ServiceMap: update dependencies and backend/app (+5 areas)
Public changes compared with the previous published export: 16 files changed. Affected areas: - backend/app: 3 files - dependencies: 5 files - deployment and CI: 2 files - documentation: 2 files - frontend/src: 1 file - project files: 1 file - tests: 2 files Changed files: - Updated: ".env.example" - Updated: "Dockerfile.unraid" - Updated: "README.md" - Updated: "backend/app/incidents.py" - Updated: "backend/app/main.py" - Updated: "backend/app/store.py" - Updated: "backend/pyproject.toml" - Updated: "backend/tests/test_docker_provider.py" - Updated: "backend/tests/test_v163_health_performance.py" - Updated: "docker-compose.broker.yml" - Updated: "docs/ARCHITECTURE.md" - Updated: "frontend/package-lock.json" - Updated: "frontend/package.json" - Updated: "frontend/src/App.tsx" - Updated: "package-lock.json" - Updated: "package.json" Export provenance/manifests refreshed; excluded from the change summary. Publish curated ServiceMap source from 8c297f756ff84aaa917ae9c48535e835f8c09ec7
This commit is contained in:
1 parent
d4c351ce73
commit
41a8eee259
18 files changed
+167
-92
No files matched your search
@@ -97,14 +97,23 @@ def _severity(service: Service) -> Criticality:
|
||||
# concurrent passes could both see "no active incident" for a root and each
|
||||
# open one (the later duplicate cleanup then left two journal entries).
|
||||
_SYNC_LOCK = threading.RLock()
|
||||
_DEFAULT_RESOLUTION_REASON = "Hoofdoorzaak hersteld; incident opgelost"
|
||||
|
||||
|
||||
def sync_incidents(store: Store, *, resolution_reason: str = "Hoofdoorzaak hersteld; incident opgelost") -> list[IncidentRecord]:
|
||||
def sync_incidents(store: Store, *, resolution_reason: str = _DEFAULT_RESOLUTION_REASON) -> list[IncidentRecord]:
|
||||
with _SYNC_LOCK:
|
||||
return _sync_incidents_locked(store, resolution_reason=resolution_reason)
|
||||
_sync_incidents_locked(store, resolution_reason=resolution_reason)
|
||||
return store.list_incidents(active_only=False)
|
||||
|
||||
|
||||
def _sync_incidents_locked(store: Store, *, resolution_reason: str) -> list[IncidentRecord]:
|
||||
def reconcile_incidents(store: Store) -> tuple[TopologyHealth, IncidentOverview]:
|
||||
with _SYNC_LOCK:
|
||||
return _sync_incidents_locked(store, resolution_reason=_DEFAULT_RESOLUTION_REASON)
|
||||
|
||||
|
||||
def _sync_incidents_locked(
|
||||
store: Store, *, resolution_reason: str
|
||||
) -> tuple[TopologyHealth, IncidentOverview]:
|
||||
"""Reconcile the current graph with the durable incident journal.
|
||||
|
||||
A root service has at most one active incident. Repeated probe cycles only
|
||||
@@ -117,7 +126,7 @@ def _sync_incidents_locked(store: Store, *, resolution_reason: str) -> list[Inci
|
||||
now = datetime.now(UTC)
|
||||
active: dict[str, IncidentRecord] = {}
|
||||
for incident in sorted(
|
||||
store.list_incidents(active_only=True),
|
||||
store.list_incidents(active_only=True, services=services),
|
||||
key=lambda item: (item.detected_at, item.id),
|
||||
):
|
||||
if incident.root_service_id in active:
|
||||
@@ -191,4 +200,4 @@ def _sync_incidents_locked(store: Store, *, resolution_reason: str) -> list[Inci
|
||||
)
|
||||
store.upsert_incident(resolved)
|
||||
store.append_incident_event(incident.id, "incident.resolved", resolution_reason, root_id)
|
||||
return store.list_incidents(active_only=False)
|
||||
return health, overview
|
||||
+42
-40
@@ -22,8 +22,8 @@ from .discovery_store import DiscoveryStore
|
||||
from .docker_discovery import discover_containers
|
||||
from .docker_provider import docker_provider_mode
|
||||
from .events import SSE_RETRY_MS, event_bus, next_event, sse_payload
|
||||
from .health import status_maps, topology_health, validate_dependencies
|
||||
from .incidents import incident_overview, sync_incidents
|
||||
from .health import TopologyHealth, status_maps, topology_health, validate_dependencies
|
||||
from .incidents import incident_overview, reconcile_incidents, sync_incidents
|
||||
from .loops import loop_registry, run_periodic
|
||||
from .models import (
|
||||
AvailabilitySummary,
|
||||
@@ -288,7 +288,7 @@ async def lifespan(_: FastAPI):
|
||||
await background_task
|
||||
|
||||
|
||||
app = FastAPI(title="ServiceMap API", version="1.6.3", lifespan=lifespan)
|
||||
app = FastAPI(title="ServiceMap API", version="1.6.4", lifespan=lifespan)
|
||||
|
||||
|
||||
# Bounded request bodies: the largest legitimate payload is a config import.
|
||||
@@ -404,7 +404,7 @@ def meta() -> dict[str, str | int | float | bool]:
|
||||
auth_mode = "misconfigured"
|
||||
return {
|
||||
"name": "ServiceMap",
|
||||
"version": "1.6.3",
|
||||
"version": "1.6.4",
|
||||
"stage": "v1",
|
||||
"scheduled_probing": PROBE_INTERVAL > 0,
|
||||
"probe_interval_seconds": PROBE_INTERVAL,
|
||||
@@ -488,21 +488,21 @@ def verify_write_access() -> dict[str, bool]:
|
||||
return {"ok": True}
|
||||
|
||||
|
||||
def _visible_services(services: list[Service]) -> list[Service]:
|
||||
return [
|
||||
service for service in services
|
||||
if service.visibility.value == "visible" and service.lifecycle.value not in {"retired", "ignored"}
|
||||
]
|
||||
|
||||
|
||||
def _service_snapshots(
|
||||
*, include_hidden: bool = False, include_context: bool = True
|
||||
) -> list[ServiceSnapshot]:
|
||||
*, include_hidden: bool = False, include_details: bool = True
|
||||
) -> tuple[list[ServiceSnapshot], TopologyHealth]:
|
||||
all_services = store.list_services()
|
||||
services = (
|
||||
all_services
|
||||
if include_hidden
|
||||
else [
|
||||
service
|
||||
for service in all_services
|
||||
if service.visibility.value == "visible" and service.lifecycle.value not in {"retired", "ignored"}
|
||||
]
|
||||
)
|
||||
latest = store.latest_probes([service.id for service in services])
|
||||
availability = store.availability_summaries([service.id for service in services], 24)
|
||||
services = all_services if include_hidden else _visible_services(all_services)
|
||||
visible_ids = [service.id for service in services]
|
||||
latest = store.latest_probes(visible_ids)
|
||||
availability = store.availability_summaries(visible_ids, 24) if include_details else {}
|
||||
evidence_by_service = store.health_evidence_for_services([service.id for service in all_services])
|
||||
# Calculate propagation over the complete canonical graph first. Hidden
|
||||
# infrastructure nodes may still sit between a visible service and its
|
||||
@@ -512,15 +512,15 @@ def _service_snapshots(
|
||||
[service.id for service in all_services], evidence_by_service=evidence_by_service
|
||||
)
|
||||
effective_all, dependency_all = status_maps(all_services, direct_statuses=direct_all)
|
||||
dependent_names = {
|
||||
service.id: [item.name for item in services if service.id in item.depends_on]
|
||||
for service in services
|
||||
}
|
||||
visible_ids = [service.id for service in services]
|
||||
windows = store.maintenance_windows()
|
||||
sources_by_service = discovery_store.sources_for_services(visible_ids) if include_context else {}
|
||||
relations_by_service = discovery_store.evidence_for_services(visible_ids) if include_context else {}
|
||||
return [
|
||||
dependent_names: dict[str, list[str]] = {}
|
||||
if include_details:
|
||||
for service in services:
|
||||
for dependency_id in set(service.depends_on):
|
||||
dependent_names.setdefault(dependency_id, []).append(service.name)
|
||||
windows = store.maintenance_windows() if include_details else {}
|
||||
sources_by_service = discovery_store.sources_for_services(visible_ids) if include_details else {}
|
||||
relations_by_service = discovery_store.evidence_for_services(visible_ids) if include_details else {}
|
||||
snapshots = [
|
||||
ServiceSnapshot(
|
||||
service=service,
|
||||
latest_probe=latest.get(service.id),
|
||||
@@ -539,16 +539,18 @@ def _service_snapshots(
|
||||
sources=sources_by_service.get(service.id, []),
|
||||
evidence=relations_by_service.get(service.id, []),
|
||||
health_evidence=evidence_by_service.get(service.id, []),
|
||||
health_resolution=resolutions.get(service.id, HealthResolution()),
|
||||
health_resolution=resolutions.get(service.id) or HealthResolution(),
|
||||
maintenance_until=windows.get(service.id),
|
||||
)
|
||||
for service in services
|
||||
]
|
||||
return snapshots, TopologyHealth(all_services, direct_all, effective_all, dependency_all)
|
||||
|
||||
|
||||
@app.get("/api/v1/services", response_model=list[ServiceSnapshot])
|
||||
def list_services(include_hidden: bool = Query(default=False)) -> list[ServiceSnapshot]:
|
||||
return _service_snapshots(include_hidden=include_hidden)
|
||||
snapshots, _ = _service_snapshots(include_hidden=include_hidden)
|
||||
return snapshots
|
||||
|
||||
|
||||
@app.post("/api/v1/services", response_model=Service, status_code=status.HTTP_201_CREATED)
|
||||
@@ -768,23 +770,23 @@ async def event_stream() -> StreamingResponse:
|
||||
|
||||
@app.get("/api/v1/overview", response_model=OverviewSummary)
|
||||
def overview() -> OverviewSummary:
|
||||
sync_incidents(store)
|
||||
# Only status and coverage are counted; discovery context is not needed.
|
||||
snapshots = _service_snapshots(include_hidden=False, include_context=False)
|
||||
overview_state = incident_overview(store)
|
||||
health, overview_state = reconcile_incidents(store)
|
||||
services = _visible_services(health.services)
|
||||
counts: dict[str, int] = {}
|
||||
coverage: dict[str, int] = {}
|
||||
for snapshot in snapshots:
|
||||
counts[snapshot.effective_status.value] = counts.get(snapshot.effective_status.value, 0) + 1
|
||||
state = snapshot.service.monitoring_state.value
|
||||
for service in services:
|
||||
status = health.effective[service.id].value
|
||||
counts[status] = counts.get(status, 0) + 1
|
||||
state = service.monitoring_state.value
|
||||
coverage[state] = coverage.get(state, 0) + 1
|
||||
services_by_id = {service.id: service for service in health.services}
|
||||
return OverviewSummary(
|
||||
generated_at=overview_state.generated_at,
|
||||
total_services=len(snapshots),
|
||||
total_services=len(services),
|
||||
health_counts=counts,
|
||||
active_incidents=len(store.list_incidents(active_only=True)),
|
||||
active_incidents=len(store.list_incidents(active_only=True, services=services_by_id)),
|
||||
likely_root_causes=[candidate.service_name for candidate in overview_state.candidates],
|
||||
flapping_services=[snapshot.service.name for snapshot in snapshots if snapshot.flapping],
|
||||
flapping_services=[service.name for service in services if service.flapping],
|
||||
recent_events=store.list_events(limit=10),
|
||||
monitoring_coverage=coverage,
|
||||
)
|
||||
@@ -982,8 +984,8 @@ def _semantic_representation(include_hidden: bool) -> CachedRepresentation:
|
||||
cached = semantic_cache.get(cache_key, revision)
|
||||
if cached is not None:
|
||||
return cached
|
||||
snapshots = _service_snapshots(include_hidden=include_hidden, include_context=False)
|
||||
incident_state = incident_overview(store)
|
||||
snapshots, health = _service_snapshots(include_hidden=include_hidden, include_details=False)
|
||||
incident_state = incident_overview(store, health=health)
|
||||
root_ids = {candidate.service_id for candidate in incident_state.candidates}
|
||||
affected_ids = {
|
||||
service_id
|
||||
|
||||
@@ -2134,8 +2134,9 @@ class Store:
|
||||
event_bus.publish({"event_type": "topology.changed", "message": "Synthetische demo afgesloten"})
|
||||
return True
|
||||
|
||||
def list_incidents(self, active_only: bool = False, limit: int = 100) -> list[IncidentRecord]:
|
||||
services = {service.id: service for service in self.list_services()}
|
||||
def list_incidents(
|
||||
self, active_only: bool = False, limit: int = 100, *, services: dict[str, Service] | None = None
|
||||
) -> list[IncidentRecord]:
|
||||
with self._connect() as db:
|
||||
query = "SELECT * FROM incidents"
|
||||
params: tuple[Any, ...] = ()
|
||||
@@ -2144,6 +2145,10 @@ class Store:
|
||||
query += " ORDER BY detected_at DESC LIMIT ?"
|
||||
params += (limit,)
|
||||
rows = db.execute(query, params).fetchall()
|
||||
if not rows:
|
||||
return []
|
||||
if services is None:
|
||||
services = {service.id: service for service in self.list_services()}
|
||||
return [self._row_to_incident(row, services) for row in rows]
|
||||
|
||||
def get_incident(self, incident_id: str) -> IncidentRecord | None:
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
[project]
|
||||
name = "servicemap-backend"
|
||||
version = "1.6.3"
|
||||
version = "1.6.4"
|
||||
description = "ServiceMap API, topology intelligence and probe engine"
|
||||
requires-python = ">=3.12"
|
||||
dependencies = [
|
||||
|
||||
@@ -53,6 +53,41 @@ def _mocked_broker_client(
|
||||
return client, options
|
||||
|
||||
|
||||
def test_reopened_inventory_client_reads_atomically_rotated_broker_token(tmp_path, monkeypatch) -> None:
|
||||
token_file = tmp_path / "servicemap-docker-broker.token"
|
||||
token_file.write_text("initial-token", encoding="utf-8")
|
||||
monkeypatch.setenv("SERVICEMAP_DOCKER_PROVIDER", "broker")
|
||||
monkeypatch.setenv("SERVICEMAP_DOCKER_BROKER_URL", "https://broker.internal")
|
||||
monkeypatch.setenv("SERVICEMAP_DOCKER_BROKER_TOKEN_FILE", str(token_file))
|
||||
|
||||
seen = []
|
||||
def inventory(request):
|
||||
seen.append(request.headers["authorization"])
|
||||
return httpx.Response(200, json={"items": []})
|
||||
|
||||
original_client = httpx.Client
|
||||
monkeypatch.setattr(
|
||||
docker_provider.httpx, "Client",
|
||||
lambda **kwargs: original_client(transport=httpx.MockTransport(inventory), **kwargs),
|
||||
)
|
||||
client = docker_provider.docker_client_for_inventory("")
|
||||
try:
|
||||
assert client.containers.list() == []
|
||||
finally:
|
||||
client.close()
|
||||
|
||||
replacement = tmp_path / "next-token"
|
||||
replacement.write_text("rotated-token", encoding="utf-8")
|
||||
replacement.replace(token_file)
|
||||
client = docker_provider.docker_client_for_inventory("")
|
||||
try:
|
||||
assert client.containers.list() == []
|
||||
finally:
|
||||
client.close()
|
||||
|
||||
assert seen == ["Bearer initial-token", "Bearer rotated-token"]
|
||||
|
||||
|
||||
def test_inventory_uses_authenticated_fixed_get_and_health_check_has_no_token(monkeypatch) -> None:
|
||||
requests = []
|
||||
|
||||
|
||||
@@ -353,12 +353,35 @@ def test_topology_request_budget_at_300_services(tmp_path: Path, monkeypatch) ->
|
||||
return original()
|
||||
|
||||
monkeypatch.setattr(store, "_connect", counting_connect)
|
||||
evidence_reads = 0
|
||||
read_evidence = store.health_evidence_for_services
|
||||
|
||||
def counting_evidence(service_ids, **kwargs):
|
||||
nonlocal evidence_reads
|
||||
evidence_reads += 1
|
||||
return read_evidence(service_ids, **kwargs)
|
||||
|
||||
monkeypatch.setattr(store, "health_evidence_for_services", counting_evidence)
|
||||
topology_reads = 0
|
||||
list_services = store.list_services
|
||||
|
||||
def counting_services():
|
||||
nonlocal topology_reads
|
||||
topology_reads += 1
|
||||
return list_services()
|
||||
|
||||
monkeypatch.setattr(store, "list_services", counting_services)
|
||||
client = TestClient(main_module.app)
|
||||
budget_ms = os.environ.get("SERVICEMAP_API_P95_BUDGET_MS")
|
||||
|
||||
for endpoint in ("/api/v1/topology/semantic", "/api/v1/overview"):
|
||||
connections = 0
|
||||
evidence_reads = 0
|
||||
topology_reads = 0
|
||||
main_module.semantic_cache.clear()
|
||||
assert client.get(endpoint).status_code == 200
|
||||
assert evidence_reads == 1, (endpoint, evidence_reads)
|
||||
assert topology_reads == 1, (endpoint, topology_reads)
|
||||
assert connections <= 30, (endpoint, connections)
|
||||
elapsed_ms = []
|
||||
for _ in range(20):
|
||||
|
||||
Reference in new issue
Block a user