bvandeusen
faecac3ec6
Bounds Docker time-series growth (the main scaling concern). New docker_metrics_hourly table + docker_006 migration; a plugin retention module (docker.run_retention capability) rolls raw docker_metrics older than the raw window into hourly averages (idempotent upsert), deletes the rolled raw rows, then prunes stale rollups + lifecycle events. Core cleanup.py drives it each hourly run via the capability (no plugin-model import), reading the three retention windows fresh from settings so changes apply without restart (rule 25). Settings → "Thresholds & Retention" gains a Docker retention card (raw / rolled-up / events windows, working defaults 7/90/30 days). Unit tests cover the hour-aligned cutoff/bucketing helpers; integration test exercises the real rollup-average + prune across both windows. Milestone 77 task #941. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_016Jg27rgypiW2efULXJDtMC
68 lines
2.8 KiB
Python
68 lines
2.8 KiB
Python
from __future__ import annotations
|
|
import logging
|
|
from datetime import datetime, timedelta, timezone
|
|
from typing import TYPE_CHECKING
|
|
|
|
from sqlalchemy import delete
|
|
|
|
from steward.models.monitors import MonitorResult
|
|
from steward.models.metrics import PluginMetric
|
|
from steward.models.ansible import AnsibleRun
|
|
|
|
if TYPE_CHECKING:
|
|
from quart import Quart
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
async def run_cleanup(app: "Quart") -> None:
|
|
"""Delete rows older than DATA_RETENTION_DAYS from time-series tables, then
|
|
run Docker-specific rollup + retention (delegated to the docker plugin)."""
|
|
retention_days: int = app.config.get("DATA_RETENTION_DAYS", 90)
|
|
now = datetime.now(timezone.utc)
|
|
cutoff = now - timedelta(days=retention_days)
|
|
|
|
async with app.db_sessionmaker() as session:
|
|
async with session.begin():
|
|
for model, ts_col in [
|
|
(MonitorResult, MonitorResult.checked_at),
|
|
(PluginMetric, PluginMetric.recorded_at),
|
|
(AnsibleRun, AnsibleRun.started_at),
|
|
]:
|
|
result = await session.execute(
|
|
delete(model).where(ts_col < cutoff)
|
|
)
|
|
if result.rowcount:
|
|
logger.info(f"Pruned {result.rowcount} rows from {model.__tablename__}")
|
|
|
|
await _run_docker_retention(session, now)
|
|
|
|
|
|
async def _run_docker_retention(session, now: datetime) -> None:
|
|
"""Drive the docker plugin's rollup + prune via its capability, if loaded.
|
|
|
|
Windows are read fresh from settings each run (rule 25 — a change in the
|
|
Settings UI takes effect on the next hourly cleanup, no restart). Kept in its
|
|
own transaction so a docker-side failure can't roll back the generic prune
|
|
above. No-op when the docker plugin is disabled (capability absent).
|
|
"""
|
|
from steward.core.capabilities import has_capability, invoke_capability
|
|
if not has_capability("docker.run_retention"):
|
|
return
|
|
from steward.core.settings import get_setting
|
|
from steward.models.users import UserRole
|
|
|
|
# Reads + rollup/prune share one transaction — get_setting's SELECT would
|
|
# otherwise autobegin one, making a later session.begin() raise.
|
|
async with session.begin():
|
|
raw_days = int(await get_setting(session, "docker.retention.metrics_raw_days") or 7)
|
|
rollup_days = int(await get_setting(session, "docker.retention.metrics_rollup_days") or 90)
|
|
events_days = int(await get_setting(session, "docker.retention.events_days") or 30)
|
|
counts = await invoke_capability(
|
|
"docker.run_retention", UserRole.viewer, session,
|
|
events_days=events_days, metrics_raw_days=raw_days,
|
|
metrics_rollup_days=rollup_days, now=now,
|
|
)
|
|
if counts and any(counts.values()):
|
|
logger.info("Docker retention: %s", counts)
|