Fix T6 occupancy parsing in host agent

agent/ec_agent.py

@@ -30,7 +30,7 @@ import urllib.parse
 import urllib.request
 import uuid
 
-VERSION = "1.4.0"
+VERSION = "1.4.1"
 LABEL = "io.electric-cherry"
 MAX_FILE = 32 * 1024 * 1024
 LIBRARY_DOWNLOAD_WORKERS = 4
@@ -169,7 +169,8 @@ def file_digest(path):
 def status_players(text):
     lines = text.splitlines()
     separator = next((index for index, line in enumerate(lines) if re.match(r"^\s*-+\s+-+", line)), None)
-    if separator is not None and any(re.search(r"\bnum\s+score\s+ping\b", line, re.I) for line in lines[:separator]):
+    # T6 includes a bot column; T5 places ping directly after score.
+    if separator is not None and any(re.search(r"\bnum\s+score\s+(?:bot\s+)?ping\b", line, re.I) for line in lines[:separator]):
         rows = [line.strip() for line in lines[separator + 1:] if line.strip()]
         if any(not re.match(r"^\d+\s+", row) for row in rows):
             return None
@@ -1013,7 +1014,16 @@ class Agent:
             raise Held("Container state is unknown; disruption held")
         result = self.rcon(instance_id, generation, "status")
         if not result.get("ok") or result.get("players") is None:
-            raise Held("Fresh local player occupancy is unavailable; disruption held")
+            # Keep diagnostics bounded and do not include raw replies: they may
+            # contain player addresses or echoed credentials.
+            reason = {
+                "unconfigured": "local RCON password is not configured",
+                "unreachable": "local RCON endpoint is unreachable",
+                "invalid-reply": "local RCON reply has an invalid packet prefix",
+                "timeout": "local RCON status query timed out",
+                "authentication": "local RCON authentication failed",
+            }.get(result.get("reason"), "local RCON status reply has no recognized player count" if result.get("ok") else "local RCON query failed")
+            raise Held(f"Fresh local player occupancy is unavailable: {reason}; disruption held")
         if result["players"] != 0:
             raise Held("Players are present; disruption held")
 

agent/tests/test_agent.py

@@ -153,6 +153,28 @@ class AgentTests(unittest.TestCase):
         self.agent.rcon = lambda *_: {"ok": True, "players": 0}
         self.assertEqual(self.agent.execute(operation)["state"], "completed")
 
+    def test_occupancy_hold_explains_local_rcon_failure_without_exposing_reply(self):
+        self.create()
+        operation = self.operation("instance.restart")
+        for reason, detail in (
+            ("unconfigured", "password is not configured"),
+            ("unreachable", "endpoint is unreachable"),
+            ("invalid-reply", "invalid packet prefix"),
+            ("timeout", "query timed out"),
+            ("authentication", "authentication failed"),
+            (None, "no recognized player count"),
+        ):
+            with self.subTest(reason=reason):
+                self.agent.rcon = lambda *_: {"ok": reason is None, "reason": reason, "players": None, "text": "sensitive raw reply"}
+                result = self.agent.execute(operation)
+                self.assertEqual(result["state"], "accepted")
+                message = json.loads(result["result"])["message"]
+                self.assertIn(detail, message)
+                self.assertNotIn("sensitive raw reply", message)
+                self.assertFalse(any(call[:2] == ("container", "restart") for call in self.docker.calls))
+        self.agent.rcon = lambda *_: {"ok": True, "players": 0}
+        self.assertEqual(self.agent.execute(operation)["state"], "completed")
+
     def test_unknown_container_health_cannot_disrupt(self):
         self.create()
         self.docker.containers[self.agent.container_name(INSTANCE, 1)]["State"] = {"Status": "unknown", "Running": False}
@@ -896,6 +918,26 @@ class AgentTests(unittest.TestCase):
         self.assertEqual(result["players"], 1)
         self.assertTrue(result["ok"])
 
+    def test_t6_bot_column_allows_empty_restart_but_holds_players_and_partial_replies(self):
+        self.create()
+        header = "map: zm_nuked\nnum score bot ping guid                             name             lastmsg address               qport  rate\n"
+        table = header + "--- ----- --- ---- -------------------------------- ---------------- ------- --------------------- ------ -----\n"
+        operation = self.operation("instance.restart")
+        for body, players in (
+            (header, None),
+            (table + "0 0 0 CNCT abc player 0 127.0.0.1:28960 123 25000\n", 1),
+            (table + "0 0 1 0 abc bot 0 loopback 0 25000\n", 1),
+            (table + "unrecognized trailing output", None),
+            (table, 0),
+        ):
+            with self.subTest(body=body):
+                self.assertEqual(module.status_players(body), players)
+                self.agent.rcon = lambda *_: {"ok": True, "players": module.status_players(body)}
+                result = self.agent.execute(operation)
+                self.assertEqual(result["state"], "completed" if players == 0 else "accepted")
+                if players != 0:
+                    self.assertFalse(any(call[:2] == ("container", "restart") for call in self.docker.calls))
+
     def test_invalid_rebuild_preflight_preserves_running_original(self):
         payload = self.create()
         result = self.agent.execute(self.operation("instance.rebuild", {**payload, "image": "--privileged", "nextGeneration": 2}))

docs/native-hosts.md

@@ -216,6 +216,16 @@ wait indefinitely, including multi-hour Zombies sessions. Before images returned
 by the agent populate existing campaign/beta rollback history. A beta cannot be
 closed while its remote activation receipt is still unresolved.
 
+An empty Fleet card does not replace this final host-local check: the card may
+use telemetry, and the controller's RCON parser is more permissive than the
+agent's disruption guard. Agent 1.4.1 includes the local RCON failure reason in
+the operation's hold message (configuration, timeout, unreachable endpoint,
+authentication, invalid packet, or an unrecognized player count). For a parser
+hold, inspect the raw `status` reply in **Diagnose connection**. Unknown counts
+remain held; RCON availability alone does not prove the lobby is empty.
+It also recognizes T6's `num score bot ping` header, which earlier agents
+rejected even when the complete player table was empty.
+
 Rebuild uses a new generation, fresh instance storage and the latest promoted
 configuration/scripts/plugins. It retains the selected game-file version and
 runtime image. Old generation commands cannot target the replacement. Start,

static/agent/ec-agent.py

@@ -30,7 +30,7 @@ import urllib.parse
 import urllib.request
 import uuid
 
-VERSION = "1.4.0"
+VERSION = "1.4.1"
 LABEL = "io.electric-cherry"
 MAX_FILE = 32 * 1024 * 1024
 LIBRARY_DOWNLOAD_WORKERS = 4
@@ -169,7 +169,8 @@ def file_digest(path):
 def status_players(text):
     lines = text.splitlines()
     separator = next((index for index, line in enumerate(lines) if re.match(r"^\s*-+\s+-+", line)), None)
-    if separator is not None and any(re.search(r"\bnum\s+score\s+ping\b", line, re.I) for line in lines[:separator]):
+    # T6 includes a bot column; T5 places ping directly after score.
+    if separator is not None and any(re.search(r"\bnum\s+score\s+(?:bot\s+)?ping\b", line, re.I) for line in lines[:separator]):
         rows = [line.strip() for line in lines[separator + 1:] if line.strip()]
         if any(not re.match(r"^\d+\s+", row) for row in rows):
             return None
@@ -1013,7 +1014,16 @@ class Agent:
             raise Held("Container state is unknown; disruption held")
         result = self.rcon(instance_id, generation, "status")
         if not result.get("ok") or result.get("players") is None:
-            raise Held("Fresh local player occupancy is unavailable; disruption held")
+            # Keep diagnostics bounded and do not include raw replies: they may
+            # contain player addresses or echoed credentials.
+            reason = {
+                "unconfigured": "local RCON password is not configured",
+                "unreachable": "local RCON endpoint is unreachable",
+                "invalid-reply": "local RCON reply has an invalid packet prefix",
+                "timeout": "local RCON status query timed out",
+                "authentication": "local RCON authentication failed",
+            }.get(result.get("reason"), "local RCON status reply has no recognized player count" if result.get("ok") else "local RCON query failed")
+            raise Held(f"Fresh local player occupancy is unavailable: {reason}; disruption held")
         if result["players"] != 0:
             raise Held("Players are present; disruption held")
 

static/agent/manifest.json

@@ -1,8 +1,8 @@
 {
-  "version": "1.4.0",
+  "version": "1.4.1",
   "protocolVersion": 1,
   "platform": "debian-amd64",
-  "sha256": "8bb6bf3c433def0540cf0de11304f23820c9afcaf1d90d8cdaa7a2a09a455f2d",
+  "sha256": "0691d8457f52ef8b53691b627733352b2f187529a4ecc465cdaaa6e27ddaaaa7",
   "updaterProtocolVersion": 1,
   "updaterSha256": "066bb896726600ece9ce72697ce523c9a707cc9a621505931ba0892dec59d3c8"
 }