Lukas
1cdc97a729
- 11.1 MainLayout health indicator: warning MessageBar when fail2ban offline - 11.2 formatDate utility + TimezoneProvider + GET /api/setup/timezone - 11.3 Responsive sidebar: auto-collapse <640px, media query listener - 11.4 PageFeedback (PageLoading/PageError/PageEmpty), BanTable updated - 11.5 prefers-reduced-motion: disable sidebar transition - 11.6 WorldMap ARIA: role/tabIndex/aria-label/onKeyDown for countries - 11.7 Health transition logging (fail2ban_came_online/went_offline) - 11.8 Global handlers: Fail2BanConnectionError/ProtocolError -> 502 - 11.9 379 tests pass, 82% coverage, ruff+mypy+tsc+eslint clean - Timezone endpoint: setup_service.get_timezone, 5 new tests
90 lines
2.9 KiB
Python
90 lines
2.9 KiB
Python
"""Health-check background task.
|
|
|
|
Registers an APScheduler job that probes the fail2ban socket every 30 seconds
|
|
and stores the result on ``app.state.server_status``. The dashboard endpoint
|
|
reads from this cache, keeping HTTP responses fast and the daemon connection
|
|
decoupled from user-facing requests.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
from typing import TYPE_CHECKING, Any
|
|
|
|
import structlog
|
|
|
|
from app.models.server import ServerStatus
|
|
from app.services import health_service
|
|
|
|
if TYPE_CHECKING: # pragma: no cover
|
|
from fastapi import FastAPI
|
|
|
|
log: structlog.stdlib.BoundLogger = structlog.get_logger()
|
|
|
|
#: How often the probe fires (seconds).
|
|
HEALTH_CHECK_INTERVAL: int = 30
|
|
|
|
|
|
async def _run_probe(app: Any) -> None:
|
|
"""Probe fail2ban and cache the result on *app.state*.
|
|
|
|
This is the APScheduler job callback. It reads ``fail2ban_socket`` from
|
|
``app.state.settings``, runs the health probe, and writes the result to
|
|
``app.state.server_status``.
|
|
|
|
Args:
|
|
app: The :class:`fastapi.FastAPI` application instance passed by the
|
|
scheduler via the ``kwargs`` mechanism.
|
|
"""
|
|
socket_path: str = app.state.settings.fail2ban_socket
|
|
prev_status: ServerStatus = getattr(
|
|
app.state, "server_status", ServerStatus(online=False)
|
|
)
|
|
status: ServerStatus = await health_service.probe(socket_path)
|
|
app.state.server_status = status
|
|
|
|
# Log transitions between online and offline states.
|
|
if status.online and not prev_status.online:
|
|
log.info("fail2ban_came_online", version=status.version)
|
|
elif not status.online and prev_status.online:
|
|
log.warning("fail2ban_went_offline")
|
|
|
|
log.debug(
|
|
"health_check_complete",
|
|
online=status.online,
|
|
version=status.version,
|
|
active_jails=status.active_jails,
|
|
)
|
|
|
|
|
|
def register(app: FastAPI) -> None:
|
|
"""Add the health-check job to the application scheduler.
|
|
|
|
Must be called after the scheduler has been started (i.e., inside the
|
|
lifespan handler, after ``scheduler.start()``).
|
|
|
|
Args:
|
|
app: The :class:`fastapi.FastAPI` application instance whose
|
|
``app.state.scheduler`` will receive the job.
|
|
"""
|
|
# Initialise the cache with an offline placeholder so the dashboard
|
|
# endpoint is always able to return a valid response even before the
|
|
# first probe fires.
|
|
app.state.server_status = ServerStatus(online=False)
|
|
|
|
app.state.scheduler.add_job(
|
|
_run_probe,
|
|
trigger="interval",
|
|
seconds=HEALTH_CHECK_INTERVAL,
|
|
kwargs={"app": app},
|
|
id="health_check",
|
|
replace_existing=True,
|
|
# Fire immediately on startup too, so the UI isn't dark for 30 s.
|
|
next_run_time=__import__("datetime").datetime.now(
|
|
tz=__import__("datetime").timezone.utc
|
|
),
|
|
)
|
|
log.info(
|
|
"health_check_scheduled",
|
|
interval_seconds=HEALTH_CHECK_INTERVAL,
|
|
)
|