feat(docker): needs attention first, stale rows say so, no sparklines on the overview (#5123)

The Docker page opens with a "Needs attention" card: every container
with something wrong, worst problem first — restarting, crash-looping,
unhealthy, OOM-killed, exited non-zero — each with what it does ("is
restarting and keeps dying · exit 1 · since 8m ago"), linked to its page.
It is ranked by the same code as the MCP tools: container_problems,
problem_since and the per-container ranking move from mcp_tools into
service.py (ranked_containers, needs_attention), and mcp_tools' _rows and
findings now read them from there.

Counts follow as stat tiles: needs attention, unhealthy, restarting (or
crash-looping), running, stopped, stale, services, hosts.

A container whose last report is older than the host agent's staleness
window (stale_after_seconds, 180 s by default) carries a Stale badge with
when it last reported, instead of a live-looking green dot.

The per-container CPU and memory history columns are gone from the
overview, and with them one history query per container per refresh; the
container's own page keeps its charts. With nothing left to range over,
the time-range control goes too, and the page now refreshes on the poll
interval. CPU and memory are graded by Settings → Thresholds rather than
50/80 and 70/90 written into the template.

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

plugins/docker/mcp_tools.py

@@ -25,58 +25,17 @@ from steward.models.hosts import Host
 from steward.models.users import UserRole
 
 from . import service as docker_svc
+from .service import PROBLEM_SAYS, PROBLEM_SEVERITY, PROBLEMS, container_problems, problem_since
 from .dedup import dedup_by_container_id
 from .models import DockerDiskUsage, DockerSwarmNode, DockerSwarmService
 from .swarm_view import build_swarm_services, replaced_tasks, task_time
 
-PROBLEMS = ("restarting", "crash_looping", "unhealthy", "oom_killed", "exited_nonzero")
 CRASH_LOOP_DEATHS = docker_svc.CRASH_LOOP_DEATHS
 LOG_LINE_CHARS = 2000
 
 
 # ── pure helpers ─────────────────────────────────────────────────────────────
 
-def container_problems(c, failures_1h: int, failed_24h: bool,
-                       crash_loop_deaths: int = CRASH_LOOP_DEATHS) -> list[str]:
-    """What is wrong with one container, most severe first; [] when nothing is."""
-    status = (c.status or "").lower()
-    out = []
-    if status == "restarting":
-        out.append("restarting")
-    if failures_1h >= crash_loop_deaths:
-        out.append("crash_looping")
-    if status == "running" and (c.health or "") == "unhealthy":
-        out.append("unhealthy")
-    if c.oom_killed:
-        out.append("oom_killed")
-    if status in ("exited", "dead") and (c.exit_code or 0) != 0 and failed_24h:
-        out.append("exited_nonzero")
-    return out
-
-
-def problem_since(problems: list[str], deaths: list, hour_ago: datetime,
-                  unhealthy_at: datetime | None = None) -> datetime | None:
-    """When a container's worst problem began; None when the retained events
-    cannot say. `deaths` are its die/oom events of the last 24h, newest first.
-
-    A restart or crash loop dates from its first death in the last hour, an
-    OOM kill or error exit from the death that ended it, and a failing
-    healthcheck from when its health last turned unhealthy.
-    """
-    if not problems:
-        return None
-    worst = problems[0]
-    if worst in ("restarting", "crash_looping"):
-        in_hour = [e.at for e in deaths if e.at >= hour_ago]
-        return min(in_hour) if in_hour else (deaths[0].at if deaths else None)
-    if worst == "unhealthy":
-        return unhealthy_at
-    if worst == "oom_killed":
-        ooms = [e.at for e in deaths if e.event == "oom"]
-        return ooms[0] if ooms else (deaths[0].at if deaths else None)
-    return deaths[0].at if deaths else None
-
-
 def cut_middle(text: str, limit: int = LOG_LINE_CHARS) -> str:
     """Keep both ends of an over-long line and say how much of the middle went."""
     if text is None or len(text) <= limit:
@@ -117,13 +76,6 @@ def container_row(c, host_names: dict, problems: list[str], now: datetime,
     }
 
 
-_SEVERITY = {"restarting": "critical", "crash_looping": "critical", "unhealthy": "critical",
-             "oom_killed": "warning", "exited_nonzero": "warning"}
-_SAYS = {"restarting": "is restarting", "crash_looping": "keeps dying",
-         "unhealthy": "fails its healthcheck", "oom_killed": "was OOM-killed",
-         "exited_nonzero": "exited with an error"}
-
-
 def _from_iso(text: str | None) -> datetime | None:
     return datetime.fromisoformat(text.replace("Z", "+00:00")) if text else None
 
@@ -137,8 +89,8 @@ def container_findings(rows: list[dict]) -> list[dict]:
         where = f" on {r['host_name']}" if r["host_name"] else ""
         detail = f" (exit {r['exit_code']})" if r["exit_code"] not in (None, 0) else ""
         out.append(finding(
-            _SEVERITY[worst], f"container_{worst}", r["container_name"],
+            PROBLEM_SEVERITY[worst], f"container_{worst}", r["container_name"],
-            f"container {r['container_name']}{where} {' and '.join(_SAYS[p] for p in r['problems'])}{detail}",
+            f"container {r['container_name']}{where} {' and '.join(PROBLEM_SAYS[p] for p in r['problems'])}{detail}",
             since=_from_iso(r["problem_since"]),
             ids={"host_id": r["host_id"], "container_name": r["container_name"]},
             next=f'get_container(container="{r["container_name"]}", host="{r["host_id"]}")'))
@@ -204,24 +156,10 @@ async def _replaced(db, containers) -> set[tuple]:
 async def _rows(db, now: datetime) -> list[dict]:
     """Every container with its problems. A swarm task Swarm replaced has none:
     its failure, if it was one, belongs to its service (service_findings)."""
-    containers = dedup_by_container_id(await docker_svc.list_containers(db))
-    replaced = await _replaced(db, containers)
-    failures = await docker_svc.recent_failures_by_container(db, now - timedelta(hours=24))
-    hour_ago = now - timedelta(hours=1)
     host_names = await _host_names(db)
-    deaths = _crash_loop_deaths()
+    return [container_row(r["container"], host_names, r["problems"], now,
-    sick = await docker_svc.unhealthy_since(
+                          replaced=r["replaced"], since=r["since"])
-        db, [(c.host_id, c.name) for c in containers if (c.health or "") == "unhealthy"])
+            for r in await docker_svc.ranked_containers(db, now, _crash_loop_deaths())]
-    rows = []
-    for c in containers:
-        key = (c.host_id, c.name)
-        gone = key in replaced
-        events = failures.get(key, [])
-        recent = sum(1 for e in events if e.at >= hour_ago)
-        problems = [] if gone else container_problems(c, recent, bool(events), deaths)
-        since = problem_since(problems, events, hour_ago, sick.get(key))
-        rows.append(container_row(c, host_names, problems, now, replaced=gone, since=since))
-    return rows
 
 
 async def _swarm(db, now: datetime):

plugins/docker/routes.py

@@ -63,7 +63,6 @@ def _sparkline(values: list[float], width: int = 80, height: int = 20) -> str:
 @require_role(UserRole.viewer)
 async def index():
     poll_interval = current_app.config.get("MONITORS_POLL_INTERVAL", 60)
-    current_range = request.args.get("range", DEFAULT_RANGE)
     # Show the Swarm link only when a manager has actually reported topology, so
     # non-swarm installs aren't offered an always-empty page.
     async with current_app.db_sessionmaker() as db:
@@ -74,7 +73,6 @@ async def index():
     return await render_template(
         "docker/index.html",
         poll_interval=poll_interval,
-        current_range=current_range,
         has_swarm=has_swarm,
         has_disk=has_disk,
     )
@@ -83,11 +81,14 @@ async def index():
 @docker_bp.get("/rows")
 @require_role(UserRole.viewer)
 async def rows():
-    """HTMX fragment: swarm services (collapsed, cluster-complete) on top, then
+    """HTMX fragment: what needs attention first, then swarm services
-    non-swarm containers grouped by host with resource sparklines."""
+    (collapsed, cluster-complete), then non-swarm containers grouped by host."""
-    since, range_key = parse_range(request.args.get("range"))
+    now = datetime.now(timezone.utc)
+    stale_after = docker_svc.stale_after_seconds(current_app.config)
 
     async with current_app.db_sessionmaker() as db:
+        ranked = await docker_svc.ranked_containers(
+            db, now, docker_svc.crash_loop_deaths(current_app.config))
         containers = await docker_svc.list_containers(db)
         services = list((await db.execute(select(DockerSwarmService))).scalars())
         nodes = list((await db.execute(select(DockerSwarmNode))).scalars())
@@ -118,13 +119,11 @@ async def rows():
                     "host_name": host.name if host else c.host_id,
                     "containers": [], "running": 0, "stopped": 0,
                 }
-            cpu_hist, mem_hist = await docker_svc.container_history(db, c.host_id, c.name, since)
             g["containers"].append({
                 "container": c,
                 "ports": json.loads(c.ports_json) if c.ports_json else [],
                 "uptime": _uptime(c.started_at) if c.status == "running" else None,
-                "sparkline_cpu": _sparkline(cpu_hist),
+                "stale": _is_stale(c, now, stale_after),
-                "sparkline_mem": _sparkline(mem_hist),
             })
             if c.status == "running":
                 g["running"] += 1
@@ -152,17 +151,28 @@ async def rows():
     swarm_services = swarm_view["services"]
     standalone_running = sum(g["running"] for g in host_groups)
     ghost_total = sum(r["count"] for s in swarm_services for r in s["replicas"] if r["ghost"])
+    attention = [dict(r, stale=_is_stale(r["container"], now, stale_after))
+                 for r in docker_svc.needs_attention(ranked)]
     return await render_template(
         "docker/rows.html",
         host_groups=host_groups,
         swarm_services=swarm_services,
+        attention=attention, says=docker_svc.PROBLEM_SAYS, severity=docker_svc.PROBLEM_SEVERITY,
+        unhealthy=sum("unhealthy" in r["problems"] for r in attention),
+        restarting=sum(bool({"restarting", "crash_looping"} & set(r["problems"])) for r in attention),
+        stale=sum(1 for g in host_groups for i in g["containers"] if i["stale"]),
         running=standalone_running + sum(s["running"] for s in swarm_services),
         stopped=len(standalone) - standalone_running,
         total=len(containers) + ghost_total,
-        range_key=range_key,
     )
 
 
+def _is_stale(c, now: datetime, stale_after: int) -> bool:
+    """A container whose last report is older than the agent's staleness
+    window: what the page shows may no longer be true."""
+    return c.scraped_at is None or (now - c.scraped_at).total_seconds() > stale_after
+
+
 @docker_bp.get("/widget")
 @require_role(UserRole.viewer)
 async def widget():

plugins/docker/service.py

@@ -6,14 +6,18 @@ Quart's pipeline) call the same code.
 """
 from __future__ import annotations
 
-from datetime import datetime
+from datetime import datetime, timedelta
 
 from sqlalchemy import func, select
 
 from steward.core.time_range import bucket_seconds
 from steward.models.hosts import Host
 
-from .models import DockerContainer, DockerEvent, DockerLog, DockerMetric, DockerMetricHourly
+from .dedup import dedup_by_container_id
+from .models import (
+    DockerContainer, DockerEvent, DockerLog, DockerMetric, DockerMetricHourly, DockerSwarmService,
+)
+from .swarm_view import replaced_tasks
 
 
 CRASH_LOOP_DEATHS = 3  # default; Settings → Thresholds sets it
@@ -278,3 +282,94 @@ def prune_extra_vars(prune_target: str) -> dict:
     if prune_target == "images":
         extra_vars["prune_all_images"] = True
     return extra_vars
+
+
+# ── what is wrong with a container ───────────────────────────────────────────
+# Shared by the Docker page's "Needs attention" and the MCP tools, so both
+# rank the same containers the same way.
+
+PROBLEMS = ("restarting", "crash_looping", "unhealthy", "oom_killed", "exited_nonzero")
+
+def container_problems(c, failures_1h: int, failed_24h: bool,
+                       crash_loop_deaths: int = CRASH_LOOP_DEATHS) -> list[str]:
+    """What is wrong with one container, most severe first; [] when nothing is."""
+    status = (c.status or "").lower()
+    out = []
+    if status == "restarting":
+        out.append("restarting")
+    if failures_1h >= crash_loop_deaths:
+        out.append("crash_looping")
+    if status == "running" and (c.health or "") == "unhealthy":
+        out.append("unhealthy")
+    if c.oom_killed:
+        out.append("oom_killed")
+    if status in ("exited", "dead") and (c.exit_code or 0) != 0 and failed_24h:
+        out.append("exited_nonzero")
+    return out
+
+
+def problem_since(problems: list[str], deaths: list, hour_ago: datetime,
+                  unhealthy_at: datetime | None = None) -> datetime | None:
+    """When a container's worst problem began; None when the retained events
+    cannot say. `deaths` are its die/oom events of the last 24h, newest first.
+
+    A restart or crash loop dates from its first death in the last hour, an
+    OOM kill or error exit from the death that ended it, and a failing
+    healthcheck from when its health last turned unhealthy.
+    """
+    if not problems:
+        return None
+    worst = problems[0]
+    if worst in ("restarting", "crash_looping"):
+        in_hour = [e.at for e in deaths if e.at >= hour_ago]
+        return min(in_hour) if in_hour else (deaths[0].at if deaths else None)
+    if worst == "unhealthy":
+        return unhealthy_at
+    if worst == "oom_killed":
+        ooms = [e.at for e in deaths if e.event == "oom"]
+        return ooms[0] if ooms else (deaths[0].at if deaths else None)
+    return deaths[0].at if deaths else None
+
+
+PROBLEM_SEVERITY = {"restarting": "critical", "crash_looping": "critical", "unhealthy": "critical",
+             "oom_killed": "warning", "exited_nonzero": "warning"}
+PROBLEM_SAYS = {"restarting": "is restarting", "crash_looping": "keeps dying",
+         "unhealthy": "fails its healthcheck", "oom_killed": "was OOM-killed",
+         "exited_nonzero": "exited with an error"}
+
+
+
+async def ranked_containers(db, now: datetime, crash_loop_deaths: int) -> list[dict]:
+    """Every container (deduplicated) with what is wrong with it, most severe
+    first: {container, problems, since, replaced}. A swarm task Swarm has
+    replaced has no problems; its failure belongs to its service."""
+    containers = dedup_by_container_id(await list_containers(db))
+    services = list((await db.execute(select(DockerSwarmService))).scalars())
+    replaced = replaced_tasks(containers, services)
+    failures = await recent_failures_by_container(db, now - timedelta(hours=24))
+    hour_ago = now - timedelta(hours=1)
+    sick = await unhealthy_since(
+        db, [(c.host_id, c.name) for c in containers if (c.health or "") == "unhealthy"])
+    out = []
+    for c in containers:
+        key = (c.host_id, c.name)
+        gone = key in replaced
+        events = failures.get(key, [])
+        recent = sum(1 for e in events if e.at >= hour_ago)
+        problems = [] if gone else container_problems(c, recent, bool(events), crash_loop_deaths)
+        out.append({"container": c, "problems": problems, "replaced": gone,
+                    "since": problem_since(problems, events, hour_ago, sick.get(key))})
+    return out
+
+
+def needs_attention(ranked: list[dict]) -> list[dict]:
+    """The containers with something wrong, worst problem first, then by host
+    and name."""
+    return sorted((r for r in ranked if r["problems"]),
+                  key=lambda r: (PROBLEMS.index(r["problems"][0]), r["container"].host_id, r["container"].name))
+
+
+def stale_after_seconds(config) -> int:
+    """How old a container's last report may be before the page calls it
+    stale: the host agent's own staleness window, since the agent reports it."""
+    return int(((config.get("PLUGINS") or {}).get("host_agent") or {}).get("stale_after_seconds", 180))

plugins/docker/templates/docker/index.html

@@ -1,24 +1,16 @@
 {% extends "base.html" %}
+{% from "_macros.html" import page_header %}
 {% block title %}Docker — Steward{% endblock %}
 {% block content %}
-<div style="display:flex;align-items:center;justify-content:space-between;margin-bottom:1.5rem;gap:1rem;flex-wrap:wrap;">
+{% call page_header("Docker", "What needs attention first, then every service and container.") %}
-  <div style="display:flex;align-items:baseline;gap:1rem;">
+  {% if has_swarm %}<a class="btn btn-sm btn-ghost" href="/plugins/docker/swarm">Swarm</a>{% endif %}
-    <h1 class="page-title" style="margin-bottom:0;">Docker</h1>
+  {% if has_disk %}<a class="btn btn-sm btn-ghost" href="/plugins/docker/disk">Disk</a>{% endif %}
-    {% if has_swarm %}
+{% endcall %}
-    <a href="/plugins/docker/swarm" style="font-size:0.85rem;color:var(--accent);text-decoration:none;">Swarm →</a>
-    {% endif %}
-    {% if has_disk %}
-    <a href="/plugins/docker/disk" style="font-size:0.85rem;color:var(--accent);text-decoration:none;">Disk →</a>
-    {% endif %}
-  </div>
-  {% include "_time_range.html" %}
-</div>
 
 <div id="docker-rows"
      hx-get="/plugins/docker/rows"
-     hx-trigger="load, rangeChange from:body"
+     hx-trigger="load, every {{ poll_interval }}s"
-     hx-include="#time-range"
      hx-swap="innerHTML">
-  <div style="color:var(--text-muted);font-size:0.9rem;padding:2rem;">Loading...</div>
+  <div class="card"><p class="field-hint">Loading…</p></div>
 </div>
 {% endblock %}

plugins/docker/templates/docker/rows.html

@@ -1,29 +1,43 @@
 {# docker/rows.html — HTMX fragment for the Docker main page, grouped by host #}
 
-{# ── Summary strip ─────────────────────────────────────────────────────────── #}
+{# docker/rows.html — HTMX fragment for the Docker page: what needs attention
-<div style="display:grid;grid-template-columns:repeat(auto-fill,minmax(140px,1fr));gap:0.75rem;margin-bottom:1.5rem;">
+   first (service.needs_attention, the ranking the MCP tools use), the counts,
-  <div class="card" style="margin-bottom:0;">
+   then swarm services and each host's containers. #}
-    <div class="section-title" style="margin-bottom:0.35rem;">Running</div>
+{% from "_macros.html" import stat_tile, status_badge, status_dot %}
-    <span class="stat-val" style="color:var(--green);">{{ running }}</span>
+
-  </div>
+{# ── Needs attention ──────────────────────────────────────────────────────── #}
-  <div class="card" style="margin-bottom:0;">
+{% if attention %}
-    <div class="section-title" style="margin-bottom:0.35rem;">Stopped</div>
+<section class="card problem-card" aria-labelledby="docker-problems">
-    <span class="stat-val" style="{% if stopped %}color:var(--text-muted){% endif %};">{{ stopped }}</span>
+  <h2 id="docker-problems" class="section-title" style="margin-bottom:0.6rem;">Needs attention</h2>
-  </div>
+  <ul class="problem-list">
-  <div class="card" style="margin-bottom:0;">
+    {% for r in attention %}
-    <div class="section-title" style="margin-bottom:0.35rem;">Total</div>
+    {% set c = r.container %}
-    <span class="stat-val">{{ total }}</span>
+    {% set worst = r.problems[0] %}
-  </div>
+    <li>
-  {% if swarm_services %}
+      {{ status_badge('crit' if severity[worst] == 'critical' else 'warn', worst|replace('_', ' ')|capitalize) }}
-  <div class="card" style="margin-bottom:0;">
+      <a href="/plugins/docker/container/{{ c.host_id }}/{{ c.name }}">{{ c.name }}</a>
-    <div class="section-title" style="margin-bottom:0.35rem;">Services</div>
+      <span class="problem-detail">
-    <span class="stat-val">{{ swarm_services | length }}</span>
+        {%- for p in r.problems %}{{ says[p] }}{{ " and " if not loop.last }}{% endfor %}
-  </div>
+        {%- if c.exit_code and worst in ('exited_nonzero', 'crash_looping', 'restarting') %} · exit {{ c.exit_code }}{% endif %}
-  {% endif %}
+        {%- if r.since %} · since {{ r.since|ago }}{% endif %}
-  <div class="card" style="margin-bottom:0;">
+        {%- if r.stale %} · {{ status_badge('stale') }} last report {{ c.scraped_at|ago if c.scraped_at else "never" }}{% endif -%}
-    <div class="section-title" style="margin-bottom:0.35rem;">Hosts</div>
+      </span>
-    <span class="stat-val">{{ host_groups | length }}</span>
+    </li>
-  </div>
+    {% endfor %}
+  </ul>
+</section>
+{% endif %}
+
+{# ── Counts ───────────────────────────────────────────────────────────────── #}
+<div class="stat-strip">
+  {{ stat_tile("Needs attention", attention|length, state="crit" if attention else "inactive") }}
+  {{ stat_tile("Unhealthy", unhealthy, state="crit" if unhealthy else "inactive") }}
+  {{ stat_tile("Restarting", restarting, state="crit" if restarting else "inactive", hint="restarting or crash-looping") }}
+  {{ stat_tile("Running", running, state="up" if running else "inactive") }}
+  {{ stat_tile("Stopped", stopped, state="inactive") }}
+  {% if stale %}{{ stat_tile("Stale", stale, state="stale", hint="no recent report") }}{% endif %}
+  {% if swarm_services %}{{ stat_tile("Services", swarm_services|length, href="/plugins/docker/swarm") }}{% endif %}
+  {{ stat_tile("Hosts", host_groups|length) }}
 </div>
 
 {# ── Swarm services: collapsed per-service, each current replica labelled with
@@ -43,7 +57,7 @@
         <span class="dot {% if s.state == 'healthy' %}dot-up{% elif s.state == 'degraded' %}dot-warn{% else %}dot-down{% endif %}"></span>
         <span style="font-weight:600;font-size:0.9rem;">{{ s.name }}</span>
         <span style="font-size:0.72rem;color:var(--text-dim);">{{ s.mode }}</span>
-        <span style="font-size:0.78rem;color:{% if s.state == 'healthy' %}var(--green){% elif s.state == 'degraded' %}var(--orange){% else %}var(--red){% endif %};font-variant-numeric:tabular-nums;">
+        <span class="tone-{{ 'ok' if s.state == 'healthy' else ('warn' if s.state == 'degraded' else 'crit') }}" style="font-size:0.78rem;font-variant-numeric:tabular-nums;">
           {{ s.running }}/{{ s.desired }} running
         </span>
       </div>
@@ -92,35 +106,33 @@
         <th>Image</th>
         <th>Ports</th>
         <th>CPU %</th>
-        <th style="min-width:100px;">CPU history</th>
         <th>Mem %</th>
-        <th style="min-width:100px;">Mem history</th>
       </tr>
     </thead>
     <tbody>
       {% for sub in g.subgroups %}
       {% if g.grouped and sub.label %}
-      <tr><td colspan="7" style="padding:0.4rem 0.75rem 0.2rem;font-size:0.72rem;color:var(--text-dim);text-transform:uppercase;letter-spacing:0.05em;font-weight:600;">{{ sub.label }}</td></tr>
+      <tr><td colspan="5" style="padding:0.4rem 0.75rem 0.2rem;font-size:0.72rem;color:var(--text-dim);text-transform:uppercase;letter-spacing:0.05em;font-weight:600;">{{ sub.label }}</td></tr>
       {% endif %}
       {% for item in sub.containers %}
       {% set c = item.container %}
       <tr>
         <td>
           <div style="display:flex;align-items:center;gap:0.5rem;">
-            <span class="dot {% if c.status == 'running' %}dot-up{% elif c.status == 'paused' %}dot-warn{% else %}dot-down{% endif %}"></span>
+            {{ status_dot('stale' if item.stale else ('up' if c.status == 'running' else ('pending' if c.status == 'paused' else 'down')), 'Stale' if item.stale else c.status|capitalize) }}
             <div>
               <div style="font-weight:500;font-size:0.9rem;">
                 <a href="/plugins/docker/container/{{ c.host_id }}/{{ c.name }}" style="color:inherit;text-decoration:none;border-bottom:1px dotted var(--border-mid);">{{ c.name }}</a>
-                {% if c.health == 'healthy' %}<span title="healthy" style="color:var(--green);font-size:0.7rem;">●</span>
+                {% if c.health == 'unhealthy' %} {{ status_badge('crit', 'Unhealthy') }}
-                {% elif c.health == 'unhealthy' %}<span title="unhealthy" style="color:var(--red);font-size:0.7rem;">●</span>
+                {% elif c.health == 'starting' %} {{ status_badge('pending', 'Starting') }}{% endif %}
-                {% elif c.health == 'starting' %}<span title="health: starting" style="color:var(--orange);font-size:0.7rem;">◐</span>{% endif %}
+                {% if item.stale %} {{ status_badge('stale', title="Last report " ~ (c.scraped_at.strftime('%Y-%m-%d %H:%M UTC') if c.scraped_at else 'never') ~ "; what is shown may no longer be true") }}{% endif %}
               </div>
               <div style="font-size:0.73rem;color:var(--text-muted);">
                 {{ c.status }}{% if item.uptime %} · up {{ item.uptime }}{% endif %}
                 {% if c.status != 'running' and c.exit_code is not none and c.exit_code != 0 %}
-                  · <span style="color:var(--red);">exit {{ c.exit_code }}{% if c.oom_killed %} (OOM){% endif %}</span>
+                  · <span class="tone-crit">exit {{ c.exit_code }}{% if c.oom_killed %} (OOM){% endif %}</span>
                 {% endif %}
-                {% if c.restart_count %} · <span title="restart count" style="color:var(--orange);">⟳{{ c.restart_count }}</span>{% endif %}
+                {% if c.restart_count %} · <span class="tone-warn" title="Restarted {{ c.restart_count }} times">⟳{{ c.restart_count }}</span>{% endif %}
               </div>
               {% if c.service_name or c.compose_project %}
               <div style="font-size:0.68rem;color:var(--text-dim);margin-top:0.1rem;">
@@ -140,24 +152,18 @@
         </td>
         <td style="font-family:ui-monospace,monospace;font-size:0.88rem;">
           {% if c.cpu_pct is not none %}
-          <span style="color:{% if c.cpu_pct > 80 %}var(--red){% elif c.cpu_pct > 50 %}var(--orange){% else %}var(--text){% endif %};">
+          <span class="tone-{{ status_class(c.cpu_pct, 'cpu') }}">{{ "%.1f" | format(c.cpu_pct) }}%</span>
-            {{ "%.1f" | format(c.cpu_pct) }}%
-          </span>
           {% else %}
           <span style="color:var(--text-dim);">—</span>
           {% endif %}
         </td>
-        <td>{{ item.sparkline_cpu | safe }}</td>
         <td style="font-family:ui-monospace,monospace;font-size:0.88rem;">
           {% if c.mem_pct is not none %}
-          <span style="color:{% if c.mem_pct > 90 %}var(--red){% elif c.mem_pct > 70 %}var(--orange){% else %}var(--text){% endif %};">
+          <span class="tone-{{ status_class(c.mem_pct, 'mem') }}">{{ "%.1f" | format(c.mem_pct) }}%</span>
-            {{ "%.1f" | format(c.mem_pct) }}%
-          </span>
           {% else %}
           <span style="color:var(--text-dim);">—</span>
           {% endif %}
         </td>
-        <td>{{ item.sparkline_mem | safe }}</td>
       </tr>
       {% endfor %}
       {% endfor %}

tests/plugins/docker/test_docker_page.py

@@ -0,0 +1,76 @@
+"""The Docker page leads with what needs attention, ranked the way the MCP
+tools rank it, and marks a container whose report is old as stale (#5123)."""
+import re
+from datetime import datetime, timedelta, timezone
+from types import SimpleNamespace as NS
+
+
+from plugins.docker import service as svc
+from plugins.docker.routes import _is_stale
+from tests._templates import render_plugin_template
+from tests.test_style_tokens import hex_colours
+
+NOW = datetime.now(timezone.utc)
+
+
+def _c(name, **kw):
+    base = dict(name=name, host_id="h1", image="nginx:1", status="running", health=None, exit_code=None,
+                oom_killed=False, restart_count=0, cpu_pct=5.0, mem_pct=10.0, compose_project=None,
+                service_name=None, ports_json=None, started_at=NOW - timedelta(hours=2),
+                scraped_at=NOW - timedelta(seconds=20))
+    base.update(kw)
+    return NS(**base)
+
+
+def test_needs_attention_ranks_the_worst_problem_first():
+    ranked = [{"container": _c(n), "problems": p, "since": None, "replaced": False} for n, p in [
+        ("fine", []), ("oom", ["oom_killed"]), ("sick", ["unhealthy"]),
+        ("loop", ["crash_looping"]), ("exit", ["exited_nonzero"]), ("bounce", ["restarting", "crash_looping"])]]
+    assert [r["container"].name for r in svc.needs_attention(ranked)] == ["bounce", "loop", "sick", "oom", "exit"]
+
+
+def test_stale_is_older_than_the_agents_window():
+    assert not _is_stale(_c("a"), NOW, 180)
+    assert _is_stale(_c("a", scraped_at=NOW - timedelta(minutes=10)), NOW, 180)
+    assert _is_stale(_c("a", scraped_at=None), NOW, 180)
+    assert svc.stale_after_seconds({"PLUGINS": {"host_agent": {"stale_after_seconds": 60}}}) == 60
+    assert svc.stale_after_seconds({}) == 180
+
+
+async def test_a_crash_loop_comes_first_and_a_stale_row_says_so(app):
+    loop = _c("api", status="restarting", exit_code=1, restart_count=7)
+    old = _c("cache", scraped_at=NOW - timedelta(minutes=30), cpu_pct=95.0)
+    attention = [{"container": loop, "problems": ["restarting", "crash_looping"], "replaced": False,
+                  "since": datetime.now(timezone.utc) - timedelta(minutes=8), "stale": False}]
+    groups = [{"host": None, "host_id": "h1", "host_name": "nas", "running": 1, "stopped": 1, "grouped": False,
+               "subgroups": [{"label": None, "containers": [
+                   {"container": loop, "ports": [], "uptime": None, "stale": False},
+                   {"container": old, "ports": [], "uptime": "2h", "stale": True}]}]}]
+    async with app.test_request_context("/plugins/docker/rows"):
+        page = await render_plugin_template(
+            app, "docker", "docker/rows.html", attention=attention, says=svc.PROBLEM_SAYS, severity=svc.PROBLEM_SEVERITY,
+            unhealthy=0, restarting=1, stale=1, host_groups=groups, swarm_services=[],
+            running=1, stopped=1, total=2)
+    page = re.sub(r">\s+<", "><", " ".join(page.split()))
+
+    card = page[page.index("Needs attention"):page.index("</section>")]
+    assert '<span class="badge badge-crit">Restarting</span>' in card
+    assert 'href="/plugins/docker/container/h1/api">api</a>' in card
+    assert "is restarting and keeps dying · exit 1 · since <time" in card and ">8m ago</time>" in card
+    assert "cache" not in card
+    assert page.index("Needs attention") < page.index("<table")
+
+    assert '<span class="stat-tile-label">Restarting</span><span class="stat-tile-value">1</span>' in page
+    assert '<span class="stat-tile-label">Stale</span><span class="stat-tile-value">1</span>' in page
+    assert 'class="badge badge-warn" title="Last report' in page          # the stale row
+    assert '<span class="tone-crit">95.0%</span>' in page                  # CPU through Settings → Thresholds
+    assert "history" not in page.lower() and "<svg" not in page            # no sparklines on the overview
+    assert not hex_colours(page), hex_colours(page)
+
+
+async def test_a_healthy_board_has_no_attention_card(app):
+    async with app.test_request_context("/plugins/docker/rows"):
+        page = await render_plugin_template(
+            app, "docker", "docker/rows.html", attention=[], says=svc.PROBLEM_SAYS, severity=svc.PROBLEM_SEVERITY,
+            unhealthy=0, restarting=0, stale=0, host_groups=[], swarm_services=[], running=0, stopped=0, total=0)
+    assert "Needs attention</h2>" not in page and "No containers reported yet" in page