SIENTIAPDE-1974: fix(read): code-review fixes for the slot v2 read path

- config: GatewayConfig.certs_from_env() reads only CERT_BUCKET/CERT_CACHE_DIR;
  IngestorManager uses it, so a malformed write/gRPC env var (GRPC_PORT,
  WRITE_MIN_PAUSE_S, IDLE_CONN_MAX_S) can no longer stop the reads.
- certs: CertProvider raises CertificateNotConfiguredError with an actionable
  message when no platform_certificates document exists; downloads once under
  an asyncio.Lock, caches the paths, writes atomically (tmp + os.replace);
  invalidate() forces a fresh download.
- pool: server_id normalized to str (1 and '1' share a session); warns when a
  get() asks for other url/server_uri/use_security than the pooled client was
  opened with (client reused, the slot config wins); invalidates the cert
  cache when a secured connect fails.
- SubscriptionEngine: tracks nodes per subscription and forgets them on
  unsubscribe (no more cycle alarms / tag count for removed tags), keeping
  addresses still monitored by another slot.
- ServerReader.shutdown(evict) / IngestorManager.remove_server(evict): a
  server that only left this pod's slots keeps the shared connection for
  NATS writes/gRPC; lost servers and config changes still evict.
- IngestorManager warns once per server when a slot still sets
  server_cert_path (server certificate pinning is no longer applied).
- slots: CONNECTION_FIELDS exported and reused by ServerReader.

Accepted, not changed: read metrics label server_name with the server_id
(same as the write path, which only knows the id); a malformed slot is
notified each lease attempt (alerts are deduplicated by notification_id).

Tests: 236 unit tests pass (+18); ruff clean; local e2e against an asyncua
server incl. slot moved (connection kept and reused by the write path) and
lost-server eviction/re-initialization.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Eduardo Rios
2026-10-06 13:54:04 -03:00
parent 50578fdba8
commit 3d8ee1c2ed
13 changed files with 421 additions and 31 deletions

View File

@@ -255,3 +255,46 @@ async def test_unsubscribe_tolerates_dead_connection():
assert 'slot1' not in eng.subscriptions
eng.pool.unpin.assert_called_once_with('1')
eng.logger.warning.assert_called_once()
async def _subscribed_engine(slots):
"""Engine with one subscription per slot: {slot: [addresses]}."""
eng = make_engine()
eng.pool = MagicMock(
get=AsyncMock(
return_value=MagicMock(create_subscription=AsyncMock(return_value=AsyncMock()))
)
)
for slot, addresses in slots.items():
await eng.create_subscription('1', {'subscription_period_ms': 500}, slot)
nodes = {
a: {'frequency': '1000', 'names': [{'collection': 'c', 'tag_name': a}]}
for a in addresses
}
await eng.subscribe(slot, nodes, 10)
return eng
@pytest.mark.asyncio
async def test_unsubscribe_forgets_the_slot_nodes():
eng = await _subscribed_engine({'s1': ['ns=2;i=1', 'ns=2;i=2'], 's2': ['ns=2;i=3']})
await eng.unsubscribe('s1')
# no more cycle checks / tag counting for addresses nobody reads anymore
assert set(eng.nodes) == {'ns=2;i=3'}
eng.emit_metric.assert_any_await(
metric_object=metrics.OPC_TAGS_SUBSCRIBED,
method='set',
value=1,
tags={'pod_id': eng.pod_id, 'server_name': '1'},
)
@pytest.mark.asyncio
async def test_unsubscribe_keeps_node_shared_with_another_slot():
eng = await _subscribed_engine({'s1': ['ns=2;i=1', 'ns=2;i=2'], 's2': ['ns=2;i=2']})
await eng.unsubscribe('s1')
assert set(eng.nodes) == {'ns=2;i=2'}