明细只保留 1 天,超过 1 天聚合到新表 channel_monitor_daily_rollups(按 monitor_id/model/bucket_date 维度),聚合保留 30 天。两张表都用 SoftDeleteMixin 软删除(DELETE 自动改为 UPDATE deleted_at = NOW())。 聚合 + 清理任务由 OpsCleanupService 的 cron 统一调度,与运维监控的清理共享 schedule(默认 0 2 * * *)和 leader lock。ChannelMonitorRunner 的 cleanupLoop 被移除,只保留 dueCheckLoop。 读取路径 ComputeAvailability* 改为 UNION 明细(今天 deleted_at IS NULL)+ 聚合(过去 windowDays 天 deleted_at IS NULL),SUM(ok)/SUM(total) 自然加权 计算可用率,AVG latency 用 SUM(sum_latency_ms)/SUM(count_latency)。 watermark 表 channel_monitor_aggregation_watermark 单行(id=1),记录 last_aggregated_date,重启后从该日期 +1 继续聚合,首次为 nil 则从 today - 30d 开始回填,单次最多 35 天上限避免长事务。 raw SQL 的 ListLatestPerModel / ListLatestForMonitorIDs / ListRecentHistoryForMonitors 都补上 deleted_at IS NULL 过滤(SoftDeleteMixin interceptor 只对 ent query 生效)。 bump version to 0.1.114.28 GroupBadge 在 MonitorKeyPickerDialog 中复用平台主题色 + 倍率/专属倍率 (顺手优化)。
61 lines
3.1 KiB
SQL
61 lines
3.1 KiB
SQL
-- Migration: 126_add_channel_monitor_aggregation
|
||
-- 渠道监控日聚合:把 channel_monitor_histories 的明细按天聚合,明细只保留 1 天,
|
||
-- 聚合保留 30 天。明细和聚合表都用软删除(deleted_at),由 ops cleanup 任务每天
|
||
-- 凌晨随运维监控清理一起跑(共享 cron)。
|
||
--
|
||
-- 设计要点:
|
||
-- - channel_monitor_histories 加 deleted_at 软删除字段(SoftDeleteMixin 全局
|
||
-- Hook 会把 DELETE 自动改写成 UPDATE deleted_at = NOW())。
|
||
-- - channel_monitor_daily_rollups 按 (monitor_id, model, bucket_date) 唯一,
|
||
-- 用 ON CONFLICT DO UPDATE 实现幂等回填,状态分布和延迟分子分母都保留,
|
||
-- 方便后续按窗口任意求加权可用率和均值。
|
||
-- - watermark 表只有一行(id=1),记录最近一次聚合到达的日期,避免重启后重复
|
||
-- 扫全表。
|
||
-- - rollup 上 (bucket_date) 索引服务清理任务的 DELETE WHERE bucket_date < cutoff。
|
||
|
||
-- 1) 给历史明细表加软删除字段
|
||
ALTER TABLE channel_monitor_histories
|
||
ADD COLUMN IF NOT EXISTS deleted_at TIMESTAMPTZ;
|
||
|
||
CREATE INDEX IF NOT EXISTS idx_channel_monitor_histories_deleted_at
|
||
ON channel_monitor_histories (deleted_at);
|
||
|
||
-- 2) 创建日聚合表
|
||
CREATE TABLE IF NOT EXISTS channel_monitor_daily_rollups (
|
||
id BIGSERIAL PRIMARY KEY,
|
||
monitor_id BIGINT NOT NULL REFERENCES channel_monitors(id) ON DELETE CASCADE,
|
||
model VARCHAR(200) NOT NULL,
|
||
bucket_date DATE NOT NULL,
|
||
total_checks INT NOT NULL DEFAULT 0,
|
||
ok_count INT NOT NULL DEFAULT 0,
|
||
operational_count INT NOT NULL DEFAULT 0,
|
||
degraded_count INT NOT NULL DEFAULT 0,
|
||
failed_count INT NOT NULL DEFAULT 0,
|
||
error_count INT NOT NULL DEFAULT 0,
|
||
sum_latency_ms BIGINT NOT NULL DEFAULT 0,
|
||
count_latency INT NOT NULL DEFAULT 0,
|
||
sum_ping_latency_ms BIGINT NOT NULL DEFAULT 0,
|
||
count_ping_latency INT NOT NULL DEFAULT 0,
|
||
computed_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
||
deleted_at TIMESTAMPTZ
|
||
);
|
||
|
||
CREATE UNIQUE INDEX IF NOT EXISTS idx_channel_monitor_daily_rollups_unique
|
||
ON channel_monitor_daily_rollups (monitor_id, model, bucket_date);
|
||
CREATE INDEX IF NOT EXISTS idx_channel_monitor_daily_rollups_bucket
|
||
ON channel_monitor_daily_rollups (bucket_date);
|
||
CREATE INDEX IF NOT EXISTS idx_channel_monitor_daily_rollups_deleted_at
|
||
ON channel_monitor_daily_rollups (deleted_at);
|
||
|
||
-- 3) 创建 watermark 表(单行:id=1)
|
||
CREATE TABLE IF NOT EXISTS channel_monitor_aggregation_watermark (
|
||
id INT PRIMARY KEY DEFAULT 1,
|
||
last_aggregated_date DATE,
|
||
updated_at TIMESTAMPTZ NOT NULL DEFAULT NOW(),
|
||
CONSTRAINT channel_monitor_aggregation_watermark_singleton CHECK (id = 1)
|
||
);
|
||
|
||
INSERT INTO channel_monitor_aggregation_watermark (id, last_aggregated_date, updated_at)
|
||
VALUES (1, NULL, NOW())
|
||
ON CONFLICT (id) DO NOTHING;
|