feat(docker): Docker's own status reaches Steward — an empty answer clears a host, a lost one greys it, a demoted manager's swarm rows go (#5152)

The agent (1.9.0) said nothing when Docker could not be asked and nothing
when Docker had no containers, so Steward could tell neither apart from the
other, and kept a host's last containers on screen as running either way.

- Agent: collect_docker returns (status, detail, containers). "ok" sends
  docker: [] when there are none; "unavailable" (no socket, nothing
  listening) and "error" (permission, timeout, bad answer) say why.
  collect_swarm says whether /info answered, and a node that is not a
  manager sends swarm_manager: false. A failed /info says nothing.
- Ingest: an empty docker list is a snapshot; the newest sample's status
  goes to the Docker plugin with the rest.
- docker_host_status (docker_011): per host, status, detail, since and
  last report. A host that never had Docker gets no row.
- An "ok" empty snapshot reaps the host's containers; an error keeps them.
  swarm_manager: false drops the host's swarm services and nodes.
- Docker page: a "Docker not answering" card names each such host, why,
  and since when; its containers read as stale. Duplicate header comment
  in rows.html removed.

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

plugins/docker/routes.py

@@ -18,7 +18,7 @@ from . import service as docker_svc
 from .dedup import dedup_by_container_id
 from .swarm_view import build_swarm_services, parse_tasks, task_summary
 from .models import (
-    DockerContainer, DockerDiskUsage, DockerEvent, DockerImage,
+    DockerContainer, DockerDiskUsage, DockerEvent, DockerHostStatus, DockerImage,
     DockerSwarmNode, DockerSwarmService,
 )
 
@@ -87,6 +87,9 @@ async def rows():
     stale_after = docker_svc.stale_after_seconds(current_app.config)
 
     async with current_app.db_sessionmaker() as db:
+        # Hosts whose Docker is not answering: their rows are what it last said.
+        docker_down = {d.host_id: d for d in (await db.execute(
+            select(DockerHostStatus).where(DockerHostStatus.status != "ok"))).scalars()}
         ranked = await docker_svc.ranked_containers(
             db, now, docker_svc.crash_loop_deaths(current_app.config))
         containers = await docker_svc.list_containers(db)
@@ -95,7 +98,7 @@ async def rows():
         hosts = await docker_svc.host_map(
             db,
             {c.host_id for c in containers}
-            | {n.host_id for n in nodes} | {s.host_id for s in services},
+            | {n.host_id for n in nodes} | {s.host_id for s in services} | set(docker_down),
         )
 
         # Swarm tasks collapse into per-service rows (each replica labelled with
@@ -123,7 +126,7 @@ async def rows():
                 "container": c,
                 "ports": json.loads(c.ports_json) if c.ports_json else [],
                 "uptime": _uptime(c.started_at) if c.status == "running" else None,
-                "stale": _is_stale(c, now, stale_after),
+                "stale": _is_stale(c, now, stale_after, docker_down),
             })
             if c.status == "running":
                 g["running"] += 1
@@ -151,12 +154,16 @@ 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))
+    attention = [dict(r, stale=_is_stale(r["container"], now, stale_after, docker_down))
                  for r in docker_svc.needs_attention(ranked)]
     return await render_template(
         "docker/rows.html",
         host_groups=host_groups,
         swarm_services=swarm_services,
+        docker_down=sorted(
+            ({"host_id": h, "host_name": hosts[h].name if h in hosts else h, "status": d.status,
+              "detail": d.detail, "since": d.since} for h, d in docker_down.items()),
+            key=lambda d: d["host_name"].lower()),
         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),
@@ -167,10 +174,12 @@ async def rows():
     )
 
 
-def _is_stale(c, now: datetime, stale_after: int) -> bool:
+def _is_stale(c, now: datetime, stale_after: int, docker_down=()) -> 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
+    window, or whose host's Docker is not answering: what the page shows may
+    no longer be true."""
+    return (c.host_id in docker_down or c.scraped_at is None
+            or (now - c.scraped_at).total_seconds() > stale_after)
 
 
 @docker_bp.get("/widget")