feat: add in-memory limiter and breaker satisfying backend contracts

This commit is contained in:
2026-07-20 06:54:07 -04:00
parent 454a8b5e0f
commit 4a176b6220
9 changed files with 645 additions and 0 deletions
+160
View File
@@ -0,0 +1,160 @@
"""进程内熔断门: CHS gate 契约的内存实现(D3 双后端)。
状态机蓝本 VT `adapters/breaker.py`(闭路→阈值开路→冷却半开→单探针),
契约形态承 CHS `provider_gate.py`: 半开探针是**带 TTL 的租约**(持有者
死亡后可被接管,防"探针永远在路上"死锁),写回经 epoch fencing 拒绝
旧世代污染。epoch 在每次进入 OPEN 时递增。时钟构造注入,纯确定性可测。
"""
from __future__ import annotations
import time
from dataclasses import dataclass, field
from typing import TYPE_CHECKING
from polygateway.ports import GateDecision, GateState, GateUpdate
if TYPE_CHECKING:
from collections.abc import Callable
from polygateway.types import BreakerConfig
@dataclass
class _SourceGate:
"""单源的门控可变状态。"""
state: GateState = GateState.CLOSED
fails: int = 0
epoch: int = 0
open_until: float = 0.0
probe_owner: str | None = None
probe_expires: float = 0.0
reasons: dict[str, str] = field(default_factory=dict)
class InMemoryGate:
"""按 source_name 分别计数的进程内熔断门。"""
def __init__(self, *, config: BreakerConfig, now: Callable[[], float] = time.monotonic) -> None:
self._cfg = config
self._now = now
self._gates: dict[str, _SourceGate] = {}
def _gate(self, source_name: str) -> _SourceGate:
if not source_name.strip():
raise ValueError("source_name 不能为空")
return self._gates.setdefault(source_name, _SourceGate())
def _grant_probe(self, g: _SourceGate, source_name: str, owner: str) -> GateDecision:
g.state = GateState.HALF_OPEN
g.probe_owner = owner
g.probe_expires = self._now() + self._cfg.probe_ttl_s
return GateDecision(
source_name=source_name, allowed=True, state=GateState.HALF_OPEN,
epoch=g.epoch, is_probe=True, probe_owner=owner, retry_after_s=0.0,
)
async def try_enter(self, source_name: str, owner: str) -> GateDecision:
"""健康普通准入;冷却到期/探针租约过期时原子授予唯一探针。"""
if not owner.strip():
raise ValueError("owner 不能为空")
g = self._gate(source_name)
now = self._now()
if g.state is GateState.CLOSED:
return GateDecision(
source_name=source_name, allowed=True, state=GateState.CLOSED,
epoch=g.epoch, is_probe=False, probe_owner=None, retry_after_s=0.0,
)
if g.state is GateState.OPEN:
if now >= g.open_until:
return self._grant_probe(g, source_name, owner)
return GateDecision(
source_name=source_name, allowed=False, state=GateState.OPEN,
epoch=g.epoch, is_probe=False, probe_owner=None,
retry_after_s=g.open_until - now,
)
# HALF_OPEN: 探针在途;租约过期则接管,否则拒绝(防惊群)
if now >= g.probe_expires:
return self._grant_probe(g, source_name, owner)
return GateDecision(
source_name=source_name, allowed=False, state=GateState.HALF_OPEN,
epoch=g.epoch, is_probe=False, probe_owner=None,
retry_after_s=g.probe_expires - now,
)
def _fenced(self, g: _SourceGate, entry: GateDecision) -> bool:
"""写回资格: 世代一致;探针写回还要求 owner 仍在位(CHS fencing 同款)。"""
if not entry.allowed:
raise ValueError("被拒决定不得写回")
if entry.epoch != g.epoch:
return False
return not (
entry.is_probe
and (g.state is not GateState.HALF_OPEN or g.probe_owner != entry.probe_owner)
)
def _snapshot(self, g: _SourceGate, applied: bool) -> GateUpdate:
return GateUpdate(
applied=applied, state=g.state, epoch=g.epoch, failure_count=g.fails,
retry_after_s=max(0.0, g.open_until - self._now()) if g.state is GateState.OPEN else 0.0,
)
def _open(self, g: _SourceGate, reason: str) -> None:
g.state = GateState.OPEN
g.epoch += 1 # 世代推进: 旧 entry 的迟到写回自此被 fencing 拒绝
g.open_until = self._now() + self._cfg.cooldown_s
g.fails = max(g.fails, self._cfg.fail_threshold)
g.probe_owner = None
g.probe_expires = 0.0
async def record_success(self, entry: GateDecision) -> GateUpdate:
g = self._gate(entry.source_name)
if not self._fenced(g, entry):
return self._snapshot(g, applied=False)
g.state = GateState.CLOSED
g.fails = 0
g.probe_owner = None
g.probe_expires = 0.0
return self._snapshot(g, applied=True)
async def record_failure(self, entry: GateDecision, reason: str, force_open: bool) -> GateUpdate:
g = self._gate(entry.source_name)
if not self._fenced(g, entry):
return self._snapshot(g, applied=False)
if entry.is_probe or force_open:
self._open(g, reason) # 探针失败重开 / SourceDead 一击即熔
return self._snapshot(g, applied=True)
g.fails += 1
if g.fails >= self._cfg.fail_threshold:
self._open(g, reason)
return self._snapshot(g, applied=True)
async def release_probe(self, entry: GateDecision) -> GateUpdate:
"""探针无果归还(如取消): 源保持可接管状态让下一 caller 接手;幂等。"""
if not (entry.allowed and entry.is_probe and entry.probe_owner is not None):
raise ValueError("release_probe 只接受在途探针决定")
g = self._gate(entry.source_name)
if not self._fenced(g, entry):
return self._snapshot(g, applied=False)
g.state = GateState.OPEN
g.open_until = self._now() # 立即可被下一 caller 以探针身份接管
g.probe_owner = None
g.probe_expires = 0.0
return self._snapshot(g, applied=True)
async def retry_after_s(self, sources: tuple[str, ...]) -> float:
"""集合中最早可尝试时间;健康/到期返回 0。"""
if not sources:
raise ValueError("sources 不能为空")
now = self._now()
waits = []
for name in sources:
g = self._gate(name)
if g.state is GateState.OPEN:
waits.append(max(0.0, g.open_until - now))
elif g.state is GateState.HALF_OPEN:
waits.append(max(0.0, g.probe_expires - now))
else:
waits.append(0.0)
return min(waits)
+170
View File
@@ -0,0 +1,170 @@
"""进程内限流后端: 与 Redis 版同一契约的六道闸实现(D3 双后端)。
语义蓝本 CHS `app/coordination/limiter.py`: 并发 = 带 TTL 的租约(持有者
死亡后过期回收);RPM/TPM = 分钟滑动窗口计数;TPM 入场按 est 预扣,settle
按实际结算多退少补且退款落 acquire 时的窗口。检查-占用在单次同步段内完成
(无 await 穿插),单事件循环下天然原子;本实现不跨进程,是单进程部署的
正确答案(多 worker 用 M2 Redis 后端)。
"""
from __future__ import annotations
import asyncio
import time
import uuid
from typing import TYPE_CHECKING
from polygateway.errors import GovernanceBackendError
from polygateway.types import GlobalLimits, SourceConfig, SourceStats
if TYPE_CHECKING:
from collections.abc import Callable
_GLOBAL = "__global__"
_WINDOW_S = 60.0
class _MemoryPermit:
"""入场许可;release/settle 幂等(CHS _RedisPermit 同款 flag 语义)。"""
def __init__(self, limiter: InMemoryLimiter, source: str, lease_id: str, est: int, window: int) -> None:
self._limiter = limiter
self._source = source
self._lease_id = lease_id
self._est = est
self._window = window
self._released = False
self._settled = False
async def release(self) -> None:
if self._released:
return
self._released = True
self._limiter._release_lease(self._source, self._lease_id)
async def settle(self, actual_tokens: int) -> None:
if self._settled:
return
self._settled = True
delta = actual_tokens - self._est
if delta:
self._limiter._settle_tpm(self._source, delta, self._window)
class InMemoryLimiter:
"""六道闸: 全局/单源 × 并发/RPM/TPM;限额 0 = 该闸不启用。"""
def __init__(
self,
*,
scope: str,
sources: dict[str, SourceConfig],
global_limits: GlobalLimits,
lease_ttl_s: float = 1500.0,
now: Callable[[], float] = time.monotonic,
sleep: Callable[[float], object] = asyncio.sleep,
poll_interval_s: float = 0.05,
) -> None:
if lease_ttl_s <= 0 or poll_interval_s <= 0:
raise ValueError("lease_ttl_s 与 poll_interval_s 必须 > 0")
self._scope = scope
self._sources = dict(sources)
self._global = global_limits
self._lease_ttl_s = lease_ttl_s
self._now = now
self._sleep = sleep
self._poll_interval_s = poll_interval_s
# source → lease_id → 过期时刻
self._leases: dict[str, dict[str, float]] = {name: {} for name in self._sources}
# (source|__global__, window_id) → 计数
self._rpm: dict[tuple[str, int], int] = {}
self._tpm: dict[tuple[str, int], int] = {}
self._progress_at: float | None = None
# —— 内部原语(同步,单事件循环下原子)——
def _cfg(self, source_key: str) -> SourceConfig:
cfg = self._sources.get(source_key)
if cfg is None:
raise GovernanceBackendError(f"未知源 {source_key!r}(scope={self._scope})")
return cfg
def _window(self) -> int:
return int(self._now() / _WINDOW_S)
def _purge_leases(self) -> None:
now = self._now()
for leases in self._leases.values():
expired = [lid for lid, expiry in leases.items() if expiry <= now]
for lid in expired:
del leases[lid]
def _inflight(self, source_key: str) -> int:
return len(self._leases[source_key])
def _inflight_total(self) -> int:
return sum(len(leases) for leases in self._leases.values())
def _release_lease(self, source_key: str, lease_id: str) -> None:
self._leases[source_key].pop(lease_id, None)
def _settle_tpm(self, source_key: str, delta: int, window: int) -> None:
# 退款/补记落 acquire 时的窗口,计数下限 0(CHS Lua 同款)
for key in (source_key, _GLOBAL):
slot = (key, window)
self._tpm[slot] = max(0, self._tpm.get(slot, 0) + delta)
def _gates_pass(self, cfg: SourceConfig, est_tokens: int, window: int) -> bool:
checks = (
(self._global.max_concurrency, self._inflight_total() + 1),
(cfg.max_concurrency, self._inflight(cfg.name) + 1),
(self._global.rpm, self._rpm.get((_GLOBAL, window), 0) + 1),
(cfg.rpm, self._rpm.get((cfg.name, window), 0) + 1),
(self._global.tpm, self._tpm.get((_GLOBAL, window), 0) + est_tokens),
(cfg.tpm, self._tpm.get((cfg.name, window), 0) + est_tokens),
)
return all(limit <= 0 or would_be <= limit for limit, would_be in checks)
# —— 契约方法 ——
async def try_acquire(self, source_key: str, est_tokens: int) -> _MemoryPermit | None:
"""非阻塞准入: 六闸全过才占用;任一满则返回 None 且零副作用。"""
cfg = self._cfg(source_key)
self._purge_leases()
window = self._window()
if not self._gates_pass(cfg, est_tokens, window):
return None
lease_id = uuid.uuid4().hex
self._leases[source_key][lease_id] = self._now() + self._lease_ttl_s
for key in (source_key, _GLOBAL):
self._rpm[(key, window)] = self._rpm.get((key, window), 0) + 1
if est_tokens:
self._tpm[(key, window)] = self._tpm.get((key, window), 0) + est_tokens
return _MemoryPermit(self, source_key, lease_id, est_tokens, window)
async def acquire(self, source_key: str, est_tokens: int) -> _MemoryPermit:
"""阻塞准入: 轮询直到拿到 permit;sleep 可被取消(取消穿透铁律)。"""
while True:
permit = await self.try_acquire(source_key, est_tokens)
if permit is not None:
return permit
await self._sleep(self._poll_interval_s)
async def source_stats(self, source_key: str) -> SourceStats:
self._cfg(source_key)
self._purge_leases()
window = self._window()
return SourceStats(
inflight=self._inflight(source_key),
rpm_used=self._rpm.get((source_key, window), 0),
tpm_used=self._tpm.get((source_key, window), 0),
)
async def mark_progress(self) -> None:
"""记录"最近一次出餐"时刻,供背压 stall 判定(M2)读取。"""
self._progress_at = self._now()
async def progress_age_s(self) -> float:
if self._progress_at is None:
return float("inf")
return self._now() - self._progress_at