anti
f775223a83
Adds decnet.fleet.reconciler — a pure async function plus a long-lived
worker — that periodically reconciles the three sources of truth on a
DECNET host:
1. decnet-state.json (CLI-canonical fleet record)
2. fleet_deckies table (DB mirror, written by engine.deployer)
3. docker inspect (actual per-container runtime state)
Drift handling:
* JSON has X, DB doesn't → INSERT (deploy ran with DB offline)
* DB has X (this host), JSON doesn't → DELETE (teardown ran with DB offline)
* Both have X, docker disagrees → flip state to running/failed/degraded
* Docker socket unreachable → leave existing state alone (don't
torch every row to torn_down)
Cross-host safety: deletions are scoped to host_uuid for the local host;
a master that runs both a local fleet and swarm workers will never
clobber a peer's slice.
CLI:
decnet reconcile --once # one-shot, prints counts
decnet reconcile [--interval N] # long-lived worker, mirrors
# orchestrator's lifecycle (control
# listener + heartbeat + tick loop)
Promotes decnet/fleet.py → decnet/fleet/ package so the reconciler can
live alongside it without name collision (build_deckies_from_ini and
all_service_names re-exported unchanged via __init__.py).
14 new tests cover state aggregation rules, all four drift directions,
host_uuid scoping, docker-unreachable safety, and worker shutdown via
the bus control event.
85 lines
2.8 KiB
Python
85 lines
2.8 KiB
Python
"""Long-lived periodic reconciler worker.
|
|
|
|
Modeled on :mod:`decnet.orchestrator.worker`: same control listener, same
|
|
heartbeat helper, same shutdown semantics. One tick = one
|
|
:func:`reconcile_once` pass.
|
|
|
|
Default interval is short (30s) because reconciliation is cheap when
|
|
nothing has drifted (three reads, no writes), and a short cadence keeps
|
|
the dashboard's view of crashed containers fresh.
|
|
"""
|
|
from __future__ import annotations
|
|
|
|
import asyncio
|
|
import contextlib
|
|
|
|
from decnet.bus.factory import get_bus
|
|
from decnet.bus.publish import (
|
|
run_control_listener,
|
|
run_health_heartbeat,
|
|
)
|
|
from decnet.fleet.reconciler import reconcile_once
|
|
from decnet.logging import get_logger
|
|
from decnet.web.db.models import LOCAL_HOST_SENTINEL
|
|
from decnet.web.db.repository import BaseRepository
|
|
|
|
logger = get_logger("fleet.reconciler")
|
|
|
|
|
|
async def fleet_reconciler_worker(
|
|
repo: BaseRepository,
|
|
*,
|
|
interval: int = 30,
|
|
host_uuid: str = LOCAL_HOST_SENTINEL,
|
|
) -> None:
|
|
"""Periodically converge JSON ↔ DB ↔ docker for the local host.
|
|
|
|
Honours the bus control topic (``system.reconciler.control``) for
|
|
graceful shutdown — same lifecycle contract as every other DECNET
|
|
worker.
|
|
"""
|
|
logger.info("fleet reconciler started interval=%ds host=%s", interval, host_uuid)
|
|
|
|
bus = None
|
|
try:
|
|
bus = get_bus(client_name="reconciler")
|
|
await bus.connect()
|
|
except Exception as exc: # noqa: BLE001
|
|
logger.warning(
|
|
"reconciler: bus unavailable, continuing without publish: %s", exc,
|
|
)
|
|
bus = None
|
|
|
|
shutdown = asyncio.Event()
|
|
heartbeat_task = asyncio.create_task(run_health_heartbeat(bus, "reconciler"))
|
|
control_task = asyncio.create_task(
|
|
run_control_listener(bus, "reconciler", shutdown),
|
|
)
|
|
|
|
try:
|
|
while not shutdown.is_set():
|
|
try:
|
|
await asyncio.wait_for(shutdown.wait(), timeout=interval)
|
|
except asyncio.TimeoutError:
|
|
pass # normal tick
|
|
if shutdown.is_set():
|
|
break
|
|
try:
|
|
counts = await reconcile_once(repo, host_uuid=host_uuid)
|
|
if any(counts.values()):
|
|
logger.info(
|
|
"reconcile inserted=%d deleted=%d state_updated=%d",
|
|
counts["inserted"], counts["deleted"],
|
|
counts["state_updated"],
|
|
)
|
|
except Exception as exc: # noqa: BLE001
|
|
logger.error("reconcile tick failed: %s", exc)
|
|
finally:
|
|
for t in (heartbeat_task, control_task):
|
|
t.cancel()
|
|
with contextlib.suppress(Exception, asyncio.CancelledError):
|
|
await t
|
|
if bus is not None:
|
|
with contextlib.suppress(Exception):
|
|
await bus.close()
|