feat(docker): swarm-aware container views — collapse replicas, show host, surface agent-less tasks

The container list and dashboard widget listed each swarm task as its own cryptic
row (svc.1.<taskid>) and could only show tasks on nodes that run a Steward agent.
Make both views service-centric and cluster-complete.

- swarm_view.build_swarm_services(): model-free builder that merges real container
  rows (collected local-per-node, so host_id = the node a task runs on) with the
  managers' placement (DockerSwarmService.placement_json). Where placement counts
  exceed the local rows on a node — i.e. nodes with no agent — it synthesizes
  "ghost" replicas so the replica list matches the cluster. Non-swarm containers
  pass through grouped by host.
- /rows + rows.html: a Swarm-services panel collapses each service to one block,
  every replica a host chip (status + cpu); ghosts render dashed "N · no agent".
  Non-swarm/compose containers keep the per-host table below. New Services stat.
- /widget + widget.html: same service collapse with host chips; standalone
  containers stay grouped by host.
- Unit tests for the builder (real+ghost merge, full coverage, partial agent-less,
  synthesized service, down state).

No migration — node_id, placement_json and swarm-node hostnames are already stored.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016Jg27rgypiW2efULXJDtMC

plugins/docker/routes.py

@@ -11,6 +11,7 @@ from steward.models.users import UserRole
 from steward.models.hosts import Host
 from steward.core.time_range import parse_range, DEFAULT_RANGE
 from .dedup import dedup_by_container_id
+from .swarm_view import build_swarm_services
 from .models import (
     DockerContainer, DockerDiskUsage, DockerEvent, DockerImage, DockerMetric,
     DockerSwarmNode, DockerSwarmService,
@@ -138,7 +139,8 @@ async def _host_map(db, host_ids: set[str]) -> dict[str, Host]:
 @docker_bp.get("/rows")
 @require_role(UserRole.viewer)
 async def rows():
-    """HTMX fragment: containers grouped by host, with resource sparklines."""
+    """HTMX fragment: swarm services (collapsed, cluster-complete) on top, then
+    non-swarm containers grouped by host with resource sparklines."""
     since, range_key = parse_range(request.args.get("range"))
 
     async with current_app.db_sessionmaker() as db:
@@ -148,12 +150,24 @@ async def rows():
                 DockerContainer.name,
             )
         )).scalars())
-        hosts = await _host_map(db, {c.host_id for c in containers})
+        services = list((await db.execute(select(DockerSwarmService))).scalars())
+        nodes = list((await db.execute(select(DockerSwarmNode))).scalars())
+        hosts = await _host_map(
+            db,
+            {c.host_id for c in containers}
+            | {n.host_id for n in nodes} | {s.host_id for s in services},
+        )
 
-        # Group by host so each container is clearly attributed to the box it
-        # runs on (names are only unique within a host now).
+        # Swarm tasks collapse into per-service rows (each replica labelled with
+        # its host, plus "ghost" replicas for tasks on agent-less nodes, drawn
+        # from the managers' placement). Non-swarm containers keep the host view.
+        node_hostname = {n.node_id: n.hostname for n in nodes if n.hostname}
+        host_name = {h.id: h.name for h in hosts.values()}
+        swarm_view = build_swarm_services(containers, services, node_hostname, host_name)
+
+        standalone = [c for c in containers if not c.service_name]
         groups: dict[str, dict] = {}
-        for c in containers:
+        for c in standalone:
             g = groups.get(c.host_id)
             if g is None:
                 host = hosts.get(c.host_id)
@@ -176,15 +190,15 @@ async def rows():
             else:
                 g["stopped"] += 1
 
-    # Sub-group each host's containers by compose project (or swarm service),
-    # preserving the running-first ordering. Containers with neither label fall
-    # into an unlabelled bucket rendered flat, so plain hosts look unchanged.
+    # Sub-group each host's containers by compose project, preserving the
+    # running-first ordering. Containers with no project fall into an unlabelled
+    # bucket rendered flat, so plain hosts look unchanged.
     for g in groups.values():
         subs: dict[str, dict] = {}
         order: list[str] = []
         for item in g["containers"]:
             c = item["container"]
-            label = c.compose_project or c.service_name or None
+            label = c.compose_project or None
             key = label or ""
             if key not in subs:
                 subs[key] = {"label": label, "containers": []}
@@ -194,14 +208,16 @@ async def rows():
         g["grouped"] = any(s["label"] for s in g["subgroups"])
 
     host_groups = sorted(groups.values(), key=lambda g: g["host_name"].lower())
-    running = sum(g["running"] for g in host_groups)
-    total = len(containers)
+    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"])
     return await render_template(
         "docker/rows.html",
         host_groups=host_groups,
-        running=running,
-        stopped=total - running,
-        total=total,
+        swarm_services=swarm_services,
+        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,
     )
 
@@ -221,6 +237,8 @@ async def widget():
                 DockerContainer.name,
             )
         )).scalars()))
+        services = list((await db.execute(select(DockerSwarmService))).scalars())
+        nodes = list((await db.execute(select(DockerSwarmNode))).scalars())
         # Recent failures are more actionable than a static "stopped" tally. Count
         # DISTINCT container names so the two managers' duplicate die/oom events
         # don't double it.
@@ -228,17 +246,28 @@ async def widget():
             select(func.count(func.distinct(DockerEvent.container_name)))
             .where(DockerEvent.event.in_(("die", "oom")), DockerEvent.at >= since_24h)
         )).scalar() or 0
-        hosts = await _host_map(db, {c.host_id for c in all_containers})
+        hosts = await _host_map(
+            db,
+            {c.host_id for c in all_containers}
+            | {n.host_id for n in nodes} | {s.host_id for s in services},
+        )
 
-    running = [c for c in all_containers if c.status == "running"]
-    display = all_containers if show_stopped else running
+    # Swarm tasks collapse into per-service rows (each replica tagged with its
+    # host); non-swarm containers stay grouped by host.
+    node_hostname = {n.node_id: n.hostname for n in nodes if n.hostname}
+    host_name = {h.id: h.name for h in hosts.values()}
+    swarm_services = build_swarm_services(all_containers, services, node_hostname, host_name)["services"]
+
+    standalone = [c for c in all_containers if not c.service_name]
+    running = [c for c in standalone if c.status == "running"]
+    display = standalone if show_stopped else running
 
     # Group for display so multi-host fleets read clearly; single-host stays flat.
     groups: dict[str, dict] = {}
     for c in display:
         g = groups.setdefault(c.host_id, {
             "host_id": c.host_id,
-            "host_name": hosts[c.host_id].name if c.host_id in hosts else c.host_id,
+            "host_name": host_name.get(c.host_id, c.host_id),
             "containers": [],
         })
         g["containers"].append(c)
@@ -248,9 +277,10 @@ async def widget():
         "docker/widget.html",
         host_groups=host_groups,
         multi_host=len(host_groups) > 1,
-        running_count=len(running),
+        swarm_services=swarm_services,
+        running_count=len(running) + sum(s["running"] for s in swarm_services),
         failed_24h=failed_24h,
-        total_count=len(all_containers),
+        total_count=len(all_containers) + len(swarm_services),
         show_stopped=show_stopped,
         widget_id=widget_id,
     )

plugins/docker/swarm_view.py

@@ -0,0 +1,140 @@
+"""Model-free builder for the swarm-aware container view.
+
+Merges two sources into one service-grouped, cluster-complete picture:
+  • docker_containers — real container rows, collected LOCAL-per-node, so each
+    carries the host (= node) it runs on plus its swarm service/node labels.
+  • DockerSwarmService.placement_json — the managers' cluster-wide task→node
+    counts, which include tasks on nodes that run no Steward agent (and so have
+    no container row of their own).
+
+For every swarm service we list the real replicas we have detail for, then add
+"ghost" replicas for the remaining placement count on each node (agent-less
+nodes). Non-swarm containers pass through grouped by host, unchanged.
+
+Kept model-free (no ORM imports) so it's unit-testable and safe to import in
+tests without the plugin-loader "table already defined" gotcha.
+"""
+from __future__ import annotations
+
+import json
+
+
+def _service_state(running: int, desired: int) -> str:
+    if desired > 0 and running >= desired:
+        return "healthy"
+    if running > 0:
+        return "degraded"
+    return "down"
+
+
+def _placement(service) -> list[dict]:
+    raw = getattr(service, "placement_json", None) or "[]"
+    try:
+        data = json.loads(raw)
+    except (ValueError, TypeError):
+        return []
+    return [p for p in data if isinstance(p, dict)]
+
+
+def build_swarm_services(containers, services, node_hostname: dict, host_name: dict) -> dict:
+    """Return {"services": [...], "standalone": [...]}.
+
+    services[i] = {name, mode, desired, running, state, replicas:[...]}
+      real replica  = {ghost:False, host, host_id, name, status, cpu_pct, mem_pct,
+                       health, restart_count}
+      ghost replica = {ghost:True, host, count}   # placement with no local row
+    standalone[i] = {host_id, host, containers:[...]}  # non-swarm, by host
+    """
+    containers = list(containers)
+    swarm_cs = [c for c in containers if getattr(c, "service_name", None)]
+    standalone_cs = [c for c in containers if not getattr(c, "service_name", None)]
+
+    # Dedup service rows by name (every manager reports the same cluster-global
+    # set); keep the first — placement is identical across managers.
+    svc_by_name: dict[str, object] = {}
+    for s in services:
+        name = getattr(s, "service_name", None)
+        if name and name not in svc_by_name:
+            svc_by_name[name] = s
+
+    # Real containers grouped by service name.
+    cs_by_service: dict[str, list] = {}
+    for c in swarm_cs:
+        cs_by_service.setdefault(c.service_name, []).append(c)
+
+    def _host_of(c):
+        return (host_name.get(getattr(c, "host_id", None))
+                or node_hostname.get(getattr(c, "node_id", None))
+                or getattr(c, "host_id", None) or "?")
+
+    out_services = []
+    for name in sorted(set(svc_by_name) | set(cs_by_service)):
+        svc = svc_by_name.get(name)
+        reals = cs_by_service.get(name, [])
+
+        replicas = []
+        for c in sorted(reals, key=lambda c: (_host_of(c).lower(), c.name)):
+            replicas.append({
+                "ghost": False,
+                "host": _host_of(c),
+                "host_id": getattr(c, "host_id", None),
+                "name": c.name,
+                "status": getattr(c, "status", "unknown"),
+                "cpu_pct": getattr(c, "cpu_pct", None),
+                "mem_pct": getattr(c, "mem_pct", None),
+                "health": getattr(c, "health", None),
+                "restart_count": getattr(c, "restart_count", 0) or 0,
+            })
+
+        # Real running replicas per node, to subtract from placement.
+        real_running_by_node: dict[str, int] = {}
+        for c in reals:
+            if (getattr(c, "status", "") or "").lower() == "running":
+                nid = getattr(c, "node_id", None)
+                if nid:
+                    real_running_by_node[nid] = real_running_by_node.get(nid, 0) + 1
+
+        ghosts = []
+        for p in _placement(svc) if svc is not None else []:
+            nid = p.get("node_id", "")
+            placed = int(p.get("running", 0) or 0)
+            ghost = placed - real_running_by_node.get(nid, 0)
+            if ghost > 0:
+                ghosts.append({
+                    "ghost": True,
+                    "host": node_hostname.get(nid) or (nid[:12] if nid else "?"),
+                    "count": ghost,
+                })
+        ghosts.sort(key=lambda g: g["host"].lower())
+        replicas.extend(ghosts)
+
+        running = getattr(svc, "running", None) if svc is not None else None
+        desired = getattr(svc, "desired", None) if svc is not None else None
+        # No service row (manager didn't report it): infer from what we can see.
+        if running is None:
+            running = sum(1 for r in replicas
+                          if (not r["ghost"] and (r["status"] or "").lower() == "running"))
+        if desired is None:
+            desired = len(replicas)
+
+        out_services.append({
+            "name": name,
+            "mode": getattr(svc, "mode", "replicated") if svc is not None else "replicated",
+            "running": running,
+            "desired": desired,
+            "state": _service_state(int(running or 0), int(desired or 0)),
+            "replicas": replicas,
+        })
+
+    # Non-swarm containers grouped by host, running-first order preserved.
+    groups: dict[str, dict] = {}
+    for c in standalone_cs:
+        hid = getattr(c, "host_id", None)
+        g = groups.get(hid)
+        if g is None:
+            g = groups[hid] = {"host_id": hid, "host": host_name.get(hid) or hid or "?",
+                               "containers": []}
+        g["containers"].append(c)
+    standalone = sorted(groups.values(), key=lambda g: g["host"].lower())
+
+    return {"services": out_services, "standalone": standalone}

plugins/docker/templates/docker/rows.html

@@ -14,12 +14,61 @@
     <div class="section-title" style="margin-bottom:0.35rem;">Total</div>
     <span class="stat-val">{{ total }}</span>
   </div>
+  {% if swarm_services %}
+  <div class="card" style="margin-bottom:0;">
+    <div class="section-title" style="margin-bottom:0.35rem;">Services</div>
+    <span class="stat-val">{{ swarm_services | length }}</span>
+  </div>
+  {% endif %}
   <div class="card" style="margin-bottom:0;">
     <div class="section-title" style="margin-bottom:0.35rem;">Hosts</div>
     <span class="stat-val">{{ host_groups | length }}</span>
   </div>
 </div>
 
+{# ── Swarm services: collapsed per-service, each replica labelled with its host.
+   Ghost replicas (tasks on nodes with no Steward agent) come from the managers'
+   placement data so the picture is cluster-complete. ──────────────────────── #}
+{% if swarm_services %}
+<div class="card-flush" style="margin-bottom:1.5rem;">
+  <div style="display:flex;align-items:baseline;gap:0.6rem;padding:0.5rem 0.75rem;border-bottom:1px solid var(--border);">
+    <span style="font-weight:600;font-size:0.95rem;">Swarm services</span>
+    <a href="/plugins/docker/swarm" style="font-size:0.78rem;color:var(--text-muted);">Topology →</a>
+  </div>
+  <div style="padding:0.5rem 0.75rem;display:grid;gap:0.85rem;">
+    {% for s in swarm_services %}
+    <div>
+      <div style="display:flex;align-items:center;gap:0.55rem;flex-wrap:wrap;margin-bottom:0.35rem;">
+        <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;">
+          {{ s.running }}/{{ s.desired }} running
+        </span>
+      </div>
+      <div style="display:grid;grid-template-columns:repeat(auto-fill,minmax(210px,1fr));gap:0.4rem;padding-left:1.1rem;">
+        {% for r in s.replicas %}
+        {% if r.ghost %}
+        <div style="display:flex;align-items:center;gap:0.4rem;font-size:0.8rem;background:var(--bg-elevated);border:1px dashed var(--border-mid);border-radius:5px;padding:0.3rem 0.55rem;" title="Running on a node with no Steward agent — detail unavailable">
+          <span class="dot dot-warn" style="opacity:0.6;"></span>
+          <span style="font-weight:500;">{{ r.host }}</span>
+          <span style="color:var(--text-dim);margin-left:auto;">{{ r.count }} · no agent</span>
+        </div>
+        {% else %}
+        <div style="display:flex;align-items:center;gap:0.4rem;font-size:0.8rem;background:var(--bg-elevated);border:1px solid var(--border);border-radius:5px;padding:0.3rem 0.55rem;">
+          <span class="dot {% if r.status == 'running' %}dot-up{% elif r.status == 'paused' %}dot-warn{% else %}dot-down{% endif %}"></span>
+          <a href="/plugins/docker/container/{{ r.host_id }}/{{ r.name }}" style="font-weight:500;color:inherit;text-decoration:none;white-space:nowrap;overflow:hidden;text-overflow:ellipsis;" title="{{ r.name }}">{{ r.host }}</a>
+          {% if r.cpu_pct is not none %}<span style="color:var(--text-muted);font-family:ui-monospace,monospace;margin-left:auto;">{{ "%.0f" | format(r.cpu_pct) }}%</span>{% endif %}
+        </div>
+        {% endif %}
+        {% endfor %}
+      </div>
+    </div>
+    {% endfor %}
+  </div>
+</div>
+{% endif %}
+
 {# ── Container tables, one per host ───────────────────────────────────────── #}
 {% if host_groups %}
 {% for g in host_groups %}
@@ -114,7 +163,7 @@
   </table>
 </div>
 {% endfor %}
-{% else %}
+{% elif not swarm_services %}
 <div class="card" style="text-align:center;padding:3rem;">
   <div style="color:var(--text-muted);font-size:0.9rem;">
     No containers reported yet. Containers are collected by the Steward host agent —

plugins/docker/templates/docker/widget.html

@@ -14,6 +14,32 @@
   </div>
 </div>
 
+{# Swarm services — collapsed, each with its replicas' host chips (dashed = a
+   node with no Steward agent, count-only from placement). #}
+{% if swarm_services %}
+<div style="font-size:0.72rem;color:var(--text-dim);text-transform:uppercase;letter-spacing:0.05em;margin:0.5rem 0 0.2rem;">Swarm services</div>
+<div style="display:grid;gap:0.4rem;margin-bottom:0.4rem;">
+  {% for s in swarm_services %}
+  <div>
+    <div style="display:flex;align-items:center;gap:0.4rem;font-size:0.82rem;">
+      <span class="dot {% if s.state == 'healthy' %}dot-up{% elif s.state == 'degraded' %}dot-warn{% else %}dot-down{% endif %}"></span>
+      <span style="font-weight:500;min-width:0;white-space:nowrap;overflow:hidden;text-overflow:ellipsis;">{{ s.name }}</span>
+      <span style="color:var(--text-muted);font-family:ui-monospace,monospace;font-size:0.74rem;margin-left:auto;flex-shrink:0;">{{ s.running }}/{{ s.desired }}</span>
+    </div>
+    <div style="display:flex;flex-wrap:wrap;gap:0.25rem;margin:0.2rem 0 0 1rem;">
+      {% for r in s.replicas %}
+      {% if r.ghost %}
+      <span title="no Steward agent on this node — count from swarm placement" style="font-size:0.68rem;color:var(--text-dim);background:var(--bg-elevated);border:1px dashed var(--border-mid);border-radius:4px;padding:0.05rem 0.35rem;">{{ r.host }} ×{{ r.count }}</span>
+      {% else %}
+      <span style="font-size:0.68rem;color:var(--text-muted);background:var(--bg-elevated);border:1px solid var(--border);border-radius:4px;padding:0.05rem 0.35rem;">{{ r.host }}</span>
+      {% endif %}
+      {% endfor %}
+    </div>
+  </div>
+  {% endfor %}
+</div>
+{% endif %}
+
 {% for g in host_groups %}
 {% if multi_host %}
 <div style="font-size:0.72rem;color:var(--text-dim);text-transform:uppercase;letter-spacing:0.05em;margin:0.5rem 0 0.2rem;">{{ g.host_name }}</div>