feat(dashboard): an Alerts widget beside the Overview; the Overview's overflow links to Alerts and Monitors (#5510)

The Alerts widget (/alerts/widget) is shaped like every widget: a verdict
("2 firing, 1 silent." / "Nothing firing: all 12 rules are quiet."), then
what is firing or silent with its reading now, what the rule fires on and
since when, then what someone acknowledged and who; a line when alert
email failed in the last 7 days or is retrying (outbox.alert_trouble),
linking admins to Settings -> System; the quiet rules counted, a covering
rule once. It sits beside the Overview in the recommended layout (6x4
each).

The Overview's "+N more" line pointed at the removed /status/; alerts now
link to /alerts/ and monitors to /monitors/.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

steward/core/outbox.py

@@ -35,7 +35,7 @@ from datetime import datetime, timedelta, timezone
 from email.message import EmailMessage
 from typing import TYPE_CHECKING, Callable
 
-from sqlalchemy import insert, or_, select, text, update
+from sqlalchemy import and_, insert, or_, select, text, update
 
 from steward.models.outbox import EmailOutbox, email_outbox_events
 
@@ -434,3 +434,27 @@ async def outgoing(db, now: datetime | None = None, limit: int = 20) -> dict:
     shown = sum(counts.get(s, 0) for s in open_states)
     return {"rows": rows, "more": max(shown - len(rows), 0), "sent": counts.get("sent", 0),
             "problems": problems}
+
+
+async def alert_trouble(db, now: datetime | None = None) -> dict:
+    """The dashboard Alerts widget's delivery line (#5510): alert emails of
+    the last SHOWN_DAYS that failed, and those still retrying (tried, not yet
+    accepted), and the line that says so ("" when neither).
+    {"failed": n, "retrying": n, "said": str}."""
+    from sqlalchemy import func
+    now = now or datetime.now(timezone.utc)
+    rows = dict((await db.execute(
+        select(EmailOutbox.state, func.count())
+        .where(EmailOutbox.kind == "alert", EmailOutbox.created_at >= now - timedelta(days=SHOWN_DAYS),
+               or_(EmailOutbox.state == "failed",
+                   and_(EmailOutbox.state.in_(("pending", "sending")), EmailOutbox.attempts > 0)))
+        .group_by(EmailOutbox.state))).all())
+    failed = rows.get("failed", 0)
+    retrying = rows.get("pending", 0) + rows.get("sending", 0)
+    said = []
+    if failed:
+        said.append(f"{failed} alert email{'s' if failed != 1 else ''} failed in the last {SHOWN_DAYS} days")
+    if retrying:
+        said.append(f"{retrying} still retrying" if failed
+                    else f"{retrying} alert email{'s' if retrying != 1 else ''} still retrying")
+    return {"failed": failed, "retrying": retrying, "said": "; ".join(said)}

steward/core/widgets.py

@@ -70,6 +70,16 @@ WIDGET_REGISTRY: dict[str, dict] = {
         "poll": True,
         "params": [],
     },
+    "alerts": {
+        "key": "alerts",
+        "label": "Alerts",
+        "description": "What is firing or silent, its reading and since when, then what someone has acknowledged; says when alert email is failing",
+        "hx_url": "/alerts/widget",
+        "detail_url": "/alerts/",
+        "plugin": None,
+        "poll": True,
+        "params": [],
+    },
     "hosts_overview": {
         "key": "hosts_overview",
         "label": "Hosts — Overview",
@@ -305,13 +315,15 @@ RETIRED = {
     "uptime_summary": "monitors",         # Monitors shows 24h/7d/30d uptime per monitor (#5250)
 }
 
-# The recommended dashboard (#5213): what needs attention first, then the
+# The recommended dashboard (#5213): what needs attention first (the
+# Overview, and beside it the Alerts, #5510), then the
 # fleet, then each integration. (key, width, height, config) on the 12-column
 # grid, one cell 70px tall; a widget whose plugin is off is skipped and the
 # rest close up. Seeded into the first dashboard and offered by
 # "Reset to the recommended layout" in edit mode.
 RECOMMENDED: list[tuple[str, int, int, dict]] = [
-    ("status_overview", 12, 3, {}),
+    ("status_overview", 6, 4, {}),
+    ("alerts", 6, 4, {}),
     ("hosts_overview", 6, 7, {}),
     ("monitors", 6, 7, {"show_down_only": "no", "limit": 10}),
     ("docker_containers", 6, 6, {}),
@@ -342,7 +354,7 @@ def recommended_layout(plugins_cfg: dict) -> list[dict]:
 # Group each widget for the picker (Core monitors / Monitoring capabilities /
 # Integrations), mirroring the Settings → Plugins taxonomy. Capability plugins
 # are host/monitoring facets; traefik/unifi are discrete vendor integrations.
-_CORE_WIDGETS = {"monitors", "status_overview", "hosts_overview"}
+_CORE_WIDGETS = {"monitors", "status_overview", "alerts", "hosts_overview"}
 _INTEGRATION_PLUGINS = {"traefik", "unifi"}
 GROUP_LABELS = {
     "core": "Core monitors",