geenforce: a PC that has gone quiet stops reading as healthy

A report records how ONE cycle went. Nothing ages it, so a PC that stops
reporting keeps the status of its last good cycle: switch a machine off after a
clean run and it shows 'ok' indefinitely. One had been offline more than a day
and still read 'ok'.

Silence is a different fact from the last cycle's outcome, so it is computed
separately rather than by rewriting the stored status. receivedat - the server's
own clock, not anything a client asserts - is compared against
geenforce_reportstaleminutes, default 30, which is roughly two missed cycles at
the usual cadence. Set it to 0 to turn the check off.

In the table 'stale' takes the badge, because a status from a machine that has
not spoken since is not evidence of anything. What it last reported stays in the
tooltip with the time it was heard. The stored status is untouched: it is still
a true record of that cycle, just not proof the PC is alive.

A site whose scope enforces less often than the threshold will read stale while
healthy, which is what the setting is for.

plugins/geenforce/api/routes.py

@@ -11,7 +11,7 @@ Two audiences:
 import ipaddress
 import os
 import time
-from datetime import datetime
+from datetime import datetime, timedelta
 from functools import wraps
 
 from flask import Blueprint, request, Response, send_file, current_app, g
@@ -1061,6 +1061,29 @@ def _attach_backup_state(facts, hostnames):
                 if revision.lastseenat else None)
 
 
+def _report_stale_cutoff():
+    """(cutoff datetime, threshold minutes) for 'this PC has gone quiet'.
+
+    A report is a record of one cycle, not a heartbeat with an expiry: nothing
+    ages it. So a PC that stops reporting keeps the status of its last good
+    cycle and reads as healthy while it is unplugged. Comparing receivedat - the
+    SERVER's clock, not anything the client asserts - against a threshold is
+    what turns silence into a visible state.
+
+    Returns (None, 0) when the check is disabled, so callers skip it entirely.
+    """
+    from shopdb.api import Setting
+    try:
+        minutes = int(Setting.get('geenforce_reportstaleminutes', 30) or 0)
+    except (TypeError, ValueError):
+        minutes = 30
+    if minutes <= 0:
+        return None, 0
+    # Same clock helper that WROTE receivedat (naive UTC), so the two cannot
+    # drift into comparing an aware datetime against a naive one.
+    return service._utcnow() - timedelta(minutes=minutes), minutes
+
+
 @geenforce_bp.route('/reports', methods=['GET'])
 @jwt_required()
 @require_permission('geenforce.manage')
@@ -1084,6 +1107,7 @@ def list_reports():
     facts = _asset_facts({report.hostname for report in reports
                           if report.hostname})
 
+    stalecutoff, stalafter = _report_stale_cutoff()
     latest_cache = {}
     data = []
     for report in reports:
@@ -1092,6 +1116,9 @@ def list_reports():
             latest_cache[key] = _current_published_version(*key)
         latest = latest_cache[key]
         known = facts.get((report.hostname or '').lower(), {})
+        isstale = (stalecutoff is not None
+                   and (report.receivedat is None
+                        or report.receivedat < stalecutoff))
         data.append({
             'reportid': report.reportid,
             'hostname': report.hostname,
@@ -1118,7 +1145,14 @@ def list_reports():
             'receivedlatest': (latest is not None
                                and report.appliedversion == latest),
             'enforcerversion': report.enforcerversion,
+            # 'status' is what the PC said about its LAST cycle and is left
+            # exactly as reported. 'isstale' is the server's own judgement that
+            # no cycle has been heard from in too long - a PC switched off after
+            # a clean run reports 'ok' forever, so the two are different facts
+            # and the table shows both.
             'status': report.status,
+            'isstale': isstale,
+            'staleafterminutes': stalafter,
             'installed': report.installedcount,
             'skipped': report.skippedcount,
             'failed': report.failedcount,