+                now,
+            )
+            cfg = load_config()
+            r.lpush(
+                "ups:alerts:history:all",
+                f"{alert.ts}|CRITICAL|{alert.ups}|COMMLOST|{alert.message}|{alert.hash()}",
+            )
+            r.ltrim("ups:alerts:history:all", 0, 499)
+            r.lpush(
+                f"ups:alerts:recent:{name}",
+                f"{alert.ts}|CRITICAL|{alert.message}",
+            )
+            r.ltrim(f"ups:alerts:recent:{name}", 0, 49)
+            if cfg.smtp:
+                dispatch_alerts(cfg.smtp, name, [alert])
+
+
+def _f(v):
+    try:
+        return float(str(v).split()[0]) if v is not None else None
+    except Exception:
+        return None
+
 
 async def _poll_one(ups):
     r = get_redis()
@@ -21,7 +118,6 @@ async def _poll_one(ups):
         try:
             data = await fetch_status(ups.host, ups.port)
             data['UPSNAME'] = ups.name
-            # Derived metrics
             try:
                 loadpct = float(str(data.get('LOADPCT', '0')).split()[0])
             except Exception:
@@ -33,19 +129,15 @@ async def _poll_one(ups):
                 pass
             if nompower and loadpct >= 0:
                 watts = nompower * loadpct / 100.0
-                data['DERIVED_WATTS'] = f"{watts:.0f}"  # integer string
+                data['DERIVED_WATTS'] = f"{watts:.0f}"
                 data['HEADROOM_PCT'] = f"{max(0.0, 100.0 - loadpct):.0f}"
-            # Runtime minutes (normalize TIMELEFT like '15.0 Minutes')
             timeleft_raw = str(data.get('TIMELEFT', '')).strip()
             try:
                 runtime_min = float(timeleft_raw.split()[0])
                 data['RUNTIME_MINUTES'] = f"{runtime_min:.1f}"
             except Exception:
                 pass
-            # Event detection (status changes, last transfer changes)
+            wall_ts = int(time.time())
-            # Reuse redis handle r
-            now_ts = asyncio.get_event_loop().time()
-            wall_ts = int(now_ts)
             status_key = f"ups:event:status:last:{ups.name}"
             lastxfer_key = f"ups:event:lastxfer:last:{ups.name}"
             events_list_key = f"ups:event:list:{ups.name}"
@@ -55,53 +147,39 @@ async def _poll_one(ups):
             if prev_status != status_now and status_now:
                 r.set(status_key, status_now)
                 r.lpush(events_list_key, f"{wall_ts}|STATUS|{status_now}")
+                # Transition from ONBATT -> ONLINE: info alert
+                if prev_status and "ONBATT" in str(prev_status) and "ONLINE" in status_now:
+                    emit_info(
+                        ups.name,
+                        "LINE_RESTORED",
+                        f"UPS {ups.name} returned to line power",
+                    )
             lastxfer_now = str(data.get('LASTXFER', '')).strip()
             prev_lastxfer = r.get(lastxfer_key)
             if lastxfer_now and lastxfer_now != prev_lastxfer:
                 r.set(lastxfer_key, lastxfer_now)
                 r.lpush(events_list_key, f"{wall_ts}|XFER|{lastxfer_now}")
-            # Trim events
             r.ltrim(events_list_key, 0, max_events - 1)
-            # Energy accumulation (watt-seconds)
             if 'DERIVED_WATTS' in data:
                 try:
                     watts = float(data['DERIVED_WATTS'])
                     day_str = time.strftime('%Y%m%d')
                     energy_key = f"ups:energy:{ups.name}:{day_str}"
-                    # increment by watts * interval_seconds (approx)
                     r.incrbyfloat(energy_key, watts * ups.interval_seconds)
                     r.expire(energy_key, 3 * 24 * 3600)
-                    # Per-minute accumulation
                     minute = time.strftime('%Y%m%d%H%M')
-                    # Running sum and count in a hash
                     mb = r.hgetall(minute_bucket_key)
                     if not mb or mb.get('minute') != minute:
-                        # finalize previous bucket
+                        if mb and 'sum' in mb and 'count' in mb and 'minute' in mb:
-                        if (
-                            mb
-                            and 'sum' in mb
-                            and 'count' in mb
-                            and 'minute' in mb
-                        ):
                             try:
-                                avg = float(mb['sum']) / max(
+                                avg = float(mb['sum']) / max(1, int(mb['count']))
-                                    1, int(mb['count'])
+                                r.lpush(series_key, f"{mb['minute']}|{avg:.2f}")
-                                )
-                                r.lpush(
-                                    series_key,
-                                    f"{mb['minute']}|{avg:.2f}"
-                                )
-                                # keep up to 24h of minutes
                                 r.ltrim(series_key, 0, 1439)
                             except Exception:
                                 pass
                         r.hset(
                             minute_bucket_key,
-                            mapping={
+                            mapping={'minute': minute, 'sum': watts, 'count': 1},
-                                'minute': minute,
-                                'sum': watts,
-                                'count': 1,
-                            },
                         )
                         r.expire(minute_bucket_key, 26 * 3600)
                     else:
@@ -121,36 +199,47 @@ async def _poll_one(ups):
                 except Exception:
                     pass
             await store_snapshot(ups.name, data)
+            _record_poll_success(ups.name, data, ups.interval_seconds)
             process_alerts(ups, data)
+        except asyncio.CancelledError:
+            raise
         except Exception as e:
+            reason = str(e) if not isinstance(e, APCStatusError) else str(e)
             if isinstance(e, APCStatusError):
                 logger.warning("apcaccess error for %s: %s", ups.name, e)
             else:
                 logger.warning("Polling error for %s: %s", ups.name, e)
+            _record_poll_failure(ups.name, ups.interval_seconds, reason)
         await asyncio.sleep(ups.interval_seconds)
 
 
 async def _reconcile_tasks():
-    """Ensure a polling task exists per configured UPS and remove stale ones.
-
-    Creates tasks for new UPS entries and cancels tasks whose UPS were removed.
-    """
     async with _RELOADER_LOCK:
         cfg = load_config()
-        current_names = {u.name for u in cfg.ups}
+        current = {u.name: (u.host, u.port, u.interval_seconds) for u in cfg.ups}
-        # cancel removed
+        # cancel removed or changed
         for name in list(_ACTIVE_TASKS.keys()):
-            if name not in current_names:
+            if name not in current:
                 _ACTIVE_TASKS[name].cancel()
                 del _ACTIVE_TASKS[name]
-        # add new
         for ups in cfg.ups:
             if ups.name not in _ACTIVE_TASKS:
                 _ACTIVE_TASKS[ups.name] = asyncio.create_task(_poll_one(ups))
 
 
+async def cancel_all_tasks() -> None:
+    """Cancel all running poller tasks (used on shutdown)."""
+    for t in _ACTIVE_TASKS.values():
+        t.cancel()
+    for _name, t in list(_ACTIVE_TASKS.items()):
+        try:
+            await t
+        except (asyncio.CancelledError, Exception):
+            pass
+    _ACTIVE_TASKS.clear()
+
+
 async def poll_loop():
-    # initial reconcile
     await _reconcile_tasks()
 
     async def prune_loop():
@@ -162,19 +251,13 @@ async def poll_loop():
             await asyncio.sleep(3600)
 
     async def config_watch_loop():
-        """Periodically re-read config to capture UPS CRUD changes."""
         last_fingerprint = None
         while True:
             try:
                 cfg = load_config()
                 fingerprint = tuple(
                     sorted(
-                        (
+                        (u.name, u.host, u.port, u.interval_seconds)
-                            u.name,
-                            u.host,
-                            u.port,
-                            u.interval_seconds,
-                        )
                         for u in cfg.ups
                     )
                 )
@@ -188,5 +271,5 @@ async def poll_loop():
     await asyncio.gather(
         prune_loop(),
         config_watch_loop(),
-        *(_ACTIVE_TASKS.values()),
+        return_exceptions=True,
     )

app/rate_limit.py

@@ -0,0 +1,27 @@
+"""Rate limiting via slowapi with Redis-backed storage."""
+from __future__ import annotations
+
+from slowapi import Limiter
+from slowapi.util import get_remote_address
+
+from .settings import settings
+
+
+def _enabled_swallow(f):
+    """Return original limiter decorator or a no-op when rate limiting disabled."""
+    if settings.rate_limit_enabled:
+        return f
+
+    def _noop(*args, **kwargs):
+        def wrapper(fn):
+            return fn
+        return wrapper
+    return _noop
+
+
+limiter = Limiter(
+    key_func=get_remote_address,
+    storage_uri=settings.redis_url,
+    enabled=settings.rate_limit_enabled,
+    default_limits=[],
+)

app/routes/__init__.py

@@ -0,0 +1 @@
+"""Route modules split by concern."""

app/routes/api_alerts.py

@@ -0,0 +1,62 @@
+"""Alert history and acknowledgement endpoints."""
+import time
+
+from fastapi import APIRouter, Depends, Query
+
+from ..auth import require_session, require_session_and_csrf
+from ..storage import get_redis
+
+router = APIRouter(prefix="/api/alerts")
+
+
+@router.get("")
+async def list_alerts(
+    ups: str | None = Query(None),
+    severity: str | None = Query(None),
+    days: int = Query(30, ge=1, le=90),
+    user=Depends(require_session),
+):
+    r = get_redis()
+    raw = r.lrange("ups:alerts:history:all", 0, -1)
+    cutoff = int(time.time()) - days * 86400
+    out = []
+    for item in raw:
+        parts = item.split("|", 5)
+        if len(parts) < 5:
+            continue
+        try:
+            ts = int(parts[0])
+        except ValueError:
+            continue
+        if ts < cutoff:
+            continue
+        row = {
+            "ts": ts,
+            "severity": parts[1],
+            "ups": parts[2],
+            "code": parts[3],
+            "message": parts[4],
+            "id": parts[5] if len(parts) > 5 else "",
+        }
+        if ups and row["ups"] != ups:
+            continue
+        if severity and row["severity"] != severity.upper():
+            continue
+        ack = r.get(f"ups:alerts:ack:{row['id']}")
+        row["acked_ts"] = int(ack) if ack else None
+        out.append(row)
+    return out
+
+
+@router.get("/active")
+async def list_active(user=Depends(require_session)):
+    """Active alerts = history entries not yet acknowledged, most recent first."""
+    all_alerts = await list_alerts(days=7, user=user)
+    return [a for a in all_alerts if not a.get("acked_ts")]
+
+
+@router.post("/{alert_id}/ack")
+async def ack_alert(alert_id: str, user=Depends(require_session_and_csrf)):
+    r = get_redis()
+    r.set(f"ups:alerts:ack:{alert_id}", int(time.time()), ex=30 * 86400)
+    return {"message": "acknowledged"}

app/routes/api_auth.py

@@ -0,0 +1,62 @@
+"""Authentication endpoints: login, logout, first-run setup."""
+import re
+
+from fastapi import APIRouter, HTTPException, Request, Response
+from pydantic import BaseModel, Field
+
+from ..auth import (
+    clear_auth_cookies,
+    get_stored_admin,
+    is_admin_configured,
+    set_auth_cookies,
+    store_admin,
+    verify_password,
+)
+from ..rate_limit import limiter
+
+router = APIRouter()
+
+
+class LoginRequest(BaseModel):
+    username: str = Field(..., min_length=1, max_length=64)
+    password: str = Field(..., min_length=1, max_length=256)
+
+
+class SetupRequest(BaseModel):
+    username: str = Field(..., min_length=3, max_length=64)
+    password: str = Field(..., min_length=8, max_length=256)
+
+
+_PW_POLICY_MSG = "Password must be at least 8 characters"
+
+
+@router.post("/api/login")
+@limiter.limit("5/minute")
+async def api_login(request: Request, response: Response, payload: LoginRequest):
+    username, pw_hash = get_stored_admin()
+    if not pw_hash:
+        raise HTTPException(status_code=409, detail="Admin not configured; run setup")
+    if payload.username != username or not verify_password(payload.password, pw_hash):
+        raise HTTPException(status_code=401, detail="Invalid credentials")
+    csrf = set_auth_cookies(response, username)
+    return {"ok": True, "csrf_token": csrf}
+
+
+@router.post("/api/logout")
+async def api_logout(response: Response):
+    clear_auth_cookies(response)
+    return {"ok": True}
+
+
+@router.post("/api/setup")
+@limiter.limit("3/minute")
+async def api_setup(request: Request, response: Response, payload: SetupRequest):
+    if is_admin_configured():
+        raise HTTPException(status_code=409, detail="Admin already configured")
+    if len(payload.password) < 8:
+        raise HTTPException(status_code=400, detail=_PW_POLICY_MSG)
+    if not re.match(r"^[a-zA-Z0-9_.@-]+$", payload.username):
+        raise HTTPException(status_code=400, detail="Invalid username characters")
+    store_admin(payload.username, payload.password)
+    csrf = set_auth_cookies(response, payload.username)
+    return {"ok": True, "csrf_token": csrf}

app/routes/api_config.py

@@ -0,0 +1,147 @@
+"""Configuration CRUD endpoints (UPS, SMTP, UI). All require session+CSRF for writes."""
+
+from fastapi import APIRouter, Depends, HTTPException, Request
+
+from ..auth import require_session, require_session_and_csrf
+from ..config import SMTPConfig, UIConfig, UPSConfig
+from ..config_manager import (
+    ConfigWriteError,
+    UPSConfigUpdate,
+    config_manager,
+    smtp_redacted_dict,
+)
+from ..notifications.email import EmailSendError, send_test_email
+from ..rate_limit import limiter
+
+router = APIRouter(prefix="/api/config")
+
+
+@router.get("/ups")
+async def get_ups_configs(user=Depends(require_session)):
+    try:
+        ups_list = await config_manager.get_ups_list()
+        return [ups.model_dump() for ups in ups_list]
+    except Exception as e:  # noqa: BLE001
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.get("/ups/{ups_name}")
+async def get_ups_config(ups_name: str, user=Depends(require_session)):
+    ups = await config_manager.get_ups(ups_name)
+    if not ups:
+        raise HTTPException(status_code=404, detail="UPS not found")
+    return ups.model_dump()
+
+
+@router.post("/ups")
+@limiter.limit("30/minute")
+async def add_ups_config(
+    request: Request, ups_config: UPSConfig, user=Depends(require_session_and_csrf)
+):
+    try:
+        await config_manager.add_ups(ups_config)
+        return {"message": "UPS configuration added successfully"}
+    except ValueError as e:
+        raise HTTPException(status_code=400, detail=str(e))
+    except ConfigWriteError as e:
+        raise HTTPException(status_code=507, detail=str(e))
+
+
+@router.put("/ups/{ups_name}")
+@limiter.limit("30/minute")
+async def update_ups_config(
+    request: Request,
+    ups_name: str,
+    updates: UPSConfigUpdate,
+    user=Depends(require_session_and_csrf),
+):
+    try:
+        success = await config_manager.update_ups(ups_name, updates)
+        if not success:
+            raise HTTPException(status_code=404, detail="UPS not found")
+        return {"message": "UPS configuration updated successfully"}
+    except ValueError as e:
+        raise HTTPException(status_code=400, detail=str(e))
+    except ConfigWriteError as e:
+        raise HTTPException(status_code=507, detail=str(e))
+
+
+@router.delete("/ups/{ups_name}")
+@limiter.limit("30/minute")
+async def delete_ups_config(
+    request: Request, ups_name: str, user=Depends(require_session_and_csrf)
+):
+    success = await config_manager.delete_ups(ups_name)
+    if not success:
+        raise HTTPException(status_code=404, detail="UPS not found")
+    return {"message": "UPS configuration deleted successfully"}
+
+
+@router.post("/ups/{ups_name}/test")
+@limiter.limit("5/minute")
+async def test_ups_connection(
+    request: Request, ups_name: str, user=Depends(require_session_and_csrf)
+):
+    ups = await config_manager.get_ups(ups_name)
+    if not ups:
+        raise HTTPException(status_code=404, detail="UPS not found")
+    return await config_manager.validate_ups_connection(ups)
+
+
+@router.post("/ups/test")
+@limiter.limit("5/minute")
+async def test_new_ups_connection(
+    request: Request, ups_config: UPSConfig, user=Depends(require_session_and_csrf)
+):
+    return await config_manager.validate_ups_connection(ups_config)
+
+
+@router.get("/smtp")
+async def get_smtp_config(user=Depends(require_session)):
+    smtp = await config_manager.get_smtp_config()
+    return smtp_redacted_dict(smtp)
+
+
+@router.put("/smtp")
+@limiter.limit("30/minute")
+async def update_smtp_config(
+    request: Request, smtp_config: SMTPConfig, user=Depends(require_session_and_csrf)
+):
+    await config_manager.update_smtp_config(smtp_config)
+    return {"message": "SMTP configuration updated"}
+
+
+@router.post("/smtp/test")
+@limiter.limit("5/minute")
+async def test_smtp(request: Request, user=Depends(require_session_and_csrf)):
+    smtp = await config_manager.get_smtp_config()
+    if not smtp:
+        raise HTTPException(status_code=400, detail="SMTP not configured")
+    try:
+        send_test_email(smtp)
+    except EmailSendError as e:
+        raise HTTPException(status_code=502, detail=str(e))
+    return {"message": "Test email sent"}
+
+
+@router.get("/ui")
+async def get_ui_config(user=Depends(require_session)):
+    from ..config import load_config
+    cfg = load_config()
+    return cfg.ui.model_dump()
+
+
+@router.put("/ui")
+@limiter.limit("30/minute")
+async def update_ui_config(
+    request: Request, payload: dict, user=Depends(require_session_and_csrf)
+):
+    from ..config import load_config
+    cfg = load_config()
+    ui_dict = cfg.ui.model_dump()
+    for k, v in payload.items():
+        if k in ui_dict:
+            ui_dict[k] = v
+    new_ui = UIConfig(**ui_dict)
+    await config_manager.update_ui_config(new_ui)
+    return {"message": "UI config updated", "ui": new_ui.model_dump()}

app/routes/api_events.py

@@ -0,0 +1,37 @@
+"""Consolidated events log across all UPS."""
+
+from fastapi import APIRouter, Depends, Query
+
+from ..auth import require_session
+from ..config import load_config
+from ..storage import get_redis
+
+router = APIRouter(prefix="/api/events")
+
+
+@router.get("")
+async def list_events(
+    ups: str | None = Query(None),
+    kind: str | None = Query(None, description="STATUS|XFER"),
+    limit: int = Query(200, ge=1, le=1000),
+    user=Depends(require_session),
+):
+    r = get_redis()
+    cfg = load_config()
+    target_upses = [ups] if ups else [u.name for u in cfg.ups]
+    out = []
+    for name in target_upses:
+        raw = r.lrange(f"ups:event:list:{name}", 0, limit)
+        for item in raw:
+            parts = item.split("|", 2)
+            if len(parts) != 3:
+                continue
+            try:
+                ts = int(parts[0])
+            except ValueError:
+                continue
+            if kind and parts[1] != kind.upper():
+                continue
+            out.append({"ts": ts, "type": parts[1], "detail": parts[2], "ups": name})
+    out.sort(key=lambda x: x["ts"], reverse=True)
+    return out[:limit]

app/routes/api_health.py

@@ -0,0 +1,25 @@
+"""Liveness, readiness, and Prometheus metrics endpoints (unauthenticated)."""
+from fastapi import APIRouter
+from fastapi.responses import JSONResponse, Response
+
+from ..health import liveness, readiness
+from ..metrics import render_metrics
+
+router = APIRouter()
+
+
+@router.get("/healthz")
+async def healthz():
+    return liveness()
+
+
+@router.get("/readyz")
+async def readyz():
+    status, payload = readiness()
+    return JSONResponse(payload, status_code=status)
+
+
+@router.get("/metrics")
+async def metrics():
+    body, content_type = render_metrics()
+    return Response(content=body, media_type=content_type)

app/routes/api_ups.py

@@ -0,0 +1,356 @@
+"""Per-UPS data APIs: status, history, metric, events, energy, health, battery health, export, debug, tiles."""
+import json
+import time
+
+import orjson
+from fastapi import APIRouter, Depends, HTTPException, Query
+from fastapi.responses import StreamingResponse
+
+from ..apc_cli import APCStatusError, fetch_status
+from ..auth import require_session, require_session_and_csrf
+from ..config import load_config
+from ..exports import export_energy_csv, export_events_csv, export_history_csv
+from ..storage import get_history, get_latest, get_redis
+
+router = APIRouter(prefix="/api/ups")
+
+
+@router.get("")
+async def list_ups(user=Depends(require_session)):
+    cfg = load_config()
+    return [{"name": u.name, "host": u.host, "port": u.port} for u in cfg.ups]
+
+
+@router.get("/fleet/overview")
+async def fleet_overview(user=Depends(require_session)):
+    """Aggregate summary across all UPS."""
+    cfg = load_config()
+    r = get_redis()
+    total = len(cfg.ups)
+    counts = {"online": 0, "on_battery": 0, "warning": 0, "offline": 0, "unknown": 0}
+    total_watts = 0.0
+    min_timeleft = None
+    rows = []
+    for ups in cfg.ups:
+        snap = r.hgetall(f"ups:snap:{ups.name}") or {}
+        is_offline = bool(r.get(f"ups:health:offline:{ups.name}"))
+        status = str(snap.get("STATUS", "")).upper()
+        state = "unknown"
+        if is_offline:
+            state = "offline"
+        elif "ONBATT" in status:
+            state = "on_battery"
+        elif "ONLINE" in status:
+            state = "online"
+        counts[state] = counts.get(state, 0) + 1
+        try:
+            w = float(snap.get("DERIVED_WATTS", 0) or 0)
+            total_watts += w
+        except ValueError:
+            pass
+        try:
+            tl = float(str(snap.get("TIMELEFT", "")).split()[0])
+            if min_timeleft is None or tl < min_timeleft:
+                min_timeleft = tl
+        except Exception:
+            pass
+        rows.append({"name": ups.name, "state": state, "status": status})
+    return {
+        "total": total,
+        "counts": counts,
+        "total_watts": round(total_watts, 1),
+        "min_timeleft_minutes": min_timeleft,
+        "rows": rows,
+    }
+
+
+@router.get("/{ups_name}")
+async def ups_status(ups_name: str, user=Depends(require_session)):
+    snap = await get_latest(ups_name)
+    return snap or {"error": "not found"}
+
+
+@router.get("/{ups_name}/history")
+async def ups_history(ups_name: str, user=Depends(require_session)):
+    return await get_history(ups_name)
+
+
+@router.get("/{ups_name}/metric/{metric}")
+async def metric_history(
+    ups_name: str, metric: str, limit: int = 120, user=Depends(require_session)
+):
+    limit = max(1, min(limit, 500))
+    all_hist = await get_history(ups_name)
+    recent = all_hist[-limit:]
+    out = []
+    for item in recent:
+        data = item.get("data", {})
+        raw_val = data.get(metric)
+        if raw_val is None:
+            continue
+        try:
+            val = float(str(raw_val).split()[0])
+        except Exception:
+            continue
+        out.append({"ts": item.get("ts"), "value": val})
+    return out
+
+
+@router.get("/{ups_name}/events")
+async def ups_events(ups_name: str, user=Depends(require_session)):
+    r = get_redis()
+    raw = r.lrange(f"ups:event:list:{ups_name}", 0, 99)
+    parsed = []
+    for item in raw:
+        if "|" in item:
+            try:
+                ts_s, kind, rest = item.split("|", 2)
+                parsed.append({"ts": int(ts_s), "type": kind, "detail": rest})
+                continue
+            except Exception:
+                pass
+        parsed.append({"raw": item})
+    return parsed
+
+
+@router.get("/{ups_name}/energy")
+async def ups_energy(ups_name: str, user=Depends(require_session)):
+    r = get_redis()
+    day_str = time.strftime("%Y%m%d")
+    key = f"ups:energy:{ups_name}:{day_str}"
+    watt_seconds = r.get(key)
+    cfg = load_config()
+    rate = cfg.ui.energy_cost_per_kwh if hasattr(cfg.ui, "energy_cost_per_kwh") else 0.0
+    if watt_seconds:
+        try:
+            ws = float(watt_seconds)
+            kwh = ws / 3600.0 / 1000.0
+            return {"kwh_today": round(kwh, 4), "cost_today": round(kwh * rate, 4)}
+        except ValueError:
+            pass
+    return {"kwh_today": None, "cost_today": None}
+
+
+@router.get("/{ups_name}/watts_per_minute")
+async def ups_watts_per_minute(ups_name: str, user=Depends(require_session)):
+    r = get_redis()
+    raw = r.lrange(f"ups:watts:permin:{ups_name}", 0, 1440)
+    out = []
+    for item in raw:
+        if "|" in item:
+            minute, avg = item.split("|", 1)
+            try:
+                out.append({"minute": minute, "avg_watts": float(avg)})
+            except ValueError:
+                continue
+    out.reverse()
+    return out
+
+
+@router.get("/{ups_name}/health")
+async def ups_health(ups_name: str, user=Depends(require_session)):
+    r = get_redis()
+    alert_raw = r.lrange(f"ups:alerts:recent:{ups_name}", 0, 19)
+    alerts = []
+    for a in alert_raw:
+        parts = a.split("|", 2)
+        if len(parts) == 3:
+            try:
+                alerts.append({"ts": int(parts[0]), "severity": parts[1], "msg": parts[2]})
+                continue
+            except ValueError:
+                pass
+        if "|" in a:
+            ts_s, msg = a.split("|", 1)
+            try:
+                alerts.append({"ts": int(ts_s), "msg": msg})
+                continue
+            except ValueError:
+                pass
+        alerts.append({"raw": a})
+    dev_samples = r.lrange(f"ups:volt:dev:samples:{ups_name}", 0, 49)
+    dev_vals = []
+    for d in dev_samples:
+        try:
+            dev_vals.append(float(d))
+        except ValueError:
+            continue
+    dev_avg = sum(dev_vals) / len(dev_vals) if dev_vals else None
+    dev_max = max(dev_vals) if dev_vals else None
+    last_ok = r.get(f"ups:health:last_ok:{ups_name}")
+    is_offline = bool(r.get(f"ups:health:offline:{ups_name}"))
+    fail_count = r.get(f"ups:health:fail_count:{ups_name}")
+    events = r.lrange(f"ups:event:list:{ups_name}", 0, 200)
+    now = int(time.time())
+    onbatt_hour = 0
+    for ev in events:
+        parts = ev.split("|")
+        if len(parts) >= 3:
+            try:
+                ts_e = int(parts[0])
+            except ValueError:
+                continue
+            if now - ts_e > 3600:
+                continue
+            if parts[1] == "STATUS" and "ONBATT" in parts[2]:
+                onbatt_hour += 1
+    return {
+        "online": not is_offline,
+        "last_ok_ts": int(last_ok) if last_ok else None,
+        "fail_count": int(fail_count) if fail_count else 0,
+        "alerts": alerts,
+        "voltage_deviation": {
+            "avg_pct": round(dev_avg, 2) if dev_avg is not None else None,
+            "max_pct": round(dev_max, 2) if dev_max is not None else None,
+            "samples": len(dev_vals),
+        },
+        "onbatt_last_hour": onbatt_hour,
+    }
+
+
+@router.get("/{ups_name}/battery_health")
+async def battery_health(ups_name: str, user=Depends(require_session)):
+    """Return battery-health trend from sampled history."""
+    r = get_redis()
+    raw = r.lrange(f"ups:battery:history:{ups_name}", 0, -1)
+    samples = []
+    for item in raw:
+        try:
+            samples.append(json.loads(item))
+        except Exception:
+            continue
+    samples.sort(key=lambda s: s.get("ts", 0))
+    # Estimate runtime-at-full-charge: TIMELEFT * (100 / BCHARGE) when BCHARGE>0
+    normalized = []
+    for s in samples:
+        bc = s.get("bcharge")
+        tl = s.get("timeleft")
+        if bc and tl and bc > 10:
+            normalized.append({"ts": s["ts"], "est_full_runtime_min": tl * 100.0 / bc})
+    slope_per_day = None
+    decline_pct_14d = None
+    if len(normalized) >= 10:
+        first = normalized[0]
+        last = normalized[-1]
+        span_days = max(1e-6, (last["ts"] - first["ts"]) / 86400.0)
+        slope_per_day = (last["est_full_runtime_min"] - first["est_full_runtime_min"]) / span_days
+        # Compute 14d decline as % of initial runtime
+        cutoff_14d = last["ts"] - 14 * 86400
+        older = [n for n in normalized if n["ts"] <= cutoff_14d]
+        if older:
+            baseline = older[0]["est_full_runtime_min"]
+            if baseline > 0:
+                decline_pct_14d = (
+                    (baseline - last["est_full_runtime_min"]) / baseline * 100.0
+                )
+    return {
+        "samples": len(samples),
+        "estimated_full_runtime_samples": normalized[-200:],
+        "slope_per_day": round(slope_per_day, 3) if slope_per_day is not None else None,
+        "decline_pct_14d": round(decline_pct_14d, 2) if decline_pct_14d is not None else None,
+    }
+
+
+@router.get("/{ups_name}/export")
+async def export_ups(
+    ups_name: str,
+    format: str = Query("csv"),
+    since_days: int = Query(7, ge=1, le=30),
+    kind: str = Query("history"),
+    user=Depends(require_session),
+):
+    if format != "csv":
+        raise HTTPException(status_code=400, detail="Only csv format is supported")
+    if kind == "history":
+        gen = export_history_csv(ups_name, since_days)
+        filename = f"{ups_name}_history_{since_days}d.csv"
+    elif kind == "events":
+        gen = export_events_csv(ups_name, since_days)
+        filename = f"{ups_name}_events_{since_days}d.csv"
+    elif kind == "energy":
+        gen = export_energy_csv(ups_name, since_days)
+        filename = f"{ups_name}_energy_{since_days}d.csv"
+    else:
+        raise HTTPException(status_code=400, detail="kind must be history|events|energy")
+    return StreamingResponse(
+        gen,
+        media_type="text/csv",
+        headers={"Content-Disposition": f'attachment; filename="{filename}"'},
+    )
+
+
+@router.get("/{ups_name}/debug")
+async def ups_debug(ups_name: str, user=Depends(require_session)):
+    cfg = load_config()
+    target = next((u for u in cfg.ups if u.name == ups_name), None)
+    if not target:
+        raise HTTPException(status_code=404, detail="UPS not found")
+    try:
+        return await fetch_status(target.host, target.port)
+    except APCStatusError as e:
+        raise HTTPException(status_code=502, detail=str(e))
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.get("/{ups_name}/ui_tiles")
+async def get_ups_ui_tiles(ups_name: str, user=Depends(require_session)):
+    r = get_redis()
+    raw = r.get(f"ups:ui:tiles:{ups_name}")
+    default = {"types": {}, "order": [], "hidden": [], "custom": [], "positions": {}}
+    if not raw:
+        return default
+    try:
+        data = orjson.loads(raw)
+        return {
+            "types": data.get("types", {}),
+            "order": data.get("order", []),
+            "hidden": data.get("hidden", []),
+            "custom": data.get("custom", []),
+            "positions": data.get("positions", {}),
+        }
+    except Exception:
+        return default
+
+
+@router.post("/{ups_name}/ui_tiles")
+async def save_ups_ui_tiles(
+    ups_name: str, payload: dict, user=Depends(require_session_and_csrf)
+):
+    types = payload.get("types") if isinstance(payload.get("types"), dict) else {}
+    order = payload.get("order") if isinstance(payload.get("order"), list) else []
+    hidden = payload.get("hidden") if isinstance(payload.get("hidden"), list) else []
+    custom = payload.get("custom") if isinstance(payload.get("custom"), list) else []
+    positions = payload.get("positions") if isinstance(payload.get("positions"), dict) else {}
+    norm_custom = []
+    for c in custom or []:
+        if not isinstance(c, dict):
+            continue
+        metric = c.get("metric")
+        chart = c.get("chart")
+        if not metric or not chart:
+            continue
+        norm_custom.append({
+            "id": c.get("id") or "",
+            "metric": metric,
+            "chart": chart,
+            "source": c.get("source", "live"),
+        })
+    doc = {
+        "types": types,
+        "order": order,
+        "hidden": hidden,
+        "custom": norm_custom,
+        "positions": positions,
+        "saved_ts": int(time.time()),
+    }
+    r = get_redis()
+    r.set(f"ups:ui:tiles:{ups_name}", orjson.dumps(doc))
+    return {"message": "saved", "count_custom": len(norm_custom)}
+
+
+@router.delete("/{ups_name}/ui_tiles")
+async def clear_ups_ui_tiles(ups_name: str, user=Depends(require_session_and_csrf)):
+    r = get_redis()
+    r.delete(f"ups:ui:tiles:{ups_name}")
+    return {"message": "cleared"}

app/routes/pages.py

@@ -0,0 +1,118 @@
+"""HTML page routes (dashboard, config, login, setup, events, alerts, settings)."""
+from fastapi import APIRouter, Request
+from fastapi.responses import HTMLResponse, RedirectResponse
+from fastapi.templating import Jinja2Templates
+
+from ..auth import CSRF_COOKIE, current_user, is_admin_configured, make_csrf_token
+from ..config import load_config
+
+router = APIRouter()
+templates = Jinja2Templates(directory="app/templates")
+
+
+def _ensure_csrf(request: Request, response):
+    token = request.cookies.get(CSRF_COOKIE)
+    if not token:
+        token = make_csrf_token()
+        response.set_cookie(
+            CSRF_COOKIE, token, httponly=False, samesite="lax", path="/"
+        )
+    return token
+
+
+def _session_or_setup_redirect(request: Request):
+    if not is_admin_configured():
+        return RedirectResponse("/setup", status_code=302)
+    user = current_user(request)
+    if not user:
+        return RedirectResponse("/login", status_code=302)
+    return None
+
+
+@router.get("/", response_class=HTMLResponse)
+async def dashboard(request: Request):
+    redirect = _session_or_setup_redirect(request)
+    if redirect:
+        return redirect
+    cfg = load_config()
+    user = current_user(request)
+    response = templates.TemplateResponse(
+        "dashboard.html",
+        {
+            "request": request,
+            "ups_list": cfg.ups,
+            "ui_cfg": cfg.ui.model_dump(),
+            "current_user": user,
+            "active_nav": "dashboard",
+        },
+    )
+    _ensure_csrf(request, response)
+    return response
+
+
+@router.get("/config", response_class=HTMLResponse)
+async def config_page(request: Request):
+    redirect = _session_or_setup_redirect(request)
+    if redirect:
+        return redirect
+    response = templates.TemplateResponse(
+        "config.html",
+        {"request": request, "current_user": current_user(request), "active_nav": "config"},
+    )
+    _ensure_csrf(request, response)
+    return response
+
+
+@router.get("/events", response_class=HTMLResponse)
+async def events_page(request: Request):
+    redirect = _session_or_setup_redirect(request)
+    if redirect:
+        return redirect
+    response = templates.TemplateResponse(
+        "events.html",
+        {"request": request, "current_user": current_user(request), "active_nav": "events"},
+    )
+    _ensure_csrf(request, response)
+    return response
+
+
+@router.get("/alerts", response_class=HTMLResponse)
+async def alerts_page(request: Request):
+    redirect = _session_or_setup_redirect(request)
+    if redirect:
+        return redirect
+    response = templates.TemplateResponse(
+        "alerts.html",
+        {"request": request, "current_user": current_user(request), "active_nav": "alerts"},
+    )
+    _ensure_csrf(request, response)
+    return response
+
+
+@router.get("/settings", response_class=HTMLResponse)
+async def settings_page(request: Request):
+    redirect = _session_or_setup_redirect(request)
+    if redirect:
+        return redirect
+    response = templates.TemplateResponse(
+        "settings.html",
+        {"request": request, "current_user": current_user(request), "active_nav": "settings"},
+    )
+    _ensure_csrf(request, response)
+    return response
+
+
+@router.get("/login", response_class=HTMLResponse)
+async def login_page(request: Request):
+    if not is_admin_configured():
+        return RedirectResponse("/setup", status_code=302)
+    if current_user(request):
+        return RedirectResponse("/", status_code=302)
+    return templates.TemplateResponse("login.html", {"request": request})
+
+
+@router.get("/setup", response_class=HTMLResponse)
+async def setup_page(request: Request):
+    if is_admin_configured():
+        return RedirectResponse("/login", status_code=302)
+    return templates.TemplateResponse("setup.html", {"request": request})

app/routes/sse.py

@@ -0,0 +1,41 @@
+"""Server-Sent Events stream with snapshots + fleet overview."""
+import asyncio
+
+import orjson
+from fastapi import APIRouter, Depends
+from fastapi.responses import StreamingResponse
+
+from ..auth import require_session
+from ..config import load_config
+from ..config_manager import get_config_version
+from ..storage import get_latest, get_redis
+
+router = APIRouter()
+
+
+@router.get("/api/stream")
+async def stream(user=Depends(require_session)):
+    async def event_gen():
+        while True:
+            cfg = load_config()
+            r = get_redis()
+            payload = {"snapshots": {}, "cfgVersion": get_config_version()}
+            for u in cfg.ups:
+                snap = await get_latest(u.name)
+                if snap:
+                    payload["snapshots"][u.name] = snap
+            payload["upsMeta"] = [
+                {
+                    "name": u.name,
+                    "host": u.host,
+                    "port": u.port,
+                    "offline": bool(r.get(f"ups:health:offline:{u.name}")),
+                }
+                for u in cfg.ups
+            ]
+            for name, snap in payload["snapshots"].items():
+                payload.setdefault(name, snap)
+            yield f"data: {orjson.dumps(payload).decode()}\n\n"
+            await asyncio.sleep(5)
+
+    return StreamingResponse(event_gen(), media_type="text/event-stream")

app/security.py

@@ -0,0 +1,41 @@
+"""Security headers middleware.
+
+Emits a conservative CSP that allows only self + jsdelivr (Chart.js) and
+standard hardening headers. HSTS only when behind HTTPS proxy.
+"""
+from __future__ import annotations
+
+from starlette.middleware.base import BaseHTTPMiddleware
+from starlette.types import ASGIApp
+
+from .settings import settings
+
+CSP_DIRECTIVES = (
+    "default-src 'self'; "
+    "script-src 'self' https://cdn.jsdelivr.net 'unsafe-inline'; "
+    "style-src 'self' 'unsafe-inline'; "
+    "img-src 'self' data:; "
+    "font-src 'self' data:; "
+    "connect-src 'self'; "
+    "base-uri 'self'; "
+    "form-action 'self'; "
+    "frame-ancestors 'none'"
+)
+
+
+class SecurityHeadersMiddleware(BaseHTTPMiddleware):
+    def __init__(self, app: ASGIApp):
+        super().__init__(app)
+
+    async def dispatch(self, request, call_next):
+        response = await call_next(request)
+        response.headers.setdefault("X-Content-Type-Options", "nosniff")
+        response.headers.setdefault("X-Frame-Options", "DENY")
+        response.headers.setdefault("Referrer-Policy", "same-origin")
+        response.headers.setdefault("Content-Security-Policy", CSP_DIRECTIVES)
+        if settings.trust_proxy:
+            response.headers.setdefault(
+                "Strict-Transport-Security",
+                "max-age=31536000; includeSubDomains",
+            )
+        return response

app/settings.py

@@ -0,0 +1,81 @@
+"""Centralized settings loaded once at startup via pydantic-settings.
+
+Reads environment variables once and exposes a cached Settings() instance.
+Replaces scattered os.environ.get() calls throughout the codebase.
+"""
+from __future__ import annotations
+
+from functools import lru_cache
+
+from pydantic import Field
+from pydantic_settings import BaseSettings, SettingsConfigDict
+
+
+class Settings(BaseSettings):
+    """Application settings pulled from environment variables."""
+
+    model_config = SettingsConfigDict(
+        env_file=".env",
+        env_file_encoding="utf-8",
+        case_sensitive=False,
+        extra="ignore",
+    )
+
+    # Redis
+    redis_url: str = Field(
+        default="redis://redis:6379/0", description="Redis connection URL"
+    )
+
+    # Auth
+    admin_username: str = Field(default="admin", description="Admin username")
+    admin_password_hash: str | None = Field(
+        default=None,
+        description="Argon2 password hash. If unset, first-run setup is required.",
+    )
+    session_secret: str | None = Field(
+        default=None,
+        description="Signing secret for session cookies (required in prod).",
+    )
+    session_max_age_seconds: int = Field(
+        default=14 * 24 * 3600, description="Session cookie max age"
+    )
+    trust_proxy: bool = Field(
+        default=False,
+        description="Set true when behind HTTPS reverse proxy (enables Secure cookie + HSTS)",
+    )
+
+    # SMTP
+    smtp_password: str | None = Field(
+        default=None, description="SMTP password (not stored in Redis)"
+    )
+
+    # Network validation
+    allow_private_ips: bool = Field(
+        default=True,
+        description="Allow private/RFC1918 hosts in UPSConfig.host. Default true for homelab.",
+    )
+
+    # Legacy / migration
+    ups_config_path: str = Field(
+        default="/config/ups.yaml",
+        description="Legacy YAML path (migration-only).",
+    )
+
+    # Logging
+    log_level: str = Field(default="INFO", description="Python log level")
+
+    # Timezone
+    tz: str = Field(default="UTC", description="Server timezone")
+
+    # Rate limit toggle (disable in tests)
+    rate_limit_enabled: bool = Field(
+        default=True, description="Enable slowapi rate limiting"
+    )
+
+
+@lru_cache(maxsize=1)
+def get_settings() -> Settings:
+    return Settings()
+
+
+settings = get_settings()

app/static/css/base.css

@@ -0,0 +1,57 @@
+/* Base layout, header/nav */
+* { box-sizing: border-box; }
+body { margin: 0; font-family: -apple-system, Segoe UI, Arial, sans-serif; background: #f5f7fa; color: #222; }
+.site-header {
+  display: flex; align-items: center; gap: 24px;
+  padding: 10px 20px; background: #2c3e50; color: #ecf0f1;
+  box-shadow: 0 2px 4px rgba(0,0,0,.1);
+}
+.site-header .brand a { color: #ecf0f1; text-decoration: none; font-weight: 600; font-size: 18px; }
+.main-nav { display: flex; gap: 16px; flex: 1; flex-wrap: wrap; }
+.main-nav a { color: #bdc3c7; text-decoration: none; padding: 6px 10px; border-radius: 4px; }
+.main-nav a:hover { background: #34495e; color: #fff; }
+.main-nav a.active { background: #3498db; color: #fff; }
+.user-menu { display: flex; align-items: center; gap: 10px; }
+.user-menu .username { font-size: 13px; opacity: 0.85; }
+.btn-logout { background: #e74c3c; color: #fff; border: 0; padding: 6px 12px; border-radius: 4px; cursor: pointer; }
+.btn-logout:hover { background: #c0392b; }
+.site-main { padding: 20px; }
+.page-header { display: flex; justify-content: space-between; align-items: center; margin-bottom: 16px; gap: 16px; flex-wrap: wrap; }
+.page-header h1 { margin: 0; font-size: 22px; }
+.filters { display: flex; gap: 8px; flex-wrap: wrap; }
+.filters select, .filters input { padding: 6px 10px; border: 1px solid #ccc; border-radius: 4px; }
+
+/* Tables */
+.data-table { width: 100%; border-collapse: collapse; background: #fff; border-radius: 6px; overflow: hidden; box-shadow: 0 1px 3px rgba(0,0,0,.08); }
+.data-table th, .data-table td { text-align: left; padding: 10px 12px; border-bottom: 1px solid #eee; font-size: 14px; }
+.data-table th { background: #f0f2f5; font-weight: 600; }
+.data-table tr:hover { background: #fafbfc; }
+.sev.CRITICAL { color: #fff; background: #c0392b; padding: 2px 8px; border-radius: 3px; font-size: 12px; }
+.sev.WARNING { color: #fff; background: #e67e22; padding: 2px 8px; border-radius: 3px; font-size: 12px; }
+.sev.INFO { color: #fff; background: #2980b9; padding: 2px 8px; border-radius: 3px; font-size: 12px; }
+
+/* Settings */
+.settings-section { background: #fff; border-radius: 6px; padding: 16px 20px; margin-bottom: 20px; box-shadow: 0 1px 3px rgba(0,0,0,.08); }
+.settings-section h2 { margin-top: 0; font-size: 18px; }
+.settings-section form { display: flex; flex-direction: column; gap: 10px; max-width: 480px; }
+.settings-section label { display: flex; flex-direction: column; gap: 4px; font-size: 14px; }
+.settings-section label[type=checkbox] input, .settings-section input[type=checkbox] + * { display: inline; }
+.settings-section input, .settings-section select { padding: 6px 8px; border: 1px solid #ccc; border-radius: 4px; }
+.settings-section button { align-self: flex-start; }
+.muted { color: #777; font-size: 13px; }
+
+/* Fleet overview */
+.fleet-overview { margin-bottom: 16px; }
+.fleet-card { background: #fff; border-radius: 6px; padding: 16px 20px; display: flex; gap: 24px; flex-wrap: wrap; box-shadow: 0 1px 3px rgba(0,0,0,.08); }
+.fleet-stat { display: flex; flex-direction: column; gap: 4px; }
+.fleet-stat-label { color: #7f8c8d; font-size: 12px; text-transform: uppercase; letter-spacing: .5px; }
+.fleet-stat-value { font-size: 22px; font-weight: 600; }
+.fleet-stat-value.online { color: #27ae60; }
+.fleet-stat-value.onbatt { color: #e67e22; }
+.fleet-stat-value.offline { color: #c0392b; }
+
+/* Card state dot */
+.state-dot { display: inline-block; width: 10px; height: 10px; border-radius: 50%; background: #95a5a6; margin-right: 6px; vertical-align: middle; }
+.state-dot.online { background: #27ae60; }
+.state-dot.offline { background: #c0392b; }
+.state-dot.onbatt { background: #e67e22; }

app/static/css/login.css

@@ -0,0 +1,11 @@
+.auth-body { background: #2c3e50; min-height: 100vh; display: flex; align-items: center; justify-content: center; }
+.auth-card { background: #fff; border-radius: 8px; padding: 32px; width: 360px; box-shadow: 0 8px 24px rgba(0,0,0,.3); }
+.auth-card.wide { width: 440px; }
+.auth-card h1 { margin: 0 0 20px; text-align: center; font-size: 22px; }
+.form-field { display: flex; flex-direction: column; gap: 6px; margin-bottom: 14px; }
+.form-field label { font-size: 13px; color: #555; }
+.form-field input { padding: 8px 10px; border: 1px solid #ccc; border-radius: 4px; font-size: 14px; }
+.btn-primary { background: #3498db; color: #fff; border: 0; padding: 10px 16px; border-radius: 4px; cursor: pointer; font-size: 14px; }
+.btn-primary:hover { background: #2980b9; }
+.btn-block { width: 100%; }
+.login-error { margin-top: 12px; padding: 8px 12px; background: #fee; color: #c0392b; border-radius: 4px; font-size: 13px; }

app/static/js/alerts.js

@@ -0,0 +1,80 @@
+// Alerts page (DOM-safe)
+(async function () {
+  function cell(text, cls) {
+    const td = document.createElement('td');
+    td.textContent = text;
+    if (cls) td.className = cls;
+    return td;
+  }
+
+  async function loadUps() {
+    const r = await window.apiFetch('/api/ups');
+    if (!r.ok) return;
+    const list = await r.json();
+    const sel = document.getElementById('filter-ups');
+    list.forEach(u => {
+      const o = document.createElement('option');
+      o.value = u.name; o.textContent = u.name;
+      sel.appendChild(o);
+    });
+  }
+
+  async function loadAlerts() {
+    const severity = document.getElementById('filter-severity').value;
+    const ups = document.getElementById('filter-ups').value;
+    const view = document.getElementById('filter-view').value;
+    const path = view === 'active' ? '/api/alerts/active' : '/api/alerts';
+    const params = new URLSearchParams();
+    if (severity) params.set('severity', severity);
+    if (ups) params.set('ups', ups);
+    if (view === 'history') params.set('days', '30');
+    const url = path + (params.toString() ? '?' + params.toString() : '');
+    const r = await window.apiFetch(url);
+    const tbody = document.getElementById('alerts-tbody');
+    tbody.textContent = '';
+    if (!r.ok) {
+      const tr = document.createElement('tr');
+      tr.appendChild(cell('Failed to load')); tbody.appendChild(tr); return;
+    }
+    const rows = await r.json();
+    if (!rows.length) {
+      const tr = document.createElement('tr');
+      tr.appendChild(cell('No alerts')); tbody.appendChild(tr); return;
+    }
+    rows.forEach(a => {
+      const tr = document.createElement('tr');
+      tr.appendChild(cell(new Date(a.ts * 1000).toLocaleString()));
+      const sevCell = document.createElement('td');
+      const sevSpan = document.createElement('span');
+      sevSpan.className = 'sev ' + a.severity;
+      sevSpan.textContent = a.severity;
+      sevCell.appendChild(sevSpan);
+      tr.appendChild(sevCell);
+      tr.appendChild(cell(a.ups));
+      tr.appendChild(cell(a.code || ''));
+      tr.appendChild(cell(a.message));
+      tr.appendChild(cell(a.acked_ts ? new Date(a.acked_ts * 1000).toLocaleString() : ''));
+      const actionCell = document.createElement('td');
+      if (!a.acked_ts) {
+        const btn = document.createElement('button');
+        btn.className = 'btn btn-secondary ack-btn';
+        btn.textContent = 'Ack';
+        btn.dataset.id = a.id;
+        btn.addEventListener('click', async () => {
+          await window.apiFetch('/api/alerts/' + btn.dataset.id + '/ack', { method: 'POST' });
+          loadAlerts();
+        });
+        actionCell.appendChild(btn);
+      }
+      tr.appendChild(actionCell);
+      tbody.appendChild(tr);
+    });
+  }
+
+  document.getElementById('refresh-btn').addEventListener('click', loadAlerts);
+  ['filter-severity', 'filter-ups', 'filter-view'].forEach(id => {
+    document.getElementById(id).addEventListener('change', loadAlerts);
+  });
+  await loadUps();
+  await loadAlerts();
+})();

app/static/js/app.js

@@ -0,0 +1,33 @@
+// Shared client-side helpers: CSRF, logout, fetch wrapper
+(function () {
+  function getCsrfToken() {
+    const meta = document.querySelector('meta[name="csrf-token"]');
+    if (meta && meta.content) return meta.content;
+    const m = document.cookie.match(/(?:^|;)\s*csrf_token=([^;]+)/);
+    return m ? decodeURIComponent(m[1]) : '';
+  }
+
+  window.apiFetch = async function (url, opts = {}) {
+    const method = (opts.method || 'GET').toUpperCase();
+    const headers = Object.assign({ Accept: 'application/json' }, opts.headers || {});
+    if (method !== 'GET' && method !== 'HEAD') {
+      headers['X-CSRF-Token'] = getCsrfToken();
+      if (opts.body && !(opts.body instanceof FormData) && !headers['Content-Type']) {
+        headers['Content-Type'] = 'application/json';
+      }
+    }
+    const resp = await fetch(url, Object.assign({ credentials: 'same-origin' }, opts, { headers }));
+    if (resp.status === 401) {
+      window.location.href = '/login';
+      throw new Error('unauthorized');
+    }
+    return resp;
+  };
+
+  document.addEventListener('click', async (e) => {
+    if (e.target && e.target.id === 'logout-btn') {
+      await window.apiFetch('/api/logout', { method: 'POST' });
+      window.location.href = '/login';
+    }
+  });
+})();

app/static/js/config.js

@@ -1,174 +1,113 @@
-// Configuration management JavaScript
+// UPS Configuration page (uses apiFetch for CSRF)
+
+function el(tag, attrs, ...children) {
+  const e = document.createElement(tag);
+  if (attrs) Object.entries(attrs).forEach(([k, v]) => {
+    if (k === 'class') e.className = v;
+    else if (k === 'dataset') Object.assign(e.dataset, v);
+    else if (k.startsWith('on') && typeof v === 'function') e.addEventListener(k.slice(2), v);
+    else if (v !== null && v !== undefined) e.setAttribute(k, v);
+  });
+  children.forEach(c => {
+    if (c == null) return;
+    e.appendChild(typeof c === 'string' ? document.createTextNode(c) : c);
+  });
+  return e;
+}
 
 class ConfigManager {
-  constructor() {
+  constructor() { this.confirmCallback = null; this.init(); }
-    this.init();
-  }
 
-  init() {
+  init() { this.bindEvents(); this.loadUPSList(); }
-    this.bindEvents();
-    this.loadUPSList();
-  }
 
   bindEvents() {
-    // Add UPS button
+    document.getElementById('add-ups-btn').addEventListener('click', () => this.showUPSModal());
-    document.getElementById('add-ups-btn').addEventListener('click', () => {
-      this.showUPSModal();
-    });
-
-    // Modal close buttons
     document.querySelectorAll('.close').forEach(close => {
-      close.addEventListener('click', (e) => {
+      close.addEventListener('click', (e) => this.closeModal(e.target.closest('.modal')));
-        this.closeModal(e.target.closest('.modal'));
     });
-    });
-
-    // UPS form submission
     document.getElementById('ups-form').addEventListener('submit', (e) => {
-      e.preventDefault();
+      e.preventDefault(); this.saveUPS();
-      this.saveUPS();
     });
-
+    document.getElementById('test-connection-btn').addEventListener('click', () => this.testConnection());
-    // Test connection button
-    document.getElementById('test-connection-btn').addEventListener('click', () => {
-      this.testConnection();
-    });
-
-    // Cancel button
     document.getElementById('cancel-btn').addEventListener('click', () => {
       this.closeModal(document.getElementById('ups-modal'));
     });
-
-    // Confirmation modal buttons
     document.getElementById('confirm-yes').addEventListener('click', () => {
-      if (this.confirmCallback) {
+      if (this.confirmCallback) this.confirmCallback();
-        this.confirmCallback();
-      }
       this.closeModal(document.getElementById('confirm-modal'));
     });
-
     document.getElementById('confirm-no').addEventListener('click', () => {
       this.closeModal(document.getElementById('confirm-modal'));
     });
-
-    // Click outside modal to close
     window.addEventListener('click', (e) => {
-      if (e.target.classList.contains('modal')) {
+      if (e.target.classList.contains('modal')) this.closeModal(e.target);
-        this.closeModal(e.target);
-      }
     });
   }
 
   async loadUPSList() {
     try {
-      const response = await fetch('/api/config/ups');
+      const resp = await window.apiFetch('/api/config/ups');
-      const upsList = await response.json();
+      const list = await resp.json();
-      
+      if (!resp.ok) throw new Error(list.detail || 'Failed to load');
-      if (!response.ok) {
+      this.renderUPSList(list);
-        throw new Error(upsList.detail || 'Failed to load UPS configurations');
+    } catch (err) { this.showToast('Error: ' + err.message, 'error'); }
   }
 
-      this.renderUPSList(upsList);
+  renderUPSList(list) {
-    } catch (error) {
-      this.showToast('Error loading UPS configurations: ' + error.message, 'error');
-    }
-  }
-
-  renderUPSList(upsList) {
     const container = document.getElementById('ups-list');
-    
+    container.textContent = '';
-    if (upsList.length === 0) {
+    if (list.length === 0) {
-      container.innerHTML = `
+      const empty = el('div', { class: 'empty-state' },
-        <div class="empty-state">
+        el('h3', null, 'No UPS Configured'),
-          <h3>No UPS Configured</h3>
+        el('p', null, 'Click "Add UPS" to configure your first UPS.'));
-          <p>Get started by clicking "Add UPS" above to configure your first UPS monitoring.</p>
+      container.appendChild(empty);
-        </div>
-      `;
       return;
     }
-
+    list.forEach(ups => container.appendChild(this.renderUPSItem(ups)));
-    container.innerHTML = upsList.map(ups => this.renderUPSItem(ups)).join('');
-    
-    // Bind action buttons
-    container.querySelectorAll('.edit-btn').forEach(btn => {
-      btn.addEventListener('click', (e) => {
-        const upsName = e.target.dataset.upsName;
-        this.editUPS(upsName);
-      });
-    });
-
-    container.querySelectorAll('.delete-btn').forEach(btn => {
-      btn.addEventListener('click', (e) => {
-        const upsName = e.target.dataset.upsName;
-        this.confirmDelete(upsName);
-      });
-    });
-
-    container.querySelectorAll('.test-btn').forEach(btn => {
-      btn.addEventListener('click', (e) => {
-        const upsName = e.target.dataset.upsName;
-        this.testUPSConnection(upsName);
-      });
-    });
   }
 
   renderUPSItem(ups) {
-    const alertsEnabled = [
+    const alerts = [];
-      ups.alert_loadpct_high && `Load: ${ups.alert_loadpct_high}%`,
+    if (ups.alert_loadpct_high) alerts.push('Load: ' + ups.alert_loadpct_high + '%');
-      ups.alert_bcharge_low && `Battery: ${ups.alert_bcharge_low}%`,
+    if (ups.alert_bcharge_low) alerts.push('Battery: ' + ups.alert_bcharge_low + '%');
-      ups.alert_on_battery && 'On Battery',
+    if (ups.alert_on_battery) alerts.push('On Battery');
-      ups.alert_runtime_low_minutes && `Runtime: ${ups.alert_runtime_low_minutes}min`
+    if (ups.alert_runtime_low_minutes) alerts.push('Runtime: ' + ups.alert_runtime_low_minutes + 'min');
-    ].filter(Boolean);
+    if (ups.alert_itemp_high) alerts.push('Temp: ' + ups.alert_itemp_high + '°C');
-
+    const testBtn = el('button', { class: 'btn btn-secondary', onclick: () => this.testUPSConnection(ups.name) }, 'Test');
-    return `
+    const editBtn = el('button', { class: 'btn btn-primary', onclick: () => this.editUPS(ups.name) }, 'Edit');
-      <div class="ups-item">
+    const delBtn = el('button', { class: 'btn btn-danger', onclick: () => this.confirmDelete(ups.name) }, 'Delete');
-        <div class="ups-item-header">
+    const item = el('div', { class: 'ups-item' },
-          <div class="ups-item-title">${ups.name}</div>
+      el('div', { class: 'ups-item-header' },
-          <div class="ups-item-actions">
+        el('div', { class: 'ups-item-title' }, ups.name),
-            <button class="btn btn-secondary test-btn" data-ups-name="${ups.name}">Test</button>
+        el('div', { class: 'ups-item-actions' }, testBtn, editBtn, delBtn)),
-            <button class="btn btn-primary edit-btn" data-ups-name="${ups.name}">Edit</button>
+      el('div', { class: 'ups-item-details' },
-            <button class="btn btn-danger delete-btn" data-ups-name="${ups.name}">Delete</button>
+        el('div', { class: 'ups-detail' },
-          </div>
+          el('div', { class: 'ups-detail-label' }, 'Host'),
-        </div>
+          el('div', { class: 'ups-detail-value' }, ups.host + ':' + ups.port)),
-        <div class="ups-item-details">
+        el('div', { class: 'ups-detail' },
-          <div class="ups-detail">
+          el('div', { class: 'ups-detail-label' }, 'Polling Interval'),
-            <div class="ups-detail-label">Host</div>
+          el('div', { class: 'ups-detail-value' }, ups.interval_seconds + 's')),
-            <div class="ups-detail-value">${ups.host}:${ups.port}</div>
+        el('div', { class: 'ups-detail' },
-          </div>
+          el('div', { class: 'ups-detail-label' }, 'Alerts'),
-          <div class="ups-detail">
+          el('div', { class: 'ups-detail-value' }, alerts.length ? alerts.join(', ') : 'None'))));
-            <div class="ups-detail-label">Polling Interval</div>
+    return item;
-            <div class="ups-detail-value">${ups.interval_seconds}s</div>
-          </div>
-          <div class="ups-detail">
-            <div class="ups-detail-label">Alerts</div>
-            <div class="ups-detail-value">${alertsEnabled.length > 0 ? alertsEnabled.join(', ') : 'None'}</div>
-          </div>
-        </div>
-      </div>
-    `;
   }
 
   showUPSModal(ups = null) {
     const modal = document.getElementById('ups-modal');
     const title = document.getElementById('modal-title');
     const form = document.getElementById('ups-form');
-    
-    // Reset form
     form.reset();
-    
     if (ups) {
-      // Edit mode
       title.textContent = 'Edit UPS';
       this.populateForm(ups);
       form.dataset.mode = 'edit';
       form.dataset.upsName = ups.name;
     } else {
-      // Add mode
       title.textContent = 'Add UPS';
       form.dataset.mode = 'add';
       delete form.dataset.upsName;
     }
-    
     modal.style.display = 'block';
   }
 
@@ -179,222 +118,107 @@ class ConfigManager {
     document.getElementById('ups-interval').value = ups.interval_seconds || 30;
     document.getElementById('ups-loadpct').value = ups.alert_loadpct_high || '';
     document.getElementById('ups-bcharge').value = ups.alert_bcharge_low || '';
-    document.getElementById('ups-onbattery').checked = ups.alert_on_battery || false;
+    document.getElementById('ups-onbattery').checked = !!ups.alert_on_battery;
     document.getElementById('ups-runtime').value = ups.alert_runtime_low_minutes || '';
+    const itemp = document.getElementById('ups-itemp');
+    if (itemp) itemp.value = ups.alert_itemp_high || '';
   }
 
-  closeModal(modal) {
+  closeModal(modal) { modal.style.display = 'none'; }
-    modal.style.display = 'none';
-  }
 
   async saveUPS() {
     const form = document.getElementById('ups-form');
-    const formData = new FormData(form);
+    const fd = new FormData(form);
     const mode = form.dataset.mode;
-    const upsName = form.dataset.upsName;
+    const name = form.dataset.upsName;
-    
     const data = {
-      name: formData.get('name'),
+      name: fd.get('name'),
-      host: formData.get('host'),
+      host: fd.get('host'),
-      port: parseInt(formData.get('port')) || 3551,
+      port: parseInt(fd.get('port')) || 3551,
-      interval_seconds: parseInt(formData.get('interval_seconds')) || 30,
+      interval_seconds: parseInt(fd.get('interval_seconds')) || 30,
-      alert_on_battery: formData.has('alert_on_battery')
+      alert_on_battery: fd.has('alert_on_battery'),
     };
-
+    const num = (k) => { const v = fd.get(k); if (v) data[k] = parseFloat(v); };
-    // Add optional alert thresholds
+    num('alert_loadpct_high'); num('alert_bcharge_low');
-    const loadpct = formData.get('alert_loadpct_high');
+    num('alert_runtime_low_minutes'); num('alert_itemp_high');
-    if (loadpct) data.alert_loadpct_high = parseFloat(loadpct);
-    
-    const bcharge = formData.get('alert_bcharge_low');
-    if (bcharge) data.alert_bcharge_low = parseFloat(bcharge);
-    
-    const runtime = formData.get('alert_runtime_low_minutes');
-    if (runtime) data.alert_runtime_low_minutes = parseFloat(runtime);
-
     try {
-      let response;
+      const url = mode === 'edit' ? '/api/config/ups/' + name : '/api/config/ups';
-      if (mode === 'edit') {
+      const method = mode === 'edit' ? 'PUT' : 'POST';
-        response = await fetch(`/api/config/ups/${upsName}`, {
+      const resp = await window.apiFetch(url, { method, body: JSON.stringify(data) });
-          method: 'PUT',
+      const result = await resp.json();
-          headers: { 'Content-Type': 'application/json' },
+      if (!resp.ok) throw new Error(result.detail || 'Save failed');
-          body: JSON.stringify(data)
-        });
-      } else {
-        response = await fetch('/api/config/ups', {
-          method: 'POST',
-          headers: { 'Content-Type': 'application/json' },
-          body: JSON.stringify(data)
-        });
-      }
-
-      const result = await response.json();
-      
-      if (!response.ok) {
-        throw new Error(result.detail || 'Failed to save UPS configuration');
-      }
-
       this.showToast(result.message, 'success');
       this.closeModal(document.getElementById('ups-modal'));
       this.loadUPSList();
-    } catch (error) {
+    } catch (err) { this.showToast('Error: ' + err.message, 'error'); }
-      this.showToast('Error saving UPS: ' + error.message, 'error');
-    }
   }
 
-  async editUPS(upsName) {
+  async editUPS(name) {
     try {
-      const response = await fetch(`/api/config/ups/${upsName}`);
+      const resp = await window.apiFetch('/api/config/ups/' + name);
-      const ups = await response.json();
+      const ups = await resp.json();
-      
+      if (!resp.ok) throw new Error(ups.detail || 'Load failed');
-      if (!response.ok) {
-        throw new Error(ups.detail || 'Failed to load UPS configuration');
-      }
-
       this.showUPSModal(ups);
-    } catch (error) {
+    } catch (err) { this.showToast('Error: ' + err.message, 'error'); }
-      this.showToast('Error loading UPS configuration: ' + error.message, 'error');
-    }
   }
 
-  confirmDelete(upsName) {
+  confirmDelete(name) {
-    const modal = document.getElementById('confirm-modal');
+    document.getElementById('confirm-message').textContent =
-    const message = document.getElementById('confirm-message');
+      'Delete UPS "' + name + '"? This cannot be undone.';
-    
+    this.confirmCallback = () => this.deleteUPS(name);
-    message.textContent = `Are you sure you want to delete the UPS configuration "${upsName}"? This action cannot be undone.`;
+    document.getElementById('confirm-modal').style.display = 'block';
-    
-    this.confirmCallback = () => this.deleteUPS(upsName);
-    modal.style.display = 'block';
   }
 
-  async deleteUPS(upsName) {
+  async deleteUPS(name) {
     try {
-      const response = await fetch(`/api/config/ups/${upsName}`, {
+      const resp = await window.apiFetch('/api/config/ups/' + name, { method: 'DELETE' });
-        method: 'DELETE'
+      const result = await resp.json();
-      });
+      if (!resp.ok) throw new Error(result.detail || 'Delete failed');
-
-      const result = await response.json();
-      
-      if (!response.ok) {
-        throw new Error(result.detail || 'Failed to delete UPS configuration');
-      }
-
       this.showToast(result.message, 'success');
       this.loadUPSList();
-    } catch (error) {
+    } catch (err) { this.showToast('Error: ' + err.message, 'error'); }
-      this.showToast('Error deleting UPS: ' + error.message, 'error');
-    }
   }
 
   async testConnection() {
     const form = document.getElementById('ups-form');
-    const formData = new FormData(form);
+    const fd = new FormData(form);
-    
     const data = {
-      name: formData.get('name') || 'test',
+      name: fd.get('name') || 'test',
-      host: formData.get('host'),
+      host: fd.get('host'),
-      port: parseInt(formData.get('port')) || 3551,
+      port: parseInt(fd.get('port')) || 3551,
       interval_seconds: 30,
-      alert_on_battery: false
+      alert_on_battery: false,
     };
-
+    if (!data.host) { this.showToast('Host required', 'error'); return; }
-    if (!data.host) {
+    const btn = document.getElementById('test-connection-btn');
-      this.showToast('Host is required for connection test', 'error');
+    const orig = btn.textContent;
-      return;
+    btn.textContent = 'Testing...'; btn.disabled = true;
-    }
-
-    const testBtn = document.getElementById('test-connection-btn');
-    const originalText = testBtn.textContent;
-    testBtn.textContent = 'Testing...';
-    testBtn.disabled = true;
-
     try {
-      const response = await fetch('/api/config/ups/test', {
+      const resp = await window.apiFetch('/api/config/ups/test', {
-        method: 'POST',
+        method: 'POST', body: JSON.stringify(data),
-        headers: { 'Content-Type': 'application/json' },
-        body: JSON.stringify(data)
       });
-
+      const r = await resp.json();
-      const result = await response.json();
+      this.showToast(r.message || (r.success ? 'OK' : 'Failed'),
-      let msg;
+        r.success ? 'success' : 'error');
-      if (result.success) {
+    } catch (err) { this.showToast('Test failed: ' + err.message, 'error'); }
-        msg = 'Connection successful';
+    finally { btn.textContent = orig; btn.disabled = false; }
-        if (result.data && result.data.STATUS) {
-          msg += ` (STATUS=${result.data.STATUS})`;
-        }
-        this.showToast(msg, 'success');
-      } else {
-        if (result.connectivity && !result.connectivity.ok) {
-          msg = 'TCP connectivity failed: ' + (result.connectivity.error || 'unknown error');
-        } else if (result.protocol && !result.protocol.ok) {
-          msg = 'Protocol error after TCP success: ' + (result.protocol.error || 'unknown error');
-        } else {
-          msg = result.message || 'Connection failed';
-        }
-        this.showToast(msg, 'error');
-      }
-    } catch (error) {
-      this.showToast('Connection test failed: ' + error.message, 'error');
-    } finally {
-      testBtn.textContent = originalText;
-      testBtn.disabled = false;
-    }
   }
 
-  async testUPSConnection(upsName) {
+  async testUPSConnection(name) {
-    const testBtn = document.querySelector(`[data-ups-name="${upsName}"].test-btn`);
-    const originalText = testBtn.textContent;
-    testBtn.textContent = 'Testing...';
-    testBtn.disabled = true;
-
     try {
-      const response = await fetch(`/api/config/ups/${upsName}/test`, {
+      const resp = await window.apiFetch('/api/config/ups/' + name + '/test', { method: 'POST' });
-        method: 'POST'
+      const r = await resp.json();
-      });
+      this.showToast(r.message || (r.success ? 'OK' : 'Failed'),
-
+        r.success ? 'success' : 'error');
-      const result = await response.json();
+    } catch (err) { this.showToast('Test failed: ' + err.message, 'error'); }
-      let msg;
-      if (result.success) {
-        msg = `Connection to ${upsName} successful`;
-        if (result.data && result.data.STATUS) {
-          msg += ` (STATUS=${result.data.STATUS})`;
-        }
-        this.showToast(msg, 'success');
-      } else {
-        if (result.connectivity && !result.connectivity.ok) {
-          msg = `TCP connectivity failed: ${result.connectivity.error}`;
-        } else if (result.protocol && !result.protocol.ok) {
-          msg = `Protocol error: ${result.protocol.error}`;
-        } else {
-          msg = result.message || 'Connection failed';
-        }
-        this.showToast(`Connection to ${upsName} failed: ${msg}`, 'error');
-      }
-    } catch (error) {
-      this.showToast(`Connection test failed: ${error.message}`, 'error');
-    } finally {
-      testBtn.textContent = originalText;
-      testBtn.disabled = false;
-    }
   }
 
   showToast(message, type = 'info') {
     const container = document.getElementById('toast-container');
-    const toast = document.createElement('div');
+    const toast = el('div', { class: 'toast ' + type }, message);
-    toast.className = `toast ${type}`;
-    toast.textContent = message;
-    
     container.appendChild(toast);
-    
+    setTimeout(() => { if (toast.parentNode) toast.parentNode.removeChild(toast); }, 5000);
-    // Auto-remove after 5 seconds
-    setTimeout(() => {
-      if (toast.parentNode) {
-        toast.parentNode.removeChild(toast);
-      }
-    }, 5000);
   }
 }
 
-// Initialize when DOM is loaded
+document.addEventListener('DOMContentLoaded', () => { new ConfigManager(); });
-document.addEventListener('DOMContentLoaded', () => {
-  new ConfigManager();
-});

app/static/js/events.js

@@ -0,0 +1,56 @@
+// Events log page (DOM-safe)
+(async function () {
+  function setCell(row, text) {
+    const td = document.createElement('td');
+    td.textContent = text;
+    row.appendChild(td);
+  }
+
+  async function loadUps() {
+    const r = await window.apiFetch('/api/ups');
+    if (!r.ok) return;
+    const list = await r.json();
+    const sel = document.getElementById('filter-ups');
+    list.forEach(u => {
+      const o = document.createElement('option');
+      o.value = u.name; o.textContent = u.name;
+      sel.appendChild(o);
+    });
+  }
+
+  async function loadEvents() {
+    const ups = document.getElementById('filter-ups').value;
+    const kind = document.getElementById('filter-kind').value;
+    const params = new URLSearchParams();
+    if (ups) params.set('ups', ups);
+    if (kind) params.set('kind', kind);
+    params.set('limit', '300');
+    const r = await window.apiFetch('/api/events?' + params.toString());
+    const tbody = document.getElementById('events-tbody');
+    tbody.textContent = '';
+    if (!r.ok) {
+      const tr = document.createElement('tr');
+      setCell(tr, 'Failed to load'); tbody.appendChild(tr); return;
+    }
+    const rows = await r.json();
+    if (!rows.length) {
+      const tr = document.createElement('tr');
+      setCell(tr, 'No events'); tbody.appendChild(tr); return;
+    }
+    rows.forEach(e => {
+      const tr = document.createElement('tr');
+      setCell(tr, new Date(e.ts * 1000).toLocaleString());
+      setCell(tr, e.ups);
+      setCell(tr, e.type);
+      setCell(tr, e.detail);
+      tbody.appendChild(tr);
+    });
+  }
+
+  document.getElementById('refresh-btn').addEventListener('click', loadEvents);
+  document.getElementById('filter-ups').addEventListener('change', loadEvents);
+  document.getElementById('filter-kind').addEventListener('change', loadEvents);
+
+  await loadUps();
+  await loadEvents();
+})();

app/static/js/settings.js

@@ -0,0 +1,79 @@
+// Settings page
+(async function () {
+  async function loadUi() {
+    const r = await window.apiFetch('/api/config/ui');
+    if (!r.ok) return;
+    const ui = await r.json();
+    const form = document.getElementById('ui-form');
+    Object.entries(ui).forEach(([k, v]) => {
+      const el = form.elements[k];
+      if (!el) return;
+      if (el.type === 'checkbox') el.checked = !!v;
+      else el.value = v;
+    });
+  }
+
+  async function loadSmtp() {
+    const r = await window.apiFetch('/api/config/smtp');
+    if (!r.ok) return;
+    const s = await r.json();
+    if (!s) return;
+    const form = document.getElementById('smtp-form');
+    Object.entries(s).forEach(([k, v]) => {
+      const el = form.elements[k];
+      if (!el) return;
+      if (k === 'to_addrs' && Array.isArray(v)) el.value = v.join(', ');
+      else if (el.type === 'checkbox') el.checked = !!v;
+      else if (v !== null && v !== undefined) el.value = v;
+    });
+  }
+
+  document.getElementById('ui-form').addEventListener('submit', async (e) => {
+    e.preventDefault();
+    const f = e.target;
+    const payload = {
+      show_events: f.show_events.checked,
+      show_energy: f.show_energy.checked,
+      color_badges: f.color_badges.checked,
+      enable_transfer_burst_alert: f.enable_transfer_burst_alert.checked,
+      enable_voltage_deviation_alert: f.enable_voltage_deviation_alert.checked,
+      energy_cost_per_kwh: parseFloat(f.energy_cost_per_kwh.value) || 0,
+    };
+    const r = await window.apiFetch('/api/config/ui', {
+      method: 'PUT', body: JSON.stringify(payload),
+    });
+    alert(r.ok ? 'UI settings saved' : 'Save failed');
+  });
+
+  document.getElementById('smtp-form').addEventListener('submit', async (e) => {
+    e.preventDefault();
+    const f = e.target;
+    const to = f.to_addrs.value.split(',').map(s => s.trim()).filter(Boolean);
+    const payload = {
+      host: f.host.value,
+      port: parseInt(f.port.value, 10),
+      username: f.username.value || null,
+      use_tls: f.use_tls.checked,
+      use_ssl: f.use_ssl.checked,
+      from_addr: f.from_addr.value || null,
+      to_addrs: to,
+      subject_prefix: f.subject_prefix.value || '[UPS]',
+      silent_hours_start: f.silent_hours_start.value ? parseInt(f.silent_hours_start.value, 10) : null,
+      silent_hours_end: f.silent_hours_end.value ? parseInt(f.silent_hours_end.value, 10) : null,
+      daily_summary_hour: f.daily_summary_hour.value ? parseInt(f.daily_summary_hour.value, 10) : null,
+    };
+    const r = await window.apiFetch('/api/config/smtp', {
+      method: 'PUT', body: JSON.stringify(payload),
+    });
+    alert(r.ok ? 'SMTP saved' : 'Save failed');
+  });
+
+  document.getElementById('test-smtp-btn').addEventListener('click', async () => {
+    const r = await window.apiFetch('/api/config/smtp/test', { method: 'POST' });
+    const body = await r.json().catch(() => ({}));
+    alert(r.ok ? 'Test email sent' : 'Failed: ' + (body.detail || 'unknown'));
+  });
+
+  await loadUi();
+  await loadSmtp();
+})();

app/storage.py

@@ -1,51 +1,62 @@
 from __future__ import annotations
-import asyncio
-import time
-from typing import Dict, Any, List
-import redis
-import json
-import os
 
-REDIS_URL = os.environ.get("REDIS_URL", "redis://redis:6379/0")
+import json
+import time
+from typing import Any
+
+import redis
+
+from .settings import settings
+
 RETENTION_SECONDS = 7 * 24 * 3600
-MAX_SAMPLES_PER_UPS = 7 * 24 * 60 * 2  # assume worst-case 30s interval -> ~20160 entries
+MAX_SAMPLES_PER_UPS = 7 * 24 * 60 * 2  # worst-case ~30s interval
 
 _redis: redis.Redis | None = None
 
+
 def get_redis() -> redis.Redis:
     global _redis
     if _redis:
         return _redis
-    _redis = redis.Redis.from_url(REDIS_URL, decode_responses=True)
+    _redis = redis.Redis.from_url(settings.redis_url, decode_responses=True)
     return _redis
 
-SNAP_KEY_PREFIX = "ups:snap:"  # latest hash per ups
-HIST_KEY_PREFIX = "ups:hist:"  # time-series list per ups (append JSON)
 
-async def store_snapshot(ups_name: str, data: Dict[str, Any]):
+def reset_redis_client() -> None:
+    """Test hook - clears cached Redis connection."""
+    global _redis
+    _redis = None
+
+
+SNAP_KEY_PREFIX = "ups:snap:"
+HIST_KEY_PREFIX = "ups:hist:"
+
+
+async def store_snapshot(ups_name: str, data: dict[str, Any]):
     r = get_redis()
     ts = int(time.time())
     pipe = r.pipeline()
-    # store latest snapshot (hash)
     pipe.hset(f"{SNAP_KEY_PREFIX}{ups_name}", mapping={**data, "_ts": ts})
-    # append to history list
     hist_key = f"{HIST_KEY_PREFIX}{ups_name}"
     pipe.rpush(hist_key, json.dumps({"ts": ts, "data": data}))
     pipe.ltrim(hist_key, -MAX_SAMPLES_PER_UPS, -1)
-    # add pruning via async task (length-based + time-based)
     pipe.execute()
 
-async def get_latest(ups_name: str) -> Dict[str, Any] | None:
+
+async def get_latest(ups_name: str) -> dict[str, Any] | None:
     r = get_redis()
     h = r.hgetall(f"{SNAP_KEY_PREFIX}{ups_name}")
     return h or None
 
-async def get_history(ups_name: str, since_seconds: int = RETENTION_SECONDS) -> List[Dict[str, Any]]:
+
+async def get_history(
+    ups_name: str, since_seconds: int = RETENTION_SECONDS
+) -> list[dict[str, Any]]:
     r = get_redis()
     key = f"{HIST_KEY_PREFIX}{ups_name}"
     raw = r.lrange(key, 0, -1)
     now = int(time.time())
-    out: List[Dict[str, Any]] = []
+    out: list[dict[str, Any]] = []
     for item in raw:
         try:
             obj = json.loads(item)
@@ -55,12 +66,12 @@ async def get_history(ups_name: str, since_seconds: int = RETENTION_SECONDS) ->
             out.append(obj)
     return out
 
+
 async def prune_old():
     r = get_redis()
     now = int(time.time())
     cutoff = now - RETENTION_SECONDS
     for key in r.scan_iter(f"{HIST_KEY_PREFIX}*"):
-        # prune from left while older than cutoff
         while True:
             item = r.lindex(key, 0)
             if not item:

app/templates/alerts.html

@@ -0,0 +1,22 @@
+{% extends "base.html" %}
+{% block title %}Alerts — UPS{% endblock %}
+{% block content %}
+<div class="page-header">
+  <h1>Alerts</h1>
+  <div class="filters">
+    <select id="filter-severity"><option value="">All Severities</option><option value="CRITICAL">CRITICAL</option><option value="WARNING">WARNING</option><option value="INFO">INFO</option></select>
+    <select id="filter-ups"><option value="">All UPS</option></select>
+    <select id="filter-view"><option value="active">Active</option><option value="history">History (30d)</option></select>
+    <button id="refresh-btn" class="btn btn-secondary">Refresh</button>
+  </div>
+</div>
+<div class="main-content">
+  <table class="data-table">
+    <thead><tr><th>Time</th><th>Severity</th><th>UPS</th><th>Code</th><th>Message</th><th>Acked</th><th></th></tr></thead>
+    <tbody id="alerts-tbody"><tr><td colspan="7">Loading...</td></tr></tbody>
+  </table>
+</div>
+{% endblock %}
+{% block scripts %}
+<script src="/static/js/alerts.js"></script>
+{% endblock %}

app/templates/base.html

@@ -0,0 +1,35 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8" />
+  <meta name="viewport" content="width=device-width, initial-scale=1" />
+  <title>{% block title %}UPS Dashboard{% endblock %}</title>
+  <meta name="csrf-token" content="{{ request.cookies.get('csrf_token', '') }}" />
+  <link rel="stylesheet" href="/static/css/base.css" />
+  <link rel="stylesheet" href="/static/css/style.css" />
+  {% block head_extra %}{% endblock %}
+</head>
+<body>
+  <header class="site-header">
+    <div class="brand"><a href="/">⚡ APC UPS Dashboard</a></div>
+    <nav class="main-nav">
+      <a href="/" class="{% if active_nav == 'dashboard' %}active{% endif %}">Dashboard</a>
+      <a href="/events" class="{% if active_nav == 'events' %}active{% endif %}">Events</a>
+      <a href="/alerts" class="{% if active_nav == 'alerts' %}active{% endif %}">Alerts</a>
+      <a href="/config" class="{% if active_nav == 'config' %}active{% endif %}">UPS Config</a>
+      <a href="/settings" class="{% if active_nav == 'settings' %}active{% endif %}">Settings</a>
+    </nav>
+    <div class="user-menu">
+      {% if current_user %}
+        <span class="username">{{ current_user }}</span>
+        <button type="button" id="logout-btn" class="btn-logout">Logout</button>
+      {% endif %}
+    </div>
+  </header>
+  <main class="site-main">
+    {% block content %}{% endblock %}
+  </main>
+  <script src="/static/js/app.js"></script>
+  {% block scripts %}{% endblock %}
+</body>
+</html>

app/templates/config.html

@@ -1,27 +1,18 @@
-<!DOCTYPE html>
+{% extends "base.html" %}
-<html lang="en">
+{% block title %}UPS Configuration{% endblock %}
-<head>
+{% block head_extra %}
-  <meta charset="UTF-8" />
-  <title>UPS Configuration</title>
-  <link rel="stylesheet" href="/static/css/style.css" />
 <link rel="stylesheet" href="/static/css/config.css" />
-</head>
+{% endblock %}
-<body>
+{% block content %}
-  <div class="header">
+<div class="page-header">
   <h1>UPS Configuration</h1>
-    <nav>
-      <a href="/">← Back to Dashboard</a>
   <button id="add-ups-btn" class="btn btn-primary">Add UPS</button>
-    </nav>
 </div>
 
 <div class="main-content">
-    <div id="ups-list" class="ups-list">
+  <div id="ups-list" class="ups-list"></div>
-      <!-- UPS configurations will be loaded here -->
-    </div>
 </div>
 
-  <!-- Add/Edit UPS Modal -->
 <div id="ups-modal" class="modal">
   <div class="modal-content">
     <div class="modal-header">
@@ -31,55 +22,42 @@
     <form id="ups-form">
       <div class="form-group">
         <label for="ups-name">Name *</label>
-          <input type="text" id="ups-name" name="name" required>
+        <input type="text" id="ups-name" name="name" required pattern="[A-Za-z0-9_-]{1,32}" maxlength="32">
-          <small>Friendly name for the UPS</small>
+        <small>1-32 alphanumeric, underscore, dash</small>
       </div>
-        
       <div class="form-group">
         <label for="ups-host">Host/IP *</label>
         <input type="text" id="ups-host" name="host" required>
-          <small>IP address or hostname of the apcupsd server</small>
+        <small>IP or hostname of the apcupsd server</small>
       </div>
-        
       <div class="form-group">
         <label for="ups-port">Port</label>
-          <input type="number" id="ups-port" name="port" value="3551">
+        <input type="number" id="ups-port" name="port" value="3551" min="1" max="65535">
-          <small>apcupsd NIS port (default: 3551)</small>
       </div>
-        
       <div class="form-group">
         <label for="ups-interval">Polling Interval (seconds)</label>
-          <input type="number" id="ups-interval" name="interval_seconds" value="30" min="1">
+        <input type="number" id="ups-interval" name="interval_seconds" value="30" min="5" max="3600">
-          <small>How often to poll the UPS for data</small>
       </div>
-        
       <h3>Alert Thresholds</h3>
-        
       <div class="form-group">
         <label for="ups-loadpct">High Load Alert (%)</label>
         <input type="number" id="ups-loadpct" name="alert_loadpct_high" min="0" max="100" step="0.1">
-          <small>Trigger alert when load percentage reaches this value</small>
       </div>
-        
       <div class="form-group">
         <label for="ups-bcharge">Low Battery Alert (%)</label>
         <input type="number" id="ups-bcharge" name="alert_bcharge_low" min="0" max="100" step="0.1">
-          <small>Trigger alert when battery charge drops to this value</small>
       </div>
-        
       <div class="form-group">
-          <label>
+        <label><input type="checkbox" id="ups-onbattery" name="alert_on_battery"> Alert when on battery power</label>
-            <input type="checkbox" id="ups-onbattery" name="alert_on_battery">
-            Alert when on battery power
-          </label>
       </div>
-        
       <div class="form-group">
         <label for="ups-runtime">Low Runtime Alert (minutes)</label>
         <input type="number" id="ups-runtime" name="alert_runtime_low_minutes" min="0" step="0.1">
-          <small>Trigger alert when remaining runtime drops to this value</small>
       </div>
-        
+      <div class="form-group">
+        <label for="ups-itemp">High Temperature Alert (°C)</label>
+        <input type="number" id="ups-itemp" name="alert_itemp_high" min="0" max="120" step="0.5">
+      </div>
       <div class="form-actions">
         <button type="button" id="test-connection-btn" class="btn btn-secondary">Test Connection</button>
         <button type="submit" class="btn btn-primary">Save</button>
@@ -89,7 +67,6 @@
   </div>
 </div>
 
-  <!-- Confirmation Dialog -->
 <div id="confirm-modal" class="modal">
   <div class="modal-content small">
     <div class="modal-header">
@@ -104,9 +81,8 @@
   </div>
 </div>
 
-  <!-- Toast notifications -->
 <div id="toast-container"></div>
-
+{% endblock %}
+{% block scripts %}
 <script src="/static/js/config.js"></script>
-</body>
+{% endblock %}
-</html>

app/templates/dashboard.html

@@ -1,25 +1,32 @@
-<!DOCTYPE html>
+{% extends "base.html" %}
-<html lang="en">
+{% block title %}Dashboard — UPS{% endblock %}
-<head>
+{% block head_extra %}
-  <meta charset="UTF-8" />
-  <title>UPS Dashboard</title>
-  <link rel="stylesheet" href="/static/css/style.css" />
 <script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
-</head>
+{% endblock %}
-<body>
+{% block content %}
-  <div class="header">
+<section class="fleet-overview" id="fleet-overview">
-    <h1>APC UPS Dashboard</h1>
+  <div class="fleet-card">
-    <nav>
+    <div class="fleet-stat">
-      <a href="/config" class="config-link">⚙️ Configuration</a>
+      <span class="fleet-stat-label">Total UPS</span>
-    </nav>
+      <span class="fleet-stat-value" id="fleet-total">-</span>
     </div>
+    <div class="fleet-stat"><span class="fleet-stat-label">Online</span><span class="fleet-stat-value online" id="fleet-online">-</span></div>
+    <div class="fleet-stat"><span class="fleet-stat-label">On Battery</span><span class="fleet-stat-value onbatt" id="fleet-onbatt">-</span></div>
+    <div class="fleet-stat"><span class="fleet-stat-label">Offline</span><span class="fleet-stat-value offline" id="fleet-offline">-</span></div>
+    <div class="fleet-stat"><span class="fleet-stat-label">Total Load (W)</span><span class="fleet-stat-value" id="fleet-watts">-</span></div>
+    <div class="fleet-stat"><span class="fleet-stat-label">Min Runtime</span><span class="fleet-stat-value" id="fleet-min-runtime">-</span></div>
+  </div>
+</section>
+
 <div id="ups-cards" data-ui='{{ ui_cfg | tojson }}'>
   {% for ups in ups_list %}
     <div class="card" id="card-{{ ups.name }}">
       <div class="card-header">
-          <h2>{{ ups.name }}</h2>
+        <h2>
+          <span class="state-dot" data-state-dot></span>
+          {{ ups.name }}
+        </h2>
         <div class="ups-connection" data-field-conn>{{ ups.host }}:{{ ups.port }}</div>
-          <button class="add-metric-btn" data-add-metric title="Add metric tile">+</button>
         <button class="reset-layout-btn" data-reset-layout title="Reset tile layout">⟳</button>
       </div>
       <div class="metrics">
@@ -41,6 +48,7 @@
     </div>
   {% endif %}
 </div>
-<script src="/static/js/dashboard.js?v=1"></script>
+{% endblock %}
-</body>
+{% block scripts %}
-</html>
+<script src="/static/js/dashboard.js?v=2"></script>
+{% endblock %}

app/templates/emails/alert.html

@@ -0,0 +1,31 @@
+<!DOCTYPE html>
+<html>
+<head><meta charset="UTF-8"></head>
+<body style="font-family: -apple-system, Segoe UI, Arial, sans-serif; margin: 0; padding: 0; background: #f4f4f4;">
+<table role="presentation" width="100%" cellpadding="0" cellspacing="0" style="background: #f4f4f4; padding: 24px 0;">
+  <tr>
+    <td align="center">
+      <table role="presentation" width="560" cellpadding="0" cellspacing="0" style="background: #ffffff; border-radius: 6px; overflow: hidden; box-shadow: 0 2px 6px rgba(0,0,0,.08);">
+        <tr>
+          <td style="background: {{ color }}; color: #ffffff; padding: 16px 20px; font-size: 18px; font-weight: 600;">
+            [{{ top_severity }}] {{ ups_name }}
+          </td>
+        </tr>
+        <tr>
+          <td style="padding: 20px;">
+            <p style="margin: 0 0 12px; color: #333;">The following alerts were triggered:</p>
+            <ul style="margin: 0 0 16px; padding-left: 20px; color: #333;">
+              {% for a in alerts %}
+              <li style="margin-bottom: 6px;"><strong>[{{ a.severity }}]</strong> {{ a.message }}</li>
+              {% endfor %}
+            </ul>
+            <p style="margin: 0 0 8px;"><a href="{{ dashboard_url }}" style="color: #2980b9;">Open dashboard</a></p>
+            <p style="margin: 0; color: #888; font-size: 12px;">Sent at {{ sent_at }}</p>
+          </td>
+        </tr>
+      </table>
+    </td>
+  </tr>
+</table>
+</body>
+</html>

app/templates/emails/summary.html

@@ -0,0 +1,21 @@
+<!DOCTYPE html>
+<html>
+<head><meta charset="UTF-8"></head>
+<body style="font-family: -apple-system, Segoe UI, Arial, sans-serif;">
+<h2>UPS Daily Summary</h2>
+<p>Sent at: {{ sent_at }}</p>
+<table cellpadding="6" cellspacing="0" border="1" style="border-collapse: collapse;">
+  <tr><th>UPS</th><th>Status</th><th>Load %</th><th>Battery %</th><th>kWh today</th><th>Alerts (24h)</th></tr>
+  {% for row in summary.ups_rows %}
+  <tr>
+    <td>{{ row.name }}</td>
+    <td>{{ row.status }}</td>
+    <td>{{ row.loadpct }}</td>
+    <td>{{ row.bcharge }}</td>
+    <td>{{ row.kwh }}</td>
+    <td>{{ row.alerts_count }}</td>
+  </tr>
+  {% endfor %}
+</table>
+</body>
+</html>

app/templates/events.html

@@ -0,0 +1,21 @@
+{% extends "base.html" %}
+{% block title %}Events — UPS{% endblock %}
+{% block content %}
+<div class="page-header">
+  <h1>Event Log</h1>
+  <div class="filters">
+    <select id="filter-ups"><option value="">All UPS</option></select>
+    <select id="filter-kind"><option value="">All Events</option><option value="STATUS">STATUS</option><option value="XFER">XFER</option></select>
+    <button id="refresh-btn" class="btn btn-secondary">Refresh</button>
+  </div>
+</div>
+<div class="main-content">
+  <table class="data-table">
+    <thead><tr><th>Time</th><th>UPS</th><th>Type</th><th>Detail</th></tr></thead>
+    <tbody id="events-tbody"><tr><td colspan="4">Loading...</td></tr></tbody>
+  </table>
+</div>
+{% endblock %}
+{% block scripts %}
+<script src="/static/js/events.js"></script>
+{% endblock %}

app/templates/login.html

@@ -0,0 +1,49 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8" />
+  <meta name="viewport" content="width=device-width, initial-scale=1" />
+  <title>Login — UPS Dashboard</title>
+  <link rel="stylesheet" href="/static/css/base.css" />
+  <link rel="stylesheet" href="/static/css/login.css" />
+</head>
+<body class="auth-body">
+  <div class="auth-card">
+    <h1>UPS Dashboard</h1>
+    <form id="login-form">
+      <div class="form-field">
+        <label for="username">Username</label>
+        <input type="text" id="username" name="username" required autocomplete="username" />
+      </div>
+      <div class="form-field">
+        <label for="password">Password</label>
+        <input type="password" id="password" name="password" required autocomplete="current-password" />
+      </div>
+      <button type="submit" class="btn-primary btn-block">Sign in</button>
+      <div id="login-error" class="login-error" hidden></div>
+    </form>
+  </div>
+  <script>
+    document.getElementById('login-form').addEventListener('submit', async (e) => {
+      e.preventDefault();
+      const errBox = document.getElementById('login-error');
+      errBox.hidden = true;
+      const resp = await fetch('/api/login', {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify({
+          username: document.getElementById('username').value,
+          password: document.getElementById('password').value,
+        }),
+      });
+      if (resp.ok) {
+        window.location.href = '/';
+      } else {
+        const err = await resp.json().catch(() => ({}));
+        errBox.textContent = err.detail || 'Login failed';
+        errBox.hidden = false;
+      }
+    });
+  </script>
+</body>
+</html>

app/templates/settings.html

@@ -0,0 +1,44 @@
+{% extends "base.html" %}
+{% block title %}Settings — UPS{% endblock %}
+{% block content %}
+<div class="page-header">
+  <h1>Settings</h1>
+</div>
+<div class="main-content">
+  <section class="settings-section">
+    <h2>UI / Feature Flags</h2>
+    <form id="ui-form">
+      <label><input type="checkbox" name="show_events"> Show events tile</label>
+      <label><input type="checkbox" name="show_energy"> Show energy</label>
+      <label><input type="checkbox" name="color_badges"> Colorize badges</label>
+      <label><input type="checkbox" name="enable_transfer_burst_alert"> Transfer-burst alerts</label>
+      <label><input type="checkbox" name="enable_voltage_deviation_alert"> Voltage-deviation alerts</label>
+      <label>Energy cost per kWh ($): <input type="number" step="0.001" min="0" name="energy_cost_per_kwh"></label>
+      <button type="submit" class="btn btn-primary">Save UI Settings</button>
+    </form>
+  </section>
+
+  <section class="settings-section">
+    <h2>SMTP / Email Alerts</h2>
+    <form id="smtp-form">
+      <label>Host <input type="text" name="host" required></label>
+      <label>Port <input type="number" name="port" min="1" max="65535" required></label>
+      <label>Username <input type="text" name="username"></label>
+      <label><input type="checkbox" name="use_tls"> Use STARTTLS</label>
+      <label><input type="checkbox" name="use_ssl"> Use SSL</label>
+      <label>From <input type="email" name="from_addr"></label>
+      <label>To (comma-separated) <input type="text" name="to_addrs"></label>
+      <label>Subject Prefix <input type="text" name="subject_prefix" value="[UPS]"></label>
+      <label>Silent hours start (0-23, optional) <input type="number" name="silent_hours_start" min="0" max="23"></label>
+      <label>Silent hours end (0-23, optional) <input type="number" name="silent_hours_end" min="0" max="23"></label>
+      <label>Daily summary hour (0-23, optional) <input type="number" name="daily_summary_hour" min="0" max="23"></label>
+      <p class="muted">Password is read from the <code>SMTP_PASSWORD</code> environment variable and never stored in Redis.</p>
+      <button type="submit" class="btn btn-primary">Save SMTP</button>
+      <button type="button" id="test-smtp-btn" class="btn btn-secondary">Send Test Email</button>
+    </form>
+  </section>
+</div>
+{% endblock %}
+{% block scripts %}
+<script src="/static/js/settings.js"></script>
+{% endblock %}

app/templates/setup.html

@@ -0,0 +1,62 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8" />
+  <meta name="viewport" content="width=device-width, initial-scale=1" />
+  <title>Setup — UPS Dashboard</title>
+  <link rel="stylesheet" href="/static/css/base.css" />
+  <link rel="stylesheet" href="/static/css/login.css" />
+</head>
+<body class="auth-body">
+  <div class="auth-card wide">
+    <h1>First-run Setup</h1>
+    <p class="muted">Create the admin account. Passwords must be at least 8 characters.</p>
+    <form id="setup-form">
+      <div class="form-field">
+        <label for="username">Admin username</label>
+        <input type="text" id="username" name="username" required minlength="3" autocomplete="username" />
+      </div>
+      <div class="form-field">
+        <label for="password">Password</label>
+        <input type="password" id="password" name="password" required minlength="8" autocomplete="new-password" />
+      </div>
+      <div class="form-field">
+        <label for="password2">Confirm password</label>
+        <input type="password" id="password2" name="password2" required minlength="8" />
+      </div>
+      <button type="submit" class="btn-primary btn-block">Create admin account</button>
+      <div id="setup-error" class="login-error" hidden></div>
+    </form>
+    <p class="muted" style="margin-top:16px;">After setup you'll be redirected to the dashboard. Configure UPS devices and SMTP from the nav.</p>
+  </div>
+  <script>
+    document.getElementById('setup-form').addEventListener('submit', async (e) => {
+      e.preventDefault();
+      const errBox = document.getElementById('setup-error');
+      errBox.hidden = true;
+      const pw = document.getElementById('password').value;
+      const pw2 = document.getElementById('password2').value;
+      if (pw !== pw2) {
+        errBox.textContent = 'Passwords do not match';
+        errBox.hidden = false;
+        return;
+      }
+      const resp = await fetch('/api/setup', {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify({
+          username: document.getElementById('username').value,
+          password: pw,
+        }),
+      });
+      if (resp.ok) {
+        window.location.href = '/';
+      } else {
+        const err = await resp.json().catch(() => ({}));
+        errBox.textContent = err.detail || 'Setup failed';
+        errBox.hidden = false;
+      }
+    });
+  </script>
+</body>
+</html>

docker-compose.yml

@@ -5,18 +5,31 @@ services:
     container_name: apcups-dashboard
     ports:
       - "10280:8000"
+    env_file: .env
     environment:
       - REDIS_URL=redis://redis:6379/0
     depends_on:
-      - redis
+      redis:
+        condition: service_healthy
     restart: always
+    healthcheck:
+      test: ["CMD", "curl", "-fsS", "http://localhost:8000/healthz"]
+      interval: 30s
+      timeout: 5s
+      retries: 3
+      start_period: 15s
   redis:
     image: redis:7-alpine
     container_name: redis
     restart: always
-    # Enable persistence so UPS configuration stored in Redis survives restarts.
+    # Uncomment to require password (must also set REDIS_URL=redis://:PASS@redis:6379/0):
-    # Using both RDB snapshots (default schedule) and AOF for durability.
+    # command: ["redis-server", "--appendonly", "yes", "--appendfsync", "everysec", "--requirepass", "${REDIS_PASSWORD}"]
     command: ["redis-server", "--appendonly", "yes", "--appendfsync", "everysec"]
+    healthcheck:
+      test: ["CMD", "redis-cli", "ping"]
+      interval: 10s
+      timeout: 3s
+      retries: 5
     volumes:
       - redis-data:/data
 volumes:

pyproject.toml

@@ -1,2 +1,42 @@
+[project]
+name = "apcupsd-client"
+version = "0.2.0"
+description = "FastAPI + Redis dashboard for monitoring multiple APC UPS devices via apcupsd NIS"
+requires-python = ">=3.12"
+readme = "README.md"
+license = { text = "MIT" }
+
 [tool.pytest.ini_options]
 pythonpath = ["."]
+testpaths = ["tests"]
+asyncio_mode = "auto"
+asyncio_default_fixture_loop_scope = "function"
+filterwarnings = [
+  "ignore::DeprecationWarning",
+]
+
+[tool.ruff]
+line-length = 100
+target-version = "py312"
+extend-exclude = [".venv", "app/static"]
+
+[tool.ruff.lint]
+select = ["E", "F", "W", "I", "B", "UP", "SIM"]
+ignore = [
+  "E501",   # handled by formatter
+  "B008",   # FastAPI dependency defaults pattern
+  "B904",   # raise without from in except (noise in handlers)
+  "SIM105", # try/except/pass vs contextlib.suppress (not always clearer)
+]
+
+[tool.ruff.lint.per-file-ignores]
+"tests/*" = ["B011", "B017", "F401", "F811"]
+
+[tool.coverage.run]
+source = ["app"]
+omit = ["app/static/*", "app/templates/*"]
+
+[tool.coverage.report]
+show_missing = true
+skip_covered = false
+fail_under = 65

requirements.txt

@@ -3,7 +3,20 @@ uvicorn[standard]==0.30.6
 pyyaml==6.0.2
 redis==5.0.7
 pydantic==2.9.2
+pydantic-settings==2.5.2
 jinja2==3.1.4
 python-multipart==0.0.9
 orjson==3.10.7
+passlib[argon2]==1.7.4
+itsdangerous==2.2.0
+slowapi==0.1.9
+python-json-logger==2.0.7
+tenacity==9.0.0
+prometheus-client==0.21.0
+
+# Dev / test dependencies (kept in main file for simple docker-based dev)
 pytest==8.3.3
+pytest-asyncio==0.24.0
+httpx==0.27.2
+fakeredis==2.25.1
+ruff==0.6.9

tests/conftest.py

@@ -0,0 +1,112 @@
+"""Shared pytest fixtures: fakeredis + app overrides.
+
+Sets env vars before app modules import settings, swaps the Redis
+client in app.storage for a fakeredis instance, and mocks apcaccess.
+"""
+from __future__ import annotations
+
+import asyncio
+import os
+import sys
+from collections.abc import Iterator
+from unittest.mock import patch
+
+import pytest
+
+# Configure env BEFORE importing any app modules
+os.environ.setdefault("SESSION_SECRET", "test-session-secret-for-pytest-use-only-32bytes")
+os.environ.setdefault("ADMIN_USERNAME", "admin")
+os.environ.setdefault("ADMIN_PASSWORD_HASH", "")
+os.environ.setdefault("RATE_LIMIT_ENABLED", "false")
+os.environ.setdefault("ALLOW_PRIVATE_IPS", "true")
+os.environ.setdefault("REDIS_URL", "redis://localhost:6379/15")
+os.environ.setdefault("LOG_LEVEL", "WARNING")
+
+import fakeredis  # noqa: E402
+
+# Ensure fresh settings after env vars set
+from app import settings as settings_mod  # noqa: E402
+
+settings_mod.get_settings.cache_clear()
+settings_mod.settings = settings_mod.get_settings()
+
+from app import storage as storage_mod  # noqa: E402
+
+
+@pytest.fixture
+def fake_redis():
+    """Provide a clean fakeredis client installed into storage module."""
+    client = fakeredis.FakeRedis(decode_responses=True)
+    storage_mod._redis = client
+    yield client
+    client.flushall()
+    storage_mod._redis = None
+
+
+@pytest.fixture
+def clean_config_cache():
+    """Reset module-level config cache to force reload from (fake) Redis."""
+    from app import config as config_mod
+    config_mod._cached = None
+    yield
+    config_mod._cached = None
+
+
+@pytest.fixture
+def app_client(fake_redis, clean_config_cache):
+    """FastAPI TestClient with fakeredis wired in."""
+    from fastapi.testclient import TestClient
+
+    from app.main import create_app
+
+    app = create_app()
+    with TestClient(app) as client:
+        yield client
+
+
+@pytest.fixture
+def authed_client(app_client, fake_redis):
+    """Test client with admin set up and logged in."""
+    from app.auth import store_admin
+    store_admin("admin", "testpassword123")
+    r = app_client.post(
+        "/api/login",
+        json={"username": "admin", "password": "testpassword123"},
+    )
+    assert r.status_code == 200
+    csrf = r.json()["csrf_token"]
+    app_client.headers.update({"X-CSRF-Token": csrf})
+    return app_client
+
+
+@pytest.fixture
+def mock_apcaccess():
+    """Patch apc_cli.fetch_status to return a canned snapshot."""
+    from app import apc_cli
+
+    async def _fake_fetch(host, port):
+        return {
+            "STATUS": "ONLINE",
+            "LOADPCT": "25.0",
+            "NOMPOWER": "900",
+            "BCHARGE": "100.0",
+            "TIMELEFT": "45.0 Minutes",
+            "LINEV": "120.0",
+            "NOMINV": "120",
+            "ITEMP": "30.5",
+            "BATTV": "13.5",
+            "NOMBATTV": "12.0",
+            "SELFTEST": "NO",
+            "REPLACEBATT": "NO",
+            "LASTXFER": "No transfers since turnon",
+        }
+
+    with patch.object(apc_cli, "fetch_status", _fake_fetch):
+        yield _fake_fetch
+
+
+def pytest_collection_modifyitems(config, items):  # noqa: ARG001
+    """Mark all async tests as asyncio automatically."""
+    for item in items:
+        if asyncio.iscoroutinefunction(item.function):
+            item.add_marker(pytest.mark.asyncio)

tests/test_alerts.py

@@ -0,0 +1,150 @@
+"""Alert rules, cooldown, silent hours, deferred drain."""
+from __future__ import annotations
+
+from datetime import datetime
+from unittest.mock import patch
+
+import pytest
+
+from app.alerts import (
+    SEV_CRITICAL,
+    SEV_INFO,
+    SEV_WARNING,
+    Alert,
+    _defer_for_silent_hours,
+    _in_silent_hours,
+    drain_deferred,
+    emit_info,
+    evaluate_alerts,
+    process_alerts,
+)
+from app.config import SMTPConfig, UPSConfig
+
+
+def _ups(**overrides):
+    base = dict(
+        name="u1", host="192.168.1.10", port=3551, interval_seconds=30,
+        alert_on_battery=True,
+        alert_runtime_low_minutes=10.0,
+        alert_bcharge_low=20.0,
+        alert_loadpct_high=80.0,
+        alert_itemp_high=45.0,
+    )
+    base.update(overrides)
+    return UPSConfig(**base)
+
+
+def _snap(**overrides):
+    base = dict(
+        STATUS="ONLINE", LOADPCT="25.0", BCHARGE="100.0",
+        TIMELEFT="60.0 Minutes", ITEMP="30.0",
+        SELFTEST="NO", REPLACEBATT="NO",
+    )
+    base.update(overrides)
+    return base
+
+
+def test_no_alerts_normal_state(fake_redis, clean_config_cache):
+    alerts = evaluate_alerts(_ups(), _snap())
+    assert alerts == []
+
+
+def test_onbatt_critical(fake_redis, clean_config_cache):
+    alerts = evaluate_alerts(_ups(), _snap(STATUS="ONBATT"))
+    assert any(a.code == "ONBATT" and a.severity == SEV_CRITICAL for a in alerts)
+
+
+def test_runtime_low_critical(fake_redis, clean_config_cache):
+    alerts = evaluate_alerts(_ups(), _snap(TIMELEFT="5.0 Minutes"))
+    assert any(a.code == "RUNTIME_LOW" and a.severity == SEV_CRITICAL for a in alerts)
+
+
+def test_bcharge_low_critical(fake_redis, clean_config_cache):
+    alerts = evaluate_alerts(_ups(), _snap(BCHARGE="15.0"))
+    assert any(a.code == "BCHARGE_LOW" and a.severity == SEV_CRITICAL for a in alerts)
+
+
+def test_load_high_warning(fake_redis, clean_config_cache):
+    alerts = evaluate_alerts(_ups(), _snap(LOADPCT="90.0"))
+    assert any(a.code == "LOAD_HIGH" and a.severity == SEV_WARNING for a in alerts)
+
+
+def test_replacebatt_warning(fake_redis, clean_config_cache):
+    alerts = evaluate_alerts(_ups(), _snap(REPLACEBATT="YES"))
+    assert any(a.code == "REPLACEBATT" and a.severity == SEV_WARNING for a in alerts)
+
+
+def test_selftest_fail_warning(fake_redis, clean_config_cache):
+    alerts = evaluate_alerts(_ups(), _snap(SELFTEST="FAIL"))
+    assert any(a.code == "SELFTEST_FAIL" and a.severity == SEV_WARNING for a in alerts)
+
+
+def test_temp_high_warning(fake_redis, clean_config_cache):
+    alerts = evaluate_alerts(_ups(), _snap(ITEMP="55.0"))
+    assert any(a.code == "TEMP_HIGH" and a.severity == SEV_WARNING for a in alerts)
+
+
+def test_multiple_alerts_coalesced(fake_redis, clean_config_cache):
+    alerts = evaluate_alerts(
+        _ups(),
+        _snap(STATUS="ONBATT", BCHARGE="10.0", LOADPCT="95.0"),
+    )
+    codes = {a.code for a in alerts}
+    assert "ONBATT" in codes
+    assert "BCHARGE_LOW" in codes
+    assert "LOAD_HIGH" in codes
+
+
+def test_alert_hash_stable():
+    a1 = Alert(SEV_CRITICAL, "msg", "CODE", "u1", 100)
+    a2 = Alert(SEV_CRITICAL, "msg", "CODE", "u1", 999)
+    assert a1.hash() == a2.hash()  # hash doesn't depend on ts
+
+
+def test_silent_hours_same_day():
+    smtp = SMTPConfig(host="x", port=25, silent_hours_start=22, silent_hours_end=23)
+    now = datetime(2024, 1, 1, 22, 30)
+    assert _in_silent_hours(smtp, now) is True
+    now2 = datetime(2024, 1, 1, 23, 30)
+    assert _in_silent_hours(smtp, now2) is False
+
+
+def test_silent_hours_wrap_midnight():
+    smtp = SMTPConfig(host="x", port=25, silent_hours_start=22, silent_hours_end=7)
+    assert _in_silent_hours(smtp, datetime(2024, 1, 1, 23, 0)) is True
+    assert _in_silent_hours(smtp, datetime(2024, 1, 1, 3, 0)) is True
+    assert _in_silent_hours(smtp, datetime(2024, 1, 1, 10, 0)) is False
+
+
+def test_silent_hours_unset():
+    smtp = SMTPConfig(host="x", port=25)
+    assert _in_silent_hours(smtp) is False
+
+
+def test_defer_and_drain(fake_redis):
+    a = Alert(SEV_WARNING, "msg1", "CODE1", "u1", 100)
+    b = Alert(SEV_WARNING, "msg2", "CODE2", "u1", 101)
+    _defer_for_silent_hours([a, b])
+    drained = drain_deferred("u1")
+    assert len(drained) == 2
+    codes = {x.code for x in drained}
+    assert codes == {"CODE1", "CODE2"}
+    # Second drain is empty
+    assert drain_deferred("u1") == []
+
+
+def test_cooldown_suppresses_second_call(fake_redis, clean_config_cache):
+    # Avoid real SMTP: no smtp in config
+    with patch("app.alerts.dispatch_alerts"):
+        fresh = process_alerts(_ups(), _snap(STATUS="ONBATT"))
+        assert any(a.code == "ONBATT" for a in fresh)
+        # Immediate second call: cooled down
+        fresh2 = process_alerts(_ups(), _snap(STATUS="ONBATT"))
+        assert all(a.code != "ONBATT" for a in fresh2)
+
+
+def test_emit_info_writes_history(fake_redis, clean_config_cache):
+    with patch("app.alerts.dispatch_alerts"):
+        emit_info("u1", "REACHABLE", "u1 came back online")
+        entries = fake_redis.lrange("ups:alerts:history:all", 0, -1)
+        assert any("REACHABLE" in e for e in entries)

tests/test_api_routes.py

@@ -0,0 +1,252 @@
+"""Integration tests for UPS/alerts/events API routes."""
+from __future__ import annotations
+
+import json
+import time
+
+
+def _add_ups(client, name="u1", host="192.168.1.10"):
+    return client.post(
+        "/api/config/ups",
+        json={"name": name, "host": host, "port": 3551, "interval_seconds": 30},
+    )
+
+
+def test_ups_list_empty(authed_client):
+    r = authed_client.get("/api/ups")
+    assert r.status_code == 200
+    assert r.json() == []
+
+
+def test_ups_list_after_add(authed_client):
+    _add_ups(authed_client)
+    r = authed_client.get("/api/ups")
+    assert r.status_code == 200
+    body = r.json()
+    assert len(body) == 1
+    assert body[0]["name"] == "u1"
+
+
+def test_fleet_overview_empty(authed_client):
+    r = authed_client.get("/api/ups/fleet/overview")
+    assert r.status_code == 200
+    body = r.json()
+    assert body["total"] == 0
+    assert body["counts"]["online"] == 0
+
+
+def test_fleet_overview_with_data(authed_client, fake_redis):
+    _add_ups(authed_client, name="u1")
+    _add_ups(authed_client, name="u2", host="192.168.1.20")
+    fake_redis.hset("ups:snap:u1", mapping={"STATUS": "ONLINE", "DERIVED_WATTS": "200", "TIMELEFT": "40"})
+    fake_redis.hset("ups:snap:u2", mapping={"STATUS": "ONBATT", "DERIVED_WATTS": "150", "TIMELEFT": "10"})
+    r = authed_client.get("/api/ups/fleet/overview")
+    body = r.json()
+    assert body["total"] == 2
+    assert body["counts"]["online"] == 1
+    assert body["counts"]["on_battery"] == 1
+    assert body["total_watts"] == 350.0
+    assert body["min_timeleft_minutes"] == 10.0
+
+
+def test_fleet_overview_detects_offline(authed_client, fake_redis):
+    _add_ups(authed_client, name="u1")
+    fake_redis.set("ups:health:offline:u1", "1")
+    fake_redis.hset("ups:snap:u1", mapping={"STATUS": "ONLINE"})
+    r = authed_client.get("/api/ups/fleet/overview")
+    body = r.json()
+    assert body["counts"]["offline"] == 1
+
+
+def test_ups_status(authed_client, fake_redis):
+    _add_ups(authed_client)
+    fake_redis.hset("ups:snap:u1", mapping={"STATUS": "ONLINE", "_ts": "100"})
+    r = authed_client.get("/api/ups/u1")
+    assert r.status_code == 200
+    body = r.json()
+    assert body["STATUS"] == "ONLINE"
+
+
+def test_ups_history(authed_client, fake_redis):
+    _add_ups(authed_client)
+    now = int(time.time())
+    fake_redis.rpush(
+        "ups:hist:u1",
+        json.dumps({"ts": now, "data": {"STATUS": "ONLINE", "LOADPCT": "50"}}),
+    )
+    r = authed_client.get("/api/ups/u1/history")
+    body = r.json()
+    assert len(body) == 1
+    assert body[0]["data"]["STATUS"] == "ONLINE"
+
+
+def test_ups_metric(authed_client, fake_redis):
+    _add_ups(authed_client)
+    now = int(time.time())
+    for v in (10, 20, 30):
+        fake_redis.rpush(
+            "ups:hist:u1",
+            json.dumps({"ts": now, "data": {"LOADPCT": str(v)}}),
+        )
+    r = authed_client.get("/api/ups/u1/metric/LOADPCT")
+    body = r.json()
+    assert len(body) == 3
+    assert body[0]["value"] == 10.0
+
+
+def test_ups_events(authed_client, fake_redis):
+    _add_ups(authed_client)
+    fake_redis.lpush("ups:event:list:u1", f"{int(time.time())}|STATUS|ONLINE")
+    r = authed_client.get("/api/ups/u1/events")
+    body = r.json()
+    assert len(body) == 1
+    assert body[0]["type"] == "STATUS"
+
+
+def test_ups_energy_empty(authed_client):
+    _add_ups(authed_client)
+    r = authed_client.get("/api/ups/u1/energy")
+    assert r.status_code == 200
+    body = r.json()
+    assert body["kwh_today"] is None
+
+
+def test_ups_energy_with_data(authed_client, fake_redis):
+    _add_ups(authed_client)
+    day_str = time.strftime("%Y%m%d")
+    fake_redis.set(f"ups:energy:u1:{day_str}", "3600000")  # 1 kWh
+    r = authed_client.get("/api/ups/u1/energy")
+    body = r.json()
+    assert body["kwh_today"] == 1.0
+
+
+def test_ups_health(authed_client, fake_redis):
+    _add_ups(authed_client)
+    fake_redis.set("ups:health:last_ok:u1", "12345")
+    r = authed_client.get("/api/ups/u1/health")
+    body = r.json()
+    assert body["online"] is True
+    assert body["last_ok_ts"] == 12345
+
+
+def test_ups_health_offline(authed_client, fake_redis):
+    _add_ups(authed_client)
+    fake_redis.set("ups:health:offline:u1", "1")
+    fake_redis.set("ups:health:fail_count:u1", "3")
+    r = authed_client.get("/api/ups/u1/health")
+    body = r.json()
+    assert body["online"] is False
+    assert body["fail_count"] == 3
+
+
+def test_battery_health_empty(authed_client):
+    _add_ups(authed_client)
+    r = authed_client.get("/api/ups/u1/battery_health")
+    assert r.status_code == 200
+
+
+def test_events_list_empty(authed_client):
+    r = authed_client.get("/api/events")
+    assert r.status_code == 200
+    assert r.json() == []
+
+
+def test_events_list_all(authed_client, fake_redis):
+    _add_ups(authed_client, name="u1")
+    _add_ups(authed_client, name="u2", host="192.168.1.20")
+    now = int(time.time())
+    fake_redis.lpush("ups:event:list:u1", f"{now}|STATUS|ONLINE")
+    fake_redis.lpush("ups:event:list:u2", f"{now - 10}|XFER|Low voltage")
+    r = authed_client.get("/api/events")
+    body = r.json()
+    assert len(body) == 2
+    # Sorted by ts desc
+    assert body[0]["ts"] >= body[1]["ts"]
+
+
+def test_events_filter_by_ups(authed_client, fake_redis):
+    _add_ups(authed_client, name="u1")
+    _add_ups(authed_client, name="u2", host="192.168.1.20")
+    now = int(time.time())
+    fake_redis.lpush("ups:event:list:u1", f"{now}|STATUS|ONLINE")
+    fake_redis.lpush("ups:event:list:u2", f"{now}|STATUS|ONBATT")
+    r = authed_client.get("/api/events?ups=u1")
+    body = r.json()
+    assert all(e["ups"] == "u1" for e in body)
+
+
+def test_events_filter_by_kind(authed_client, fake_redis):
+    _add_ups(authed_client)
+    now = int(time.time())
+    fake_redis.lpush("ups:event:list:u1", f"{now}|STATUS|ONLINE")
+    fake_redis.lpush("ups:event:list:u1", f"{now}|XFER|Low voltage")
+    r = authed_client.get("/api/events?kind=XFER")
+    body = r.json()
+    assert all(e["type"] == "XFER" for e in body)
+
+
+def test_alerts_list_empty(authed_client):
+    r = authed_client.get("/api/alerts")
+    assert r.status_code == 200
+    assert r.json() == []
+
+
+def test_alerts_list_with_data(authed_client, fake_redis):
+    now = int(time.time())
+    fake_redis.lpush(
+        "ups:alerts:history:all",
+        f"{now}|CRITICAL|u1|ONBATT|On battery power|abc123",
+    )
+    r = authed_client.get("/api/alerts")
+    body = r.json()
+    assert len(body) == 1
+    assert body[0]["severity"] == "CRITICAL"
+    assert body[0]["ups"] == "u1"
+    assert body[0]["acked_ts"] is None
+
+
+def test_alerts_filter_by_severity(authed_client, fake_redis):
+    now = int(time.time())
+    fake_redis.lpush("ups:alerts:history:all", f"{now}|CRITICAL|u1|C|x|a")
+    fake_redis.lpush("ups:alerts:history:all", f"{now}|WARNING|u1|W|y|b")
+    r = authed_client.get("/api/alerts?severity=warning")
+    body = r.json()
+    assert all(a["severity"] == "WARNING" for a in body)
+
+
+def test_alerts_active_filters_acked(authed_client, fake_redis):
+    now = int(time.time())
+    fake_redis.lpush(
+        "ups:alerts:history:all",
+        f"{now}|CRITICAL|u1|A|msg|alert1",
+    )
+    fake_redis.lpush(
+        "ups:alerts:history:all",
+        f"{now}|CRITICAL|u1|B|msg2|alert2",
+    )
+    fake_redis.set("ups:alerts:ack:alert1", str(now))
+    r = authed_client.get("/api/alerts/active")
+    body = r.json()
+    assert all(a["id"] != "alert1" for a in body)
+
+
+def test_alerts_ack(authed_client, fake_redis):
+    r = authed_client.post("/api/alerts/myid/ack")
+    assert r.status_code == 200
+    assert fake_redis.get("ups:alerts:ack:myid") is not None
+
+
+def test_healthz_no_auth_required(app_client):
+    r = app_client.get("/healthz")
+    assert r.status_code == 200
+
+
+def test_readyz_reports_redis_status(app_client, fake_redis):
+    r = app_client.get("/readyz")
+    assert r.status_code in (200, 503)
+
+
+def test_metrics_endpoint(app_client, fake_redis):
+    r = app_client.get("/metrics")
+    assert r.status_code == 200
+    assert "ups_" in r.text or "# HELP" in r.text

tests/test_auth.py

@@ -0,0 +1,148 @@
+"""Auth: hash/verify, session token, login/logout, CSRF, setup."""
+from __future__ import annotations
+
+import pytest
+
+from app.auth import (
+    create_session_token,
+    get_stored_admin,
+    hash_password,
+    is_admin_configured,
+    store_admin,
+    verify_password,
+    verify_session_token,
+)
+
+
+def test_hash_and_verify_password(fake_redis):
+    h = hash_password("supersecret")
+    assert h.startswith("$argon2")
+    assert verify_password("supersecret", h) is True
+    assert verify_password("wrong", h) is False
+
+
+def test_verify_password_bad_hash_returns_false(fake_redis):
+    assert verify_password("anything", "not-a-valid-hash") is False
+
+
+def test_session_token_roundtrip(fake_redis):
+    tok = create_session_token("admin")
+    assert verify_session_token(tok) == "admin"
+
+
+def test_session_token_rejects_tampered(fake_redis):
+    tok = create_session_token("admin")
+    bad = tok[:-4] + "abcd"
+    assert verify_session_token(bad) is None
+
+
+def test_store_and_get_admin(fake_redis):
+    store_admin("alice", "pw-strong-1")
+    user, h = get_stored_admin()
+    assert user == "alice"
+    assert h and h.startswith("$argon2")
+    assert is_admin_configured() is True
+
+
+def test_login_rejects_before_setup(app_client, fake_redis):
+    r = app_client.post(
+        "/api/login",
+        json={"username": "admin", "password": "test"},
+    )
+    assert r.status_code == 409
+
+
+def test_login_happy_path(app_client, fake_redis):
+    store_admin("admin", "testpassword123")
+    r = app_client.post(
+        "/api/login",
+        json={"username": "admin", "password": "testpassword123"},
+    )
+    assert r.status_code == 200
+    body = r.json()
+    assert body["ok"] is True
+    assert "csrf_token" in body
+    # Cookies set
+    assert "ups_session" in r.cookies
+    assert "csrf_token" in r.cookies
+
+
+def test_login_bad_password(app_client, fake_redis):
+    store_admin("admin", "testpassword123")
+    r = app_client.post(
+        "/api/login",
+        json={"username": "admin", "password": "wrong"},
+    )
+    assert r.status_code == 401
+
+
+def test_login_bad_username(app_client, fake_redis):
+    store_admin("admin", "testpassword123")
+    r = app_client.post(
+        "/api/login",
+        json={"username": "notadmin", "password": "testpassword123"},
+    )
+    assert r.status_code == 401
+
+
+def test_logout_clears_cookies(authed_client):
+    r = authed_client.post("/api/logout")
+    assert r.status_code == 200
+
+
+def test_setup_creates_admin(app_client, fake_redis):
+    r = app_client.post(
+        "/api/setup",
+        json={"username": "newadmin", "password": "setup-pw-1234"},
+    )
+    assert r.status_code == 200
+    assert is_admin_configured() is True
+
+
+def test_setup_rejects_when_configured(app_client, fake_redis):
+    store_admin("admin", "already-set-password")
+    r = app_client.post(
+        "/api/setup",
+        json={"username": "newadmin", "password": "setup-pw-1234"},
+    )
+    assert r.status_code == 409
+
+
+def test_setup_rejects_short_password(app_client, fake_redis):
+    r = app_client.post(
+        "/api/setup",
+        json={"username": "admin", "password": "short"},
+    )
+    assert r.status_code == 422  # pydantic validation
+
+
+def test_setup_rejects_bad_username(app_client, fake_redis):
+    r = app_client.post(
+        "/api/setup",
+        json={"username": "bad user!", "password": "good-password-1"},
+    )
+    assert r.status_code == 400
+
+
+def test_protected_endpoint_requires_auth(app_client):
+    r = app_client.get("/api/config/ups")
+    assert r.status_code == 401
+
+
+def test_csrf_required_for_mutations(app_client, fake_redis):
+    store_admin("admin", "testpassword123")
+    r = app_client.post(
+        "/api/login",
+        json={"username": "admin", "password": "testpassword123"},
+    )
+    assert r.status_code == 200
+    # DELETE without CSRF header
+    r2 = app_client.delete("/api/config/ups/nonexistent")
+    assert r2.status_code == 403
+
+
+def test_mutations_work_with_csrf(authed_client):
+    # Authed client already has CSRF header set in fixture
+    r = authed_client.delete("/api/config/ups/nonexistent")
+    # Expected 404 because UPS doesn't exist (NOT 401/403)
+    assert r.status_code == 404

tests/test_config_api.py

@@ -0,0 +1,135 @@
+"""Config CRUD API: auth, CSRF, create/list/update/delete."""
+from __future__ import annotations
+
+
+def test_unauthed_cannot_list_ups(app_client):
+    r = app_client.get("/api/config/ups")
+    assert r.status_code == 401
+
+
+def test_unauthed_cannot_add_ups(app_client):
+    r = app_client.post(
+        "/api/config/ups",
+        json={"name": "u1", "host": "192.168.1.10"},
+    )
+    assert r.status_code == 401
+
+
+def test_add_and_list_ups(authed_client):
+    r = authed_client.post(
+        "/api/config/ups",
+        json={"name": "u1", "host": "192.168.1.10", "port": 3551, "interval_seconds": 30},
+    )
+    assert r.status_code == 200
+    r2 = authed_client.get("/api/config/ups")
+    assert r2.status_code == 200
+    body = r2.json()
+    assert len(body) == 1
+    assert body[0]["name"] == "u1"
+    assert body[0]["host"] == "192.168.1.10"
+
+
+def test_add_duplicate_rejects(authed_client):
+    authed_client.post(
+        "/api/config/ups",
+        json={"name": "u1", "host": "192.168.1.10"},
+    )
+    r = authed_client.post(
+        "/api/config/ups",
+        json={"name": "u1", "host": "192.168.1.11"},
+    )
+    assert r.status_code == 400
+
+
+def test_add_ssrf_host_rejected(authed_client):
+    r = authed_client.post(
+        "/api/config/ups",
+        json={"name": "bad", "host": "127.0.0.1"},
+    )
+    assert r.status_code == 422
+
+
+def test_update_ups(authed_client):
+    authed_client.post(
+        "/api/config/ups",
+        json={"name": "u1", "host": "192.168.1.10"},
+    )
+    r = authed_client.put(
+        "/api/config/ups/u1",
+        json={"host": "192.168.2.20", "interval_seconds": 60},
+    )
+    assert r.status_code == 200
+    r2 = authed_client.get("/api/config/ups/u1")
+    body = r2.json()
+    assert body["host"] == "192.168.2.20"
+    assert body["interval_seconds"] == 60
+
+
+def test_update_nonexistent(authed_client):
+    r = authed_client.put(
+        "/api/config/ups/missing",
+        json={"host": "192.168.2.20"},
+    )
+    assert r.status_code == 404
+
+
+def test_delete_ups(authed_client):
+    authed_client.post(
+        "/api/config/ups",
+        json={"name": "u1", "host": "192.168.1.10"},
+    )
+    r = authed_client.delete("/api/config/ups/u1")
+    assert r.status_code == 200
+    r2 = authed_client.get("/api/config/ups")
+    assert r2.json() == []
+
+
+def test_delete_nonexistent(authed_client):
+    r = authed_client.delete("/api/config/ups/missing")
+    assert r.status_code == 404
+
+
+def test_get_smtp_none(authed_client):
+    r = authed_client.get("/api/config/smtp")
+    assert r.status_code == 200
+    assert r.json() is None
+
+
+def test_smtp_password_always_redacted(authed_client, monkeypatch):
+    from app import settings as settings_mod
+    monkeypatch.setattr(settings_mod.settings, "smtp_password", "supersecret")
+    r = authed_client.put(
+        "/api/config/smtp",
+        json={
+            "host": "smtp.example.com",
+            "port": 587,
+            "username": "user@example.com",
+            "use_tls": True,
+            "from_addr": "alerts@example.com",
+            "to_addrs": ["ops@example.com"],
+        },
+    )
+    assert r.status_code == 200
+    r2 = authed_client.get("/api/config/smtp")
+    body = r2.json()
+    # Never raw password
+    assert body.get("password") in ("***", None)
+    assert body["host"] == "smtp.example.com"
+
+
+def test_ui_config_default(authed_client):
+    r = authed_client.get("/api/config/ui")
+    assert r.status_code == 200
+    body = r.json()
+    assert "show_events" in body
+
+
+def test_ui_config_update(authed_client):
+    r = authed_client.put(
+        "/api/config/ui",
+        json={"show_energy": True, "energy_cost_per_kwh": 0.15},
+    )
+    assert r.status_code == 200
+    body = r.json()
+    assert body["ui"]["show_energy"] is True
+    assert body["ui"]["energy_cost_per_kwh"] == 0.15

tests/test_exports.py

@@ -0,0 +1,74 @@
+"""CSV exports shape and content."""
+from __future__ import annotations
+
+import json
+import time
+
+import pytest
+
+from app.exports import export_energy_csv, export_events_csv, export_history_csv
+
+
+async def _collect(gen):
+    return [chunk async for chunk in gen]
+
+
+async def test_history_export_header_only_when_empty(fake_redis):
+    rows = await _collect(export_history_csv("u1", since_days=1))
+    assert len(rows) == 1
+    assert rows[0].startswith("ts,status,loadpct,bcharge,timeleft,linev,derived_watts")
+
+
+async def test_history_export_with_data(fake_redis):
+    now = int(time.time())
+    entry = {"ts": now, "data": {
+        "STATUS": "ONLINE", "LOADPCT": "25.0",
+        "BCHARGE": "100", "TIMELEFT": "45",
+        "LINEV": "120", "DERIVED_WATTS": "225",
+    }}
+    fake_redis.rpush("ups:hist:u1", json.dumps(entry))
+    rows = await _collect(export_history_csv("u1", since_days=1))
+    assert len(rows) == 2  # header + 1
+    data_row = rows[1]
+    assert "ONLINE" in data_row
+    assert "225" in data_row
+
+
+async def test_events_export_header_only_empty(fake_redis):
+    rows = await _collect(export_events_csv("u1", since_days=1))
+    assert len(rows) == 1
+    assert rows[0].startswith("ts,type,detail")
+
+
+async def test_events_export_with_data(fake_redis):
+    now = int(time.time())
+    fake_redis.rpush("ups:event:list:u1", f"{now}|STATUS|ONBATT")
+    fake_redis.rpush("ups:event:list:u1", f"{now}|XFER|Low voltage")
+    rows = await _collect(export_events_csv("u1", since_days=1))
+    assert len(rows) == 3
+    body = "\n".join(rows)
+    assert "STATUS" in body
+    assert "ONBATT" in body
+    assert "XFER" in body
+
+
+async def test_events_export_applies_cutoff(fake_redis):
+    # Event older than 1 day should be filtered out
+    old_ts = int(time.time()) - (2 * 24 * 3600)
+    fake_redis.rpush("ups:event:list:u1", f"{old_ts}|STATUS|OLD")
+    rows = await _collect(export_events_csv("u1", since_days=1))
+    assert len(rows) == 1  # header only
+
+
+async def test_energy_export_header(fake_redis):
+    rows = await _collect(export_energy_csv("u1", since_days=3))
+    assert rows[0].startswith("date,kwh")
+
+
+async def test_energy_export_with_data(fake_redis):
+    day_str = time.strftime("%Y%m%d")
+    fake_redis.set(f"ups:energy:u1:{day_str}", "3600000")  # 3,600,000 Ws = 1 kWh
+    rows = await _collect(export_energy_csv("u1", since_days=1))
+    # header + today's data
+    assert len(rows) >= 2
+    assert "1.0000" in "\n".join(rows)

tests/test_poller.py

@@ -0,0 +1,107 @@
+"""Connection health transitions + battery history append."""
+from __future__ import annotations
+
+import json
+import time
+
+from app.poller import (
+    HEALTH_OFFLINE_MIN_SECONDS,
+    _get_offline_threshold,
+    _record_poll_failure,
+    _record_poll_success,
+)
+
+
+def test_offline_threshold_floor():
+    assert _get_offline_threshold(30) == HEALTH_OFFLINE_MIN_SECONDS
+    assert _get_offline_threshold(60) == 180
+
+
+def test_offline_threshold_scales_with_interval():
+    assert _get_offline_threshold(120) == 360
+
+
+def test_poll_success_sets_health_key(fake_redis):
+    snap = {"BCHARGE": "100", "TIMELEFT": "45 Minutes", "BATTV": "13.5", "NOMBATTV": "12"}
+    _record_poll_success("u1", snap, 30)
+    assert fake_redis.get("ups:health:last_ok:u1") is not None
+    assert fake_redis.get("ups:health:fail_count:u1") is None
+
+
+def test_poll_success_clears_fail_count(fake_redis):
+    fake_redis.set("ups:health:fail_count:u1", "5")
+    _record_poll_success("u1", {}, 30)
+    assert fake_redis.get("ups:health:fail_count:u1") is None
+
+
+def test_poll_success_recovery_emits_info(fake_redis, clean_config_cache):
+    from unittest.mock import patch
+    fake_redis.set("ups:health:offline:u1", "1")
+    with patch("app.alerts.dispatch_alerts"):
+        _record_poll_success("u1", {}, 30)
+    # Offline flag cleared
+    assert fake_redis.get("ups:health:offline:u1") is None
+    # History has recovery entry
+    entries = fake_redis.lrange("ups:alerts:history:all", 0, -1)
+    assert any("REACHABLE" in e for e in entries)
+
+
+def test_battery_history_appended(fake_redis):
+    snap = {"BCHARGE": "100", "TIMELEFT": "45 Minutes", "BATTV": "13.5", "NOMBATTV": "12"}
+    _record_poll_success("u1", snap, 30)
+    entries = fake_redis.lrange("ups:battery:history:u1", 0, -1)
+    assert len(entries) == 1
+    parsed = json.loads(entries[0])
+    assert parsed["bcharge"] == 100.0
+    assert parsed["battv"] == 13.5
+
+
+def test_battery_history_throttled_at_60s(fake_redis):
+    snap = {"BCHARGE": "100", "TIMELEFT": "45", "BATTV": "13", "NOMBATTV": "12"}
+    _record_poll_success("u1", snap, 30)
+    # Immediately call again: no new sample (60s gate)
+    _record_poll_success("u1", snap, 30)
+    entries = fake_redis.lrange("ups:battery:history:u1", 0, -1)
+    assert len(entries) == 1
+
+
+def test_poll_failure_increments_count(fake_redis):
+    _record_poll_failure("u1", 30, "timeout")
+    assert fake_redis.get("ups:health:fail_count:u1") == "1"
+    _record_poll_failure("u1", 30, "timeout")
+    assert fake_redis.get("ups:health:fail_count:u1") == "2"
+
+
+def test_poll_failure_emits_offline_after_threshold(fake_redis, clean_config_cache):
+    from unittest.mock import patch
+    # Simulate an old last_ok so we cross the threshold
+    fake_redis.set("ups:health:last_ok:u1", str(int(time.time()) - 1000))
+    with patch("app.alerts.dispatch_alerts"):
+        _record_poll_failure("u1", 30, "connection refused")
+    # Offline flag now set
+    assert fake_redis.get("ups:health:offline:u1") == "1"
+    # Alert recorded in global history
+    entries = fake_redis.lrange("ups:alerts:history:all", 0, -1)
+    assert any("COMMLOST" in e for e in entries)
+
+
+def test_poll_failure_idempotent_while_offline(fake_redis, clean_config_cache):
+    from unittest.mock import patch
+    fake_redis.set("ups:health:last_ok:u1", str(int(time.time()) - 1000))
+    with patch("app.alerts.dispatch_alerts"):
+        _record_poll_failure("u1", 30, "x")
+        _record_poll_failure("u1", 30, "x")
+    # Only one COMMLOST alert
+    entries = fake_redis.lrange("ups:alerts:history:all", 0, -1)
+    assert sum("COMMLOST" in e for e in entries) == 1
+
+
+def test_poll_failure_below_threshold_no_alert(fake_redis, clean_config_cache):
+    from unittest.mock import patch
+    # Recent last_ok -> under threshold
+    fake_redis.set("ups:health:last_ok:u1", str(int(time.time())))
+    with patch("app.alerts.dispatch_alerts"):
+        _record_poll_failure("u1", 30, "transient")
+    assert fake_redis.get("ups:health:offline:u1") is None
+    entries = fake_redis.lrange("ups:alerts:history:all", 0, -1)
+    assert not any("COMMLOST" in e for e in entries)

tests/test_validators.py

@@ -0,0 +1,113 @@
+"""SSRF / config validators + SMTP password redaction."""
+from __future__ import annotations
+
+import pytest
+
+from app.config import SMTPConfig, UPSConfig
+from app.config_manager import smtp_redacted_dict
+from app.settings import settings
+
+
+def test_valid_ups_config():
+    u = UPSConfig(name="rack1", host="192.168.1.10", port=3551, interval_seconds=30)
+    assert u.host == "192.168.1.10"
+    assert u.interval_seconds == 30
+
+
+def test_reject_loopback_literal():
+    with pytest.raises(Exception, match="loopback"):
+        UPSConfig(name="bad", host="localhost")
+
+
+def test_reject_loopback_ip():
+    with pytest.raises(Exception, match="not a valid polling target"):
+        UPSConfig(name="bad", host="127.0.0.1")
+
+
+def test_reject_ipv6_loopback():
+    with pytest.raises(Exception, match="not a valid polling target"):
+        UPSConfig(name="bad", host="::1")
+
+
+def test_reject_link_local():
+    with pytest.raises(Exception, match="not a valid polling target"):
+        UPSConfig(name="bad", host="169.254.1.1")
+
+
+def test_reject_private_ip_when_disabled(monkeypatch):
+    monkeypatch.setattr(settings, "allow_private_ips", False)
+    with pytest.raises(Exception, match="private IP"):
+        UPSConfig(name="bad", host="10.0.0.5")
+
+
+def test_accept_private_ip_when_enabled(monkeypatch):
+    monkeypatch.setattr(settings, "allow_private_ips", True)
+    u = UPSConfig(name="good", host="10.0.0.5")
+    assert u.host == "10.0.0.5"
+
+
+def test_reject_empty_name():
+    with pytest.raises(Exception):
+        UPSConfig(name="", host="192.168.1.10")
+
+
+def test_reject_name_too_long():
+    with pytest.raises(Exception):
+        UPSConfig(name="a" * 33, host="192.168.1.10")
+
+
+def test_reject_bad_name_chars():
+    with pytest.raises(Exception, match="alphanumeric"):
+        UPSConfig(name="bad name!", host="192.168.1.10")
+
+
+def test_accept_valid_name_chars():
+    u = UPSConfig(name="ups_rack-01", host="192.168.1.10")
+    assert u.name == "ups_rack-01"
+
+
+def test_reject_invalid_port():
+    with pytest.raises(Exception):
+        UPSConfig(name="x", host="192.168.1.10", port=0)
+    with pytest.raises(Exception):
+        UPSConfig(name="x", host="192.168.1.10", port=70000)
+
+
+def test_reject_invalid_interval():
+    with pytest.raises(Exception):
+        UPSConfig(name="x", host="192.168.1.10", interval_seconds=1)
+    with pytest.raises(Exception):
+        UPSConfig(name="x", host="192.168.1.10", interval_seconds=99999)
+
+
+def test_accept_valid_hostname():
+    u = UPSConfig(name="dns", host="ups.lan.example.com")
+    assert u.host == "ups.lan.example.com"
+
+
+def test_reject_bad_hostname():
+    with pytest.raises(Exception):
+        UPSConfig(name="x", host="not valid host")
+
+
+def test_smtp_model_has_no_password_field():
+    s = SMTPConfig(host="smtp.example.com", port=587)
+    assert not hasattr(s, "password")
+
+
+def test_smtp_redacted_none():
+    assert smtp_redacted_dict(None) is None
+
+
+def test_smtp_redacted_env_set(monkeypatch):
+    monkeypatch.setattr(settings, "smtp_password", "secret")
+    s = SMTPConfig(host="smtp.example.com", port=587)
+    d = smtp_redacted_dict(s)
+    assert d["password"] == "***"
+
+
+def test_smtp_redacted_env_unset(monkeypatch):
+    monkeypatch.setattr(settings, "smtp_password", None)
+    s = SMTPConfig(host="smtp.example.com", port=587)
+    d = smtp_redacted_dict(s)
+    assert d["password"] is None