feat(host_agent): each drive's health, read with smartctl, alertable (#5789)

The root helper gains one read-only action, `drives`: it finds the drives
itself (smartctl --scan-open), reads each as JSON, and writes a summary the
agent carries on every sample (agent 1.20.0). Steward judges the figures in
drives.py (failing / worn / watch / ok / no SMART / unreadable), records them
per drive and as host roll-ups for covering rules, keeps the newest report on
the registration and lists it on the host page.

install-deps.sh installs smartmontools with apt, dnf, yum, zypper, pacman or
apk, run by install.sh and every agent playbook; one it can't install is
reported, never fatal.

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

plugins/host_agent/routes.py

@@ -27,7 +27,7 @@ from steward.hosts import service as host_svc
 from steward.models.hosts import Host
 from steward.models.metrics import PluginMetric
 from .downtime import recent as recent_downtime, record_events
-from . import maintenance
+from . import drives as drive_health, maintenance
 from .models import HostAgentRegistration
 from .scheduler import arrivals
 # Query helpers live in a model-free module so integration tests can import them
@@ -193,6 +193,16 @@ def apply_updates(reg, sample: dict, now: datetime) -> bool:
                       reg.reboot_required_since, reg.reboot_packages)
 
 
+def apply_drives(reg, sample: dict, now: datetime) -> bool:
+    """Keep the registration's copy of the host's newest drive report (#5789),
+    which the host page lists: model, state and why, beside the figures."""
+    report = sample.get("drives")
+    if not isinstance(report, dict) or report == reg.drive_health:
+        return False
+    reg.drive_health, reg.drive_health_at = report, _ts_or(report.get("checked_at"), now)
+    return True
+
+
 def _expand_sample_to_metrics(
     sample: dict, host_name: str, recorded_at: datetime
 ) -> list[PluginMetric]:
@@ -331,6 +341,10 @@ def _expand_sample_to_metrics(
     if max_temp is not None:
         row("temp_c_max", host_name, max_temp)
 
+    # Each drive's health (#5789): the helper's hourly read rides every sample.
+    for metric, resource, value in drive_health.metric_rows(sample.get("drives"), host_name):
+        row(metric, resource, value)
+
     # Pressure stall information (PSI) — host-level gauges (mem/cpu/io some|full).
     for k, v in (sample.get("psi") or {}).items():
         if isinstance(v, (int, float)):
@@ -527,6 +541,8 @@ async def ingest():
                 changed = True
             if newest_sample is not None and apply_updates(reg, newest_sample, received_at):
                 changed = True
+            if newest_sample is not None and apply_drives(reg, newest_sample, received_at):
+                changed = True
             # Mirror the reported IP into Host.address only when it's blank.
             new_address = pick_host_address(host.address, host_ip)
             if new_address is not None:
@@ -685,6 +701,7 @@ async def agent_source():
 AGENT_FILES_DIR = Path(__file__).parent / "files"
 AGENT_FILES = {
     "maint.py": "text/x-python",
+    "install-deps.sh": "text/x-shellscript",
     "steward-agent.service": "text/plain",
     "steward-maint.service": "text/plain",
     "steward-maint.path": "text/plain",
@@ -842,6 +859,8 @@ def _split_host_metrics(host_name: str, latest: dict) -> tuple:
             disks_io[suffix[len("diskio:"):]] = metrics
         elif suffix.startswith("temp:"):
             temps[suffix[len("temp:"):]] = metrics.get("temp_c")
+        elif suffix.startswith("drive:"):
+            continue                        # listed from the drive report (#5789)
         else:
             mounts[suffix] = metrics
     cores.sort(key=lambda c: c[0])
@@ -882,12 +901,15 @@ async def host_detail_metrics(host_id: str):
         latest = await _latest_metrics_for_host(session, host.name)
     hostlvl, cores, nets, disks_io, temps, mounts = _split_host_metrics(host.name, latest)
     temps, unwired = split_temps(host.name, temps)
+    drive_report = reg.drive_health if reg else None
     ls = reg.last_seen_at if reg else None
     stale = ls is None or (datetime.now(timezone.utc) - ls).total_seconds() > _stale_after_seconds()
     return await render_template(
         "_host_metrics.html",
         host=host, reg=reg, stale=stale, hostlvl=hostlvl,
         cores=cores, nets=nets, disks_io=disks_io, temps=temps, unwired_temps=unwired, mounts=mounts,
+        drives=drive_health.drives_of(drive_report), drives_error=(drive_report or {}).get("error"),
+        drives_at=reg.drive_health_at if reg else None,
     )