Add MQTT diagnostics, broker connection probe, and switch to Debian base

The MQTT publisher now tracks connection state, publish counters, discovery
and availability timestamps, and the last error, exposed at /api/mqtt and on
the dashboard as an MQTT card and a Home Assistant section. Connection
failures, disconnects, and dropped publishes are logged; paho's
on_connect_fail callback was not registered before, so failed attempts were
silent. MQTT connect, disconnect, and unreachable events go to the timeline.

When a connection attempt fails the service runs a probe from inside the
container: system resolver, each nameserver from resolv.conf queried directly,
/etc/hosts, and a TCP connect to every address found. The result is shown as
a summary and raw JSON on the dashboard and can be rerun via
POST /api/mqtt/probe.

The image base moves from Alpine to Debian slim. musl queries all
nameservers in parallel and accepts the first reply, so a fast public
NXDOMAIN beats a slower local server that knows the name. glibc asks the
nameservers in order.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
Milan PandurovandClaude Fable 5.1 committed 2026-09-29 12:08:48 +02:00
1 parent d80c40065e
commit 62b57a7e56
8 files changed
+737 -14

No files matched your search

+140
View File
@@ -0,0 +1,140 @@
import time
import unittest
from types import SimpleNamespace
from unittest import mock
import paho.mqtt.client as mqtt
import hamqtt
def make_cfg(**overrides):
values = dict(device_id="pi", device_name="Pi", mqtt_host="broker", mqtt_port=1883,
mqtt_username="", mqtt_password="", mqtt_discovery_prefix="homeassistant",
wifi_interface="")
values.update(overrides)
return SimpleNamespace(**values)
class FakeStorage:
def __init__(self):
self.events = []
def add_event(self, kind, message):
self.events.append((kind, message))
def ok_publish(*args, **kwargs):
return SimpleNamespace(rc=mqtt.MQTT_ERR_SUCCESS)
def no_conn_publish(*args, **kwargs):
return SimpleNamespace(rc=mqtt.MQTT_ERR_NO_CONN)
class PublisherDiagnosticsTest(unittest.TestCase):
def setUp(self):
patcher = mock.patch("hamqtt.mqtt.Client")
self.client_cls = patcher.start()
self.addCleanup(patcher.stop)
self.client = self.client_cls.return_value
self.client.publish.side_effect = ok_publish
self.storage = FakeStorage()
self.publisher = hamqtt.HomeAssistantPublisher(make_cfg(), self.storage)
def test_initial_state(self):
diag = self.publisher.diagnostics()
self.assertTrue(diag["enabled"])
self.assertEqual(diag["status"], "not_started")
self.assertFalse(diag["connected"])
self.assertEqual(diag["host"], "broker")
self.assertEqual(diag["client_id"], "healthcheck_pi")
self.assertEqual(diag["state_topic"], "healthcheck/pi/state")
self.assertIsNone(diag["username"])
def test_start_marks_connecting(self):
self.publisher.start()
self.assertEqual(self.publisher.diagnostics()["status"], "connecting")
self.assertEqual(self.client.on_connect_fail, self.publisher._on_connect_fail)
def test_connect_announces_and_records(self):
self.publisher.publish_state({"internet_up": True})
self.publisher._on_connect(self.client, None, {}, SimpleNamespace(is_failure=False), None)
diag = self.publisher.diagnostics()
self.assertEqual(diag["status"], "connected")
self.assertTrue(diag["connected"])
self.assertIsNotNone(diag["connected_at"])
self.assertEqual(diag["discovery_count"], 1)
self.assertIsNotNone(diag["availability_at"])
self.assertEqual(diag["publish_dropped"], 0)
self.assertGreater(diag["publish_count"], len(hamqtt.SENSORS))
self.assertEqual(diag["last_payload"]["internet_up"], True)
self.assertEqual(self.storage.events[0][0], "mqtt_connected")
def test_connect_fail_records_first_event_only(self):
self.publisher._on_connect_fail(self.client, None)
self.publisher._on_connect_fail(self.client, None)
diag = self.publisher.diagnostics()
self.assertEqual(diag["status"], "failed")
self.assertEqual(diag["connect_failures"], 2)
self.assertIn("failed", diag["last_error"])
self.assertEqual([k for k, _ in self.storage.events], ["mqtt_connect_failed"])
def test_refused(self):
self.publisher._on_connect(self.client, None, {}, SimpleNamespace(is_failure=True), None)
diag = self.publisher.diagnostics()
self.assertEqual(diag["status"], "refused")
self.assertEqual(diag["connect_failures"], 1)
self.assertFalse(diag["connected"])
def test_disconnect(self):
self.publisher._on_connect(self.client, None, {}, SimpleNamespace(is_failure=False), None)
self.publisher._on_disconnect(self.client, None, {}, "Keep alive timeout", None)
diag = self.publisher.diagnostics()
self.assertEqual(diag["status"], "disconnected")
self.assertEqual(diag["disconnects"], 1)
self.assertIn("Keep alive timeout", diag["last_error"])
self.assertEqual(self.storage.events[-1][0], "mqtt_disconnected")
def test_dropped_publish_is_counted(self):
self.client.publish.side_effect = no_conn_publish
self.publisher.publish_state({"internet_up": False})
diag = self.publisher.diagnostics()
self.assertEqual(diag["publish_count"], 0)
self.assertEqual(diag["publish_dropped"], 1)
self.assertIsNotNone(diag["last_publish_error"])
self.assertIsNone(diag["last_publish_at"])
def test_ha_status_message(self):
message = SimpleNamespace(topic="homeassistant/status", payload=b"online")
self.publisher._on_message(self.client, None, message)
diag = self.publisher.diagnostics()
self.assertEqual(diag["ha_status"], "online")
self.assertEqual(diag["discovery_count"], 1)
def test_connect_fail_runs_probe_once_per_interval(self):
probe = {"ts": time.time(), "summary": ["System resolver failed"]}
with mock.patch("hamqtt.netprobe.probe", return_value=probe) as probe_fn, \
mock.patch("hamqtt.threading.Thread") as thread_cls:
thread_cls.return_value.start.side_effect = lambda: self.publisher.run_probe()
self.publisher._on_connect_fail(self.client, None)
self.publisher._on_connect_fail(self.client, None)
probe_fn.assert_called_once_with("broker", 1883)
diag = self.publisher.diagnostics()
self.assertEqual(diag["probe"], probe)
self.assertFalse(diag["probe_running"])
def test_run_probe_survives_crash(self):
with mock.patch("hamqtt.netprobe.probe", side_effect=RuntimeError("boom")):
result = self.publisher.run_probe()
self.assertIn("boom", result["error"])
self.assertFalse(self.publisher.diagnostics()["probe_running"])
def test_publish_state_dedupes(self):
self.publisher.publish_state({"internet_up": True})
self.publisher.publish_state({"internet_up": True})
self.assertEqual(self.publisher.diagnostics()["publish_count"], 1)
if __name__ == "__main__":
unittest.main()
+125
View File
@@ -0,0 +1,125 @@
import os
import socket
import struct
import tempfile
import threading
import unittest
import netprobe
def answer(ident, name, rcode=0, addresses=(), cname=None):
header = struct.pack(">HHHHHH", ident, 0x8180 | rcode, 1, len(addresses) + (1 if cname else 0), 0, 0)
question = netprobe.build_query(name, ident)[12:]
records = b""
if cname:
target = b"".join(bytes([len(l)]) + l.encode() for l in cname.split(".")) + b"\x00"
records += b"\xc0\x0c" + struct.pack(">HHIH", 5, 1, 60, len(target)) + target
for address in addresses:
records += b"\xc0\x0c" + struct.pack(">HHIH", 1, 1, 60, 4) + socket.inet_aton(address)
return header + question + records
class ParseResponseTest(unittest.TestCase):
def test_addresses(self):
data = answer(7, "broker.lan", addresses=["192.168.0.10", "192.168.0.11"])
parsed = netprobe.parse_response(data, 7)
self.assertEqual(parsed["rcode"], "NOERROR")
self.assertEqual(parsed["addresses"], ["192.168.0.10", "192.168.0.11"])
def test_nxdomain(self):
parsed = netprobe.parse_response(answer(1, "nope.invalid", rcode=3), 1)
self.assertEqual(parsed["rcode"], "NXDOMAIN")
self.assertEqual(parsed["addresses"], [])
def test_cname(self):
parsed = netprobe.parse_response(answer(2, "ha.example", cname="real.example", addresses=["10.0.0.1"]), 2)
self.assertEqual(parsed["cnames"], ["real.example"])
self.assertEqual(parsed["addresses"], ["10.0.0.1"])
def test_id_mismatch(self):
with self.assertRaises(ValueError):
netprobe.parse_response(answer(3, "x.y"), 4)
class DnsQueryTest(unittest.TestCase):
def test_against_fake_server(self):
server = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
server.bind(("127.0.0.1", 0))
port = server.getsockname()[1]
def serve():
data, addr = server.recvfrom(512)
ident = struct.unpack(">H", data[:2])[0]
server.sendto(answer(ident, "broker.lan", addresses=["192.168.0.10"]), addr)
server.close()
threading.Thread(target=serve, daemon=True).start()
result = netprobe.dns_query("127.0.0.1", "broker.lan", timeout=2, port=port)
self.assertEqual(result["addresses"], ["192.168.0.10"])
self.assertEqual(result["rcode"], "NOERROR")
class ResolvConfTest(unittest.TestCase):
def test_parse(self):
with tempfile.NamedTemporaryFile("w", delete=False) as fh:
fh.write("# comment\nnameserver 192.168.0.1\nnameserver 8.8.8.8 # public\nsearch lan home\noptions ndots:1\n")
try:
parsed = netprobe.read_resolv_conf(fh.name)
finally:
os.unlink(fh.name)
self.assertEqual(parsed["nameservers"], ["192.168.0.1", "8.8.8.8"])
self.assertEqual(parsed["search"], ["lan", "home"])
self.assertEqual(parsed["options"], ["ndots:1"])
def test_missing(self):
parsed = netprobe.read_resolv_conf("/nonexistent/resolv.conf")
self.assertEqual(parsed["nameservers"], [])
self.assertIn("error", parsed)
def test_hosts(self):
with tempfile.NamedTemporaryFile("w", delete=False) as fh:
fh.write("127.0.0.1 localhost\n192.168.0.5 ha.local homeassistant\n")
try:
self.assertEqual(netprobe.hosts_entries("homeassistant", fh.name), ["192.168.0.5"])
self.assertEqual(netprobe.hosts_entries("other", fh.name), [])
finally:
os.unlink(fh.name)
class TcpTest(unittest.TestCase):
def test_connect_ok_and_refused(self):
listener = socket.socket()
listener.bind(("127.0.0.1", 0))
listener.listen(1)
port = listener.getsockname()[1]
try:
ok = netprobe.tcp_connect("127.0.0.1", port, 2)
finally:
listener.close()
self.assertTrue(ok["ok"])
refused = netprobe.tcp_connect("127.0.0.1", port, 2)
self.assertFalse(refused["ok"])
self.assertIn("error", refused)
class SummaryTest(unittest.TestCase):
def test_disagreement_flagged(self):
result = {
"host": "ha.milans.cloud", "port": 1883, "hosts_file": [],
"resolv_conf": {"nameservers": ["192.168.0.1", "8.8.8.8"]},
"getaddrinfo": {"addresses": [], "error": "Name does not resolve"},
"nameservers": [
{"server": "192.168.0.1", "rcode": "NOERROR", "addresses": ["192.168.0.20"]},
{"server": "8.8.8.8", "rcode": "NXDOMAIN", "addresses": []},
],
"tcp": [{"address": "192.168.0.20", "ok": True, "ms": 2.0}],
}
lines = netprobe.summarize(result)
self.assertTrue(any("disagree" in line for line in lines))
self.assertTrue(any("System resolver failed" in line for line in lines))
self.assertTrue(any("connected in" in line for line in lines))
if __name__ == "__main__":
unittest.main()