alex/manage
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor(observability): drop file-SD writer for http_sd_configs

Browse Source 
Slice 4 of observability-service-registry. Removes the shared-file
Prometheus bridge; external Prometheus now consumes node-exporter targets
via http_sd_configs against GET /api/monitoring/prometheus-targets.

- services/targets.py: removed write_prometheus_targets() (the file
  writer) and its json/Path/get_settings imports; updated module docstring.
  build_node_exporter_targets() is unchanged and still powers the HTTP
  endpoint.
- main.py: removed the startup write_prometheus_targets call.
- routers/settings.py: removed the _write_prometheus_targets helper and
  its three post machine create/update/delete call sites + the now-unused
  targets import.
- config.py: removed the prometheus_file_sd_dir field.
- docker-compose.yml / docker-compose.dev.yml: removed the
  PROMETHEUS_FILE_SD_DIR backend env var.
- tests: removed TestWritePrometheusTargets + the write_prometheus_targets
  import in test_targets.py; rewrote the two TestSettingsMachines tests to
  assert machines appear/disappear from /api/monitoring/prometheus-targets
  (the surviving HTTP path) instead of the removed file-writer side effect.

ruff clean; 239 backend tests pass.
This commit is contained in:
Developer
2026-06-24 08:37:20 +00:00
parent 0c5698c903
commit b200025daa
9 changed files with 57 additions and 118 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backend/src/media_library_viewer_api/config.py
-1
View File
@@ -54,7 +54,6 @@ class Settings(BaseSettings):
# Observability # Observability
prometheus_enabled: bool = True prometheus_enabled: bool = True
prometheus_file_sd_dir: str = "/app/backend/.cache/prometheus-file-sd"
alertmanager_url: str = "" alertmanager_url: str = ""
alertmanager_webhook_url: str = "" # Optional receiver for alertmanager webhook notifications alertmanager_webhook_url: str = "" # Optional receiver for alertmanager webhook notifications
backend/src/media_library_viewer_api/main.py
-6
View File
@@ -44,12 +44,6 @@ async def lifespan(app: FastAPI):
validate_encryption_key() validate_encryption_key()
logger.info("Backend startup complete: %s", describe_settings(settings)) logger.info("Backend startup complete: %s", describe_settings(settings))
logger.info("Managed known_hosts will be populated lazily on first successful SSH connection") logger.info("Managed known_hosts will be populated lazily on first successful SSH connection")
try:
from media_library_viewer_api.services.targets import write_prometheus_targets
write_prometheus_targets(get_settings_store())
except Exception:
logger.exception("Failed to write Prometheus file-SD targets during startup")
try: try:
get_settings_store().ensure_defaults() get_settings_store().ensure_defaults()
except Exception: except Exception:
backend/src/media_library_viewer_api/routers/settings.py
-12
View File
@@ -17,7 +17,6 @@ from media_library_viewer_api.services.db_maintenance import remove_sqlite_datab
from media_library_viewer_api.services.known_hosts import has_known_host from media_library_viewer_api.services.known_hosts import has_known_host
from media_library_viewer_api.services.media_index import MediaIndex from media_library_viewer_api.services.media_index import MediaIndex
from media_library_viewer_api.services.settings_store import SettingsStore from media_library_viewer_api.services.settings_store import SettingsStore
from media_library_viewer_api.services.targets import write_prometheus_targets
logger = logging.getLogger(__name__) logger = logging.getLogger(__name__)
@@ -121,14 +120,6 @@ def _validate_saved_machine_ssh(machine: MonitoringMachineInput, store: Settings
client.close() client.close()
def _write_prometheus_targets(store: SettingsStore) -> None:
"""Regenerate Prometheus file-SD targets after machine changes."""
try:
write_prometheus_targets(store)
except Exception:
logger.exception("Failed to write Prometheus file-SD targets")
@router.post("/machines/test-ssh") @router.post("/machines/test-ssh")
def test_machine_ssh( def test_machine_ssh(
machine: MonitoringMachineInput, machine: MonitoringMachineInput,
@@ -189,7 +180,6 @@ def post_machine(
store: SettingsStore = Depends(get_settings_store), store: SettingsStore = Depends(get_settings_store),
) -> dict[str, Any]: ) -> dict[str, Any]:
saved = store.upsert_machine(machine.model_dump(exclude_none=True), machine.id) saved = store.upsert_machine(machine.model_dump(exclude_none=True), machine.id)
_write_prometheus_targets(store)
saved_machine = MonitoringMachineInput.model_validate(saved) saved_machine = MonitoringMachineInput.model_validate(saved)
_validate_saved_machine_ssh(saved_machine, store) _validate_saved_machine_ssh(saved_machine, store)
return saved return saved
@@ -204,7 +194,6 @@ def put_machine(
if not store.get_machine(machine_id): if not store.get_machine(machine_id):
raise HTTPException(status_code=404, detail="Machine not found") raise HTTPException(status_code=404, detail="Machine not found")
saved = store.upsert_machine(machine.model_dump(exclude_none=True), machine_id) saved = store.upsert_machine(machine.model_dump(exclude_none=True), machine_id)
_write_prometheus_targets(store)
saved_machine = MonitoringMachineInput.model_validate(saved) saved_machine = MonitoringMachineInput.model_validate(saved)
_validate_saved_machine_ssh(saved_machine, store) _validate_saved_machine_ssh(saved_machine, store)
return saved return saved
@@ -215,7 +204,6 @@ def delete_machine(machine_id: str, store: SettingsStore = Depends(get_settings_
if not store.get_machine(machine_id): if not store.get_machine(machine_id):
raise HTTPException(status_code=404, detail="Machine not found") raise HTTPException(status_code=404, detail="Machine not found")
store.delete_machine(machine_id) store.delete_machine(machine_id)
_write_prometheus_targets(store)
return {"status": "deleted"} return {"status": "deleted"}
backend/src/media_library_viewer_api/services/targets.py
+6 -25
View File
@@ -1,19 +1,16 @@
"""Prometheus file-based service discovery target management. """Prometheus Node Exporter target discovery.
The backend owns the list of remote Node Exporter targets so that operators can The backend owns the list of remote Node Exporter targets so that operators can
enable scraping per machine from the Manage UI. Prometheus reads the generated enable scraping per machine from the Manage UI. The list is exposed over HTTP at
JSON file via `file_sd_configs`; this keeps Prometheus config static and pushes ``GET /api/monitoring/prometheus-targets`` and consumed by an external Prometheus
machine-specific changes into a file it can reload. via ``http_sd_configs`` (no shared volume required).
""" """
from __future__ import annotations from __future__ import annotations
import json
import logging import logging
from pathlib import Path
from typing import Any from typing import Any
from media_library_viewer_api.config import get_settings
from media_library_viewer_api.services.settings_store import SettingsStore from media_library_viewer_api.services.settings_store import SettingsStore
logger = logging.getLogger(__name__) logger = logging.getLogger(__name__)
@@ -34,10 +31,9 @@ def _scrape_address(machine: dict[str, Any]) -> str | None:
def build_node_exporter_targets(store: SettingsStore) -> list[dict[str, Any]]: def build_node_exporter_targets(store: SettingsStore) -> list[dict[str, Any]]:
"""Build a file-SD target list for all enabled SSH machines. """Build an http-SD target list for all enabled SSH machines.
Local machines are excluded because the Compose-managed node-exporter Local machines are excluded because the Docker host is scraped directly.
service already covers the Docker host.
""" """
targets: list[dict[str, Any]] = [] targets: list[dict[str, Any]] = []
for machine in store.list_machines(): for machine in store.list_machines():
@@ -60,18 +56,3 @@ def build_node_exporter_targets(store: SettingsStore) -> list[dict[str, Any]]:
} }
) )
return targets return targets
def write_prometheus_targets(store: SettingsStore, file_sd_dir: Path | None = None) -> Path:
"""Render and persist Prometheus file-SD targets.
Returns the path written so callers can log or expose it.
"""
settings = get_settings()
file_sd_dir = file_sd_dir or Path(settings.prometheus_file_sd_dir)
file_sd_dir.mkdir(parents=True, exist_ok=True)
file_path = file_sd_dir / "node_exporter_targets.json"
targets = build_node_exporter_targets(store)
file_path.write_text(json.dumps(targets, indent=2), encoding="utf-8")
logger.info("Wrote %s node_exporter targets to %s", len(targets), file_path)
return file_path
backend/tests/test_api.py
+24 -23
View File
@@ -702,27 +702,26 @@ class TestResolveServiceRecord:
class TestSettingsMachines: class TestSettingsMachines:
def test_post_machine_rewrites_prometheus_targets(self, test_client): def test_machine_appears_in_prometheus_targets(self, test_client):
with patch("media_library_viewer_api.routers.settings.write_prometheus_targets") as write_targets: store = app.dependency_overrides[get_settings_store]()
with patch("media_library_viewer_api.routers.settings._validate_saved_machine_ssh"): store.upsert_machine(
response = test_client.post( {
"/api/settings/machines", "name": "remote1",
json={ "mode": "ssh",
"name": "remote1", "enabled": True,
"mode": "ssh", "services": ["monitoring"],
"enabled": True, "host": "10.0.0.5",
"services": ["monitoring"], "username": "u",
"host": "10.0.0.5", "node_exporter_enabled": True,
"username": "u", "node_exporter_port": 9200,
"node_exporter_enabled": True, "node_exporter_scrape_host": "1.2.3.4",
"node_exporter_port": 9200, }
"node_exporter_scrape_host": "1.2.3.4", )
}, targets = test_client.get("/api/monitoring/prometheus-targets").json()
) assert len(targets) == 1
assert response.status_code == 201 assert targets[0]["targets"] == ["1.2.3.4:9200"]
write_targets.assert_called_once()
def test_delete_machine_rewrites_prometheus_targets(self, test_client): def test_delete_machine_removed_from_prometheus_targets(self, test_client):
store = app.dependency_overrides[get_settings_store]() store = app.dependency_overrides[get_settings_store]()
machine = store.upsert_machine( machine = store.upsert_machine(
{ {
@@ -732,12 +731,14 @@ class TestSettingsMachines:
"services": ["monitoring"], "services": ["monitoring"],
"host": "10.0.0.5", "host": "10.0.0.5",
"username": "u", "username": "u",
"node_exporter_enabled": True,
"node_exporter_port": 9200,
"node_exporter_scrape_host": "1.2.3.4",
} }
) )
with patch("media_library_viewer_api.routers.settings.write_prometheus_targets") as write_targets: response = test_client.delete(f"/api/settings/machines/{machine['id']}")
response = test_client.delete(f"/api/settings/machines/{machine['id']}")
assert response.status_code == 200 assert response.status_code == 200
write_targets.assert_called_once() assert test_client.get("/api/monitoring/prometheus-targets").json() == []
def _am_service(name="Alertmanager", **config): def _am_service(name="Alertmanager", **config):
backend/tests/test_targets.py
+2 -24
View File
@@ -1,14 +1,11 @@
"""Tests for Prometheus file-based service discovery target generation.""" """Tests for Prometheus Node Exporter target discovery."""
from pathlib import Path from pathlib import Path
import pytest import pytest
from media_library_viewer_api.services.settings_store import SettingsStore from media_library_viewer_api.services.settings_store import SettingsStore
from media_library_viewer_api.services.targets import ( from media_library_viewer_api.services.targets import build_node_exporter_targets
build_node_exporter_targets,
write_prometheus_targets,
)
@pytest.fixture @pytest.fixture
@@ -88,22 +85,3 @@ class TestBuildNodeExporterTargets:
} }
) )
assert build_node_exporter_targets(store) == [] assert build_node_exporter_targets(store) == []
class TestWritePrometheusTargets:
def test_writes_valid_json(self, store: SettingsStore, tmp_path: Path):
store.upsert_machine(
{
"name": "remote1",
"mode": "ssh",
"host": "10.0.0.5",
"username": "u",
"node_exporter_enabled": True,
"node_exporter_port": 9200,
}
)
file_path = write_prometheus_targets(store, tmp_path)
assert file_path.exists()
assert file_path.name == "node_exporter_targets.json"
targets = build_node_exporter_targets(store)
assert len(targets) == 1
docker-compose.dev.yml
-1
View File
@@ -23,7 +23,6 @@ services:
LOG_FORMAT: ${LOG_FORMAT:-text} LOG_FORMAT: ${LOG_FORMAT:-text}
PROMETHEUS_ENABLED: "true" PROMETHEUS_ENABLED: "true"
SSH_KNOWN_HOSTS_PATH: /app/backend/.cache/known_hosts SSH_KNOWN_HOSTS_PATH: /app/backend/.cache/known_hosts
PROMETHEUS_FILE_SD_DIR: /app/backend/.cache/prometheus-file-sd
ALERTMANAGER_URL: ${ALERTMANAGER_URL:-} ALERTMANAGER_URL: ${ALERTMANAGER_URL:-}
ALERTMANAGER_WEBHOOK_URL: ${ALERTMANAGER_WEBHOOK_URL:-} ALERTMANAGER_WEBHOOK_URL: ${ALERTMANAGER_WEBHOOK_URL:-}
MANAGE_ENCRYPTION_KEY: ${MANAGE_ENCRYPTION_KEY:?set MANAGE_ENCRYPTION_KEY in your .env} MANAGE_ENCRYPTION_KEY: ${MANAGE_ENCRYPTION_KEY:?set MANAGE_ENCRYPTION_KEY in your .env}
docker-compose.yml
-1
View File
@@ -34,7 +34,6 @@ services:
SMTP_USE_SSL: ${SMTP_USE_SSL:-false} SMTP_USE_SSL: ${SMTP_USE_SSL:-false}
SMTP_TIMEOUT: ${SMTP_TIMEOUT:-30} SMTP_TIMEOUT: ${SMTP_TIMEOUT:-30}
SSH_KNOWN_HOSTS_PATH: /app/backend/.cache/known_hosts SSH_KNOWN_HOSTS_PATH: /app/backend/.cache/known_hosts
PROMETHEUS_FILE_SD_DIR: ${PROMETHEUS_FILE_SD_DIR:-/app/backend/.cache/prometheus-file-sd}
ALERTMANAGER_URL: ${ALERTMANAGER_URL:-} ALERTMANAGER_URL: ${ALERTMANAGER_URL:-}
ALERTMANAGER_WEBHOOK_URL: ${ALERTMANAGER_WEBHOOK_URL:-} ALERTMANAGER_WEBHOOK_URL: ${ALERTMANAGER_WEBHOOK_URL:-}
MANAGE_ENCRYPTION_KEY: ${MANAGE_ENCRYPTION_KEY:?generate one with python -c "from cryptography.fernet import Fernet; print(Fernet.generate_key().decode())"} MANAGE_ENCRYPTION_KEY: ${MANAGE_ENCRYPTION_KEY:?generate one with python -c "from cryptography.fernet import Fernet; print(Fernet.generate_key().decode())"}
frontend/src/components/ObservabilityPage.tsx
+25 -25
View File
@@ -621,33 +621,33 @@ export function ObservabilityPage() {
</Select> </Select>
</CardHeader> </CardHeader>
<CardContent className="space-y-4"> <CardContent className="space-y-4">
{selectedMachine ? ( {selectedMachine ? (
GRAFANA_BASE_URL ? ( GRAFANA_BASE_URL ? (
<> <>
<GrafanaLinkCard <GrafanaLinkCard
title={`${selectedMachine.name} metrics`} title={`${selectedMachine.name} metrics`}
description="Open the Node Exporter overview dashboard for this machine in Grafana." description="Open the Node Exporter overview dashboard for this machine in Grafana."
href={nodeExporterDashboardUrl} href={nodeExporterDashboardUrl}
/>
<GrafanaLinkCard
title={`${selectedMachine.name} logs`}
description="Explore Loki logs for this machine in Grafana."
href={logsUrl}
/>
</>
) : (
<EmptyState
icon={Gauge}
title="No Grafana service configured"
description="Add a Grafana service instance to enable deep-links to dashboards and logs."
action={
<Button variant="outline" size="sm" asChild>
<Link to="/services">Open Services</Link>
</Button>
}
/> />
<GrafanaLinkCard )
title={`${selectedMachine.name} logs`}
description="Explore Loki logs for this machine in Grafana."
href={logsUrl}
/>
</>
) : ( ) : (
<EmptyState
icon={Gauge}
title="No Grafana service configured"
description="Add a Grafana service instance to enable deep-links to dashboards and logs."
action={
<Button variant="outline" size="sm" asChild>
<Link to="/services">Open Services</Link>
</Button>
}
/>
)
) : (
<EmptyState <EmptyState
icon={ServerOff} icon={ServerOff}
title="No machine selected" title="No machine selected"