feat(log): decouple log storage from ClickHouse with switchable logstore

- New internal/repository/logstore abstraction: exported domain interfaces
  (AccessLogStore/ObservabilityStore/UserAccessLogStore/StatusStore),
  config-driven provider (Active/Build/Migrating/SetConfigReader), GORM
  implementation for PostgreSQL/SQLite (incl. hourly rollups computed in
  real time, migration listers, PG partition maintenance), and a ClickHouse
  wrapper preserving the native batch path; repository facade delegates to
  logstore; import-lint test enforces apps never import analyticsrepo.
- ClickHouse is now optional: the log DB is either the main DB (postgres
  when database.enabled, else sqlite) or clickhouse; boot validation +
  first-run seed; log_database / log_db_migration are protected keys.
- New user task 切换日志数据库 (of_log_db_switch): freeze log writes,
  drain batch writers, copy all 6 raw log tables by id (preserving IDs)
  with target-partition pre-creation for PG, flip log_database on success,
  clear the freeze flag on failure.
- Per-store retention (log_retention_days_*) with expiry cleanup folded
  into the daily system_cleanup task; legacy database_auto_cleanup_* and
  of_database_auto_cleanup decommissioned.
- goose migrations: 6 log tables in PG (2 monthly-partitioned) + SQLite,
  retention config seeds, schedule cleanup; GET
  /api/v1/admin/status/log-database endpoint; frontend retention settings,
  switch-task UI and status badge; changelog and docs updated.

docs(plan): log database decoupling implementation plan

docs(design): log database decoupling design (ClickHouse optional)
This commit is contained in:
ryan
2026-08-08 11:36:56 +08:00
parent 734fe45baa
commit 7d71f1e4e1
95 changed files with 10569 additions and 3728 deletions
@@ -0,0 +1,731 @@
// Copyright 2026 Arctel.net
// SPDX-License-Identifier: Apache-2.0
package logstore
import (
"context"
"errors"
"fmt"
"math"
"time"
"github.com/ClickHouse/clickhouse-go/v2/lib/driver"
db "github.com/Rain-kl/Wavelet/internal/infra/persistence"
"github.com/Rain-kl/Wavelet/internal/model"
analyticsmodel "github.com/Rain-kl/Wavelet/internal/model/analytics"
analyticsrepo "github.com/Rain-kl/Wavelet/internal/repository/analytics"
)
// clickhouseLogStore 实现 AccessLogStore / ObservabilityStore / StatusStore,
// 逐方法委托 analyticsrepo(CH 原生 batch 写入,零性能损耗)。
// UserAccessLogStore 由 clickhouseUserAccessLogStore 实现(List/Count 方法名已被
// AccessLogStore 占用,Go 不允许同名不同签名方法)。
type clickhouseLogStore struct {
// skipFreeze 为 true 时跳过迁移冻结检查(仅迁移目标 store 使用)。
skipFreeze bool
}
func newClickHouseStore() *clickhouseLogStore { return &clickhouseLogStore{} }
// 编译期断言。
var (
_ AccessLogStore = (*clickhouseLogStore)(nil)
_ ObservabilityStore = (*clickhouseLogStore)(nil)
_ StatusStore = (*clickhouseLogStore)(nil)
_ UserAccessLogStore = (*clickhouseUserAccessLogStore)(nil)
)
func chConnErr() error {
if !db.ChConnReady() {
return errors.New("clickhouse connection is not initialized")
}
return nil
}
// ensureWritable 迁移冻结期拒绝写入。
func (s *clickhouseLogStore) ensureWritable(ctx context.Context) error {
if !s.skipFreeze && Migrating(ctx) {
return ErrMigrating
}
return nil
}
// ---- AccessLogStore ----
// InsertBatch 节点访问日志写入入口:冻结检查后经 hook 入队(异步),不直接落库。
func (s *clickhouseLogStore) InsertBatch(ctx context.Context, records []*model.OpenFlareAccessLog) error {
if err := s.ensureWritable(ctx); err != nil {
return err
}
rows := make([]analyticsmodel.NodeAccessLog, 0, len(records))
for _, r := range records {
if r == nil {
continue
}
rows = append(rows, toAnalyticsNodeAccessLog(r))
}
if h := currentAccessLogHooks().QueueNodeAccessLogs; h != nil {
h(rows)
}
return nil
}
// BatchInsertNodeAccessLogs 是 batchwriter flush 目标:CH 原生批量写入。
func (s *clickhouseLogStore) BatchInsertNodeAccessLogs(ctx context.Context, rows []analyticsmodel.NodeAccessLog) error {
if err := s.ensureWritable(ctx); err != nil {
return err
}
return analyticsrepo.BatchInsertNodeAccessLogs(ctx, rows)
}
func (s *clickhouseLogStore) List(ctx context.Context, query model.OpenFlareAccessLogQuery) ([]*model.OpenFlareAccessLog, error) {
rows, err := analyticsrepo.ListNodeAccessLogs(ctx, toNodeAccessLogFilter(query))
if err != nil {
return nil, err
}
return fromAnalyticsNodeAccessLogs(rows), nil
}
func (s *clickhouseLogStore) Count(ctx context.Context, query model.OpenFlareAccessLogQuery) (int64, int64, int64, error) {
return analyticsrepo.CountNodeAccessLogs(ctx, toNodeAccessLogFilter(query))
}
func (s *clickhouseLogStore) RegionCounts(ctx context.Context, nodeID string, since time.Time, limit int) ([]*model.OpenFlareAccessLogRegionCount, error) {
rows, err := analyticsrepo.RegionCountsNodeAccessLogs(ctx, nodeID, since, limit)
if err != nil {
return nil, err
}
out := make([]*model.OpenFlareAccessLogRegionCount, len(rows))
for i, r := range rows {
out[i] = &model.OpenFlareAccessLogRegionCount{Region: r.Region, Count: r.Count}
}
return out, nil
}
func (s *clickhouseLogStore) BucketAggregates(ctx context.Context, query model.OpenFlareAccessLogQuery, bucketSeconds int64) ([]analyticsmodel.NodeAccessLogBucketAggregate, error) {
return analyticsrepo.BucketAggregatesNodeAccessLogs(ctx, toNodeAccessLogFilter(query), bucketSeconds)
}
func (s *clickhouseLogStore) CountBuckets(ctx context.Context, query model.OpenFlareAccessLogQuery, bucketSeconds int64) (int64, error) {
return analyticsrepo.CountBucketAggregatesNodeAccessLogs(ctx, toNodeAccessLogFilter(query), bucketSeconds)
}
func (s *clickhouseLogStore) BucketDimensions(ctx context.Context, query model.OpenFlareAccessLogQuery, column string, bucketSeconds int64) ([]analyticsmodel.NodeAccessLogBucketDimension, error) {
return analyticsrepo.BucketDimensionsNodeAccessLogs(ctx, toNodeAccessLogFilter(query), column, bucketSeconds)
}
func (s *clickhouseLogStore) IPAggregates(ctx context.Context, query model.OpenFlareAccessLogQuery, exactRemoteAddr bool) ([]analyticsmodel.NodeAccessLogIPAggregate, error) {
return analyticsrepo.IPAggregatesNodeAccessLogs(ctx, toNodeAccessLogFilter(query), exactRemoteAddr)
}
func (s *clickhouseLogStore) IPSummaries(ctx context.Context, query model.OpenFlareAccessLogQuery, recentSince time.Time) ([]analyticsmodel.NodeAccessLogIPSummary, error) {
return analyticsrepo.IPSummariesNodeAccessLogs(ctx, toNodeAccessLogFilter(query), recentSince)
}
func (s *clickhouseLogStore) CountIPSummaries(ctx context.Context, query model.OpenFlareAccessLogQuery) (int64, error) {
return analyticsrepo.CountIPSummaryNodeAccessLogs(ctx, toNodeAccessLogFilter(query))
}
func (s *clickhouseLogStore) WAFIPAggregates(ctx context.Context, query model.OpenFlareAccessLogQuery) ([]analyticsmodel.NodeAccessLogWAFIPAggregate, error) {
return analyticsrepo.IPAggregatesForWAFNodeAccessLogs(ctx, toNodeAccessLogFilter(query))
}
func (s *clickhouseLogStore) IPTrend(ctx context.Context, query model.OpenFlareAccessLogQuery, bucketSeconds int64) ([]analyticsmodel.NodeAccessLogIPTrend, error) {
return analyticsrepo.IPTrendNodeAccessLogs(ctx, toNodeAccessLogFilter(query), bucketSeconds)
}
func (s *clickhouseLogStore) TrafficSummary(ctx context.Context, query model.OpenFlareAccessLogQuery) (model.OpenFlareAccessLogTrafficSummary, error) {
row, err := analyticsrepo.TrafficSummaryNodeAccessLogs(ctx, toNodeAccessLogFilter(query))
if err != nil {
return model.OpenFlareAccessLogTrafficSummary{}, err
}
return model.OpenFlareAccessLogTrafficSummary{
RequestCount: row.RequestCount,
ErrorCount: row.ErrorCount,
UniqueIPCount: row.UniqueIPCount,
BytesSent: row.BytesSent,
RequestLength: row.RequestLength,
NodeCount: row.NodeCount,
}, nil
}
func (s *clickhouseLogStore) ValueCounts(ctx context.Context, query model.OpenFlareAccessLogQuery, column string, limit int) ([]model.OpenFlareAccessLogValueCount, error) {
rows, err := analyticsrepo.ValueCountsNodeAccessLogs(ctx, toNodeAccessLogFilter(query), column, limit)
if err != nil {
return nil, err
}
out := make([]model.OpenFlareAccessLogValueCount, len(rows))
for i, r := range rows {
out[i] = model.OpenFlareAccessLogValueCount{Value: r.Value, Count: r.Count}
}
return out, nil
}
func (s *clickhouseLogStore) NodeAggregates(ctx context.Context, query model.OpenFlareAccessLogQuery) ([]model.OpenFlareAccessLogNodeAggregate, error) {
rows, err := analyticsrepo.NodeAggregatesNodeAccessLogs(ctx, toNodeAccessLogFilter(query))
if err != nil {
return nil, err
}
out := make([]model.OpenFlareAccessLogNodeAggregate, len(rows))
for i, r := range rows {
out[i] = model.OpenFlareAccessLogNodeAggregate{NodeID: r.NodeID, RequestCount: r.RequestCount, ErrorCount: r.ErrorCount, UniqueIPCount: r.UniqueIPCount}
}
return out, nil
}
func (s *clickhouseLogStore) DeleteAll(ctx context.Context) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteAllNodeAccessLogs(ctx)
}
func (s *clickhouseLogStore) DeleteBefore(ctx context.Context, cutoff time.Time) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteNodeAccessLogsBefore(ctx, cutoff)
}
func (s *clickhouseLogStore) DeleteByNodeBefore(ctx context.Context, nodeID string, before time.Time) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteNodeAccessLogsByNodeBefore(ctx, nodeID, before)
}
// ListForMigration 按 id 升序分页读取(迁移复制用):直接查询 CH 原生表。
func (s *clickhouseLogStore) ListForMigration(ctx context.Context, afterID uint64, limit int) ([]analyticsmodel.NodeAccessLog, error) {
if err := chConnErr(); err != nil {
return nil, err
}
rows, err := db.ChConn.Query(ctx, `
SELECT `+analyticsmodel.NodeAccessLog{}.InsertColumns()+`
FROM `+analyticsmodel.NodeAccessLog{}.TableName()+`
WHERE id > ?
ORDER BY id ASC
LIMIT ?`, afterID, limitOr(limit, migrationPageSize))
if err != nil {
return nil, fmt.Errorf("list node access logs for migration: %w", err)
}
defer func() { _ = rows.Close() }()
var result []analyticsmodel.NodeAccessLog
for rows.Next() {
var item analyticsmodel.NodeAccessLog
if err := rows.Scan(
&item.ID,
&item.NodeID,
&item.LoggedAt,
&item.RemoteAddr,
&item.Region,
&item.Host,
&item.Path,
&item.UserAgent,
&item.CacheStatus,
&item.StatusCode,
&item.BytesSent,
&item.RequestLength,
&item.RequestTimeMs,
&item.CreatedAt,
); err != nil {
return nil, fmt.Errorf("scan node access log row: %w", err)
}
item.LoggedAt = item.LoggedAt.UTC()
item.CreatedAt = item.CreatedAt.UTC()
result = append(result, item)
}
return result, nil
}
// ---- ObservabilityStore ----
// InsertMetricSnapshot 写入入口:冻结检查 + 经 hook 入队(异步),不直接落库。
func (s *clickhouseLogStore) InsertMetricSnapshot(ctx context.Context, record *model.OpenFlareMetricSnapshot) error {
if record == nil {
return nil
}
if err := s.ensureWritable(ctx); err != nil {
return err
}
if h := currentObservabilityHooks().QueueMetricSnapshot; h != nil {
h(toAnalyticsNodeMetricSnapshot(record))
}
return nil
}
func (s *clickhouseLogStore) ListMetricSnapshots(ctx context.Context, nodeID string, since time.Time, limit int) ([]*model.OpenFlareMetricSnapshot, error) {
rows, err := analyticsrepo.ListNodeMetricSnapshots(ctx, toNodeObservabilityFilter(nodeID, since, limit))
if err != nil {
return nil, err
}
return fromAnalyticsNodeMetricSnapshots(rows), nil
}
func (s *clickhouseLogStore) DeleteAllMetricSnapshots(ctx context.Context) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteAllNodeMetricSnapshots(ctx)
}
// ListTrafficHourly 委托 analyticsrepo 读 of_access_log_hourly rollup(M5 口径,UV 恒 0)。
func (s *clickhouseLogStore) ListTrafficHourly(ctx context.Context, nodeID string, since time.Time) ([]analyticsmodel.NodeTrafficHourly, error) {
return analyticsrepo.ListNodeTrafficHourly(ctx, toNodeObservabilitySince(nodeID, since))
}
// ListAccessLogHourly 委托 analyticsrepo 读 of_access_log_hourly rollup。
func (s *clickhouseLogStore) ListAccessLogHourly(ctx context.Context, nodeID string, since time.Time) ([]analyticsmodel.AccessLogHourly, error) {
return analyticsrepo.ListAccessLogHourly(ctx, toNodeObservabilitySince(nodeID, since))
}
// ListMetricHourly 委托 analyticsrepo ListNodeMetricHourly:rollup 覆盖窗口时读
// of_node_metric_capacity_hourly,否则按 mergeNodeMetricHourlyPreferRollup 合并 raw 兜底。
func (s *clickhouseLogStore) ListMetricHourly(ctx context.Context, nodeID string, since time.Time) ([]analyticsmodel.NodeMetricHourly, error) {
return analyticsrepo.ListNodeMetricHourly(ctx, toNodeObservabilitySince(nodeID, since))
}
func (s *clickhouseLogStore) DeleteMetricSnapshotsBefore(ctx context.Context, cutoff time.Time) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteNodeMetricSnapshotsBefore(ctx, cutoff)
}
// BatchInsertNodeMetricSnapshots 是 batchwriter flush 目标:CH 原生批量写入。
func (s *clickhouseLogStore) BatchInsertNodeMetricSnapshots(ctx context.Context, rows []analyticsmodel.NodeMetricSnapshot) error {
if err := s.ensureWritable(ctx); err != nil {
return err
}
return analyticsrepo.BatchInsertNodeMetricSnapshots(ctx, rows)
}
// InsertEdgeHealth 写入入口:冻结检查 + 经 hook 入队(异步),不直接落库。
func (s *clickhouseLogStore) InsertEdgeHealth(ctx context.Context, record *model.OpenFlareEdgeHealth) error {
if record == nil {
return nil
}
if err := s.ensureWritable(ctx); err != nil {
return err
}
if h := currentObservabilityHooks().QueueEdgeHealth; h != nil {
h(toAnalyticsNodeEdgeHealth(record))
}
return nil
}
func (s *clickhouseLogStore) ListEdgeHealth(ctx context.Context, nodeID string, since time.Time, limit int) ([]*model.OpenFlareEdgeHealth, error) {
rows, err := analyticsrepo.ListNodeEdgeHealth(ctx, toNodeObservabilityFilter(nodeID, since, limit))
if err != nil {
return nil, err
}
return fromAnalyticsNodeEdgeHealths(rows), nil
}
func (s *clickhouseLogStore) DeleteAllEdgeHealth(ctx context.Context) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteAllNodeEdgeHealth(ctx)
}
func (s *clickhouseLogStore) DeleteEdgeHealthBefore(ctx context.Context, cutoff time.Time) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteNodeEdgeHealthBefore(ctx, cutoff)
}
// BatchInsertNodeEdgeHealth 是 batchwriter flush 目标:CH 原生批量写入。
func (s *clickhouseLogStore) BatchInsertNodeEdgeHealth(ctx context.Context, rows []analyticsmodel.NodeEdgeHealth) error {
if err := s.ensureWritable(ctx); err != nil {
return err
}
return analyticsrepo.BatchInsertNodeEdgeHealth(ctx, rows)
}
// InsertNodeObservationFrps 写入入口:冻结检查 + 经 hook 入队(异步),不直接落库。
func (s *clickhouseLogStore) InsertNodeObservationFrps(ctx context.Context, record *model.OpenFlareNodeObservationFrps) error {
if record == nil {
return nil
}
if err := s.ensureWritable(ctx); err != nil {
return err
}
if h := currentObservabilityHooks().QueueNodeObsFrps; h != nil {
h(toAnalyticsNodeObsFrps(record))
}
return nil
}
func (s *clickhouseLogStore) ListNodeObservationFrps(ctx context.Context, nodeID string, since time.Time, limit int) ([]*model.OpenFlareNodeObservationFrps, error) {
rows, err := analyticsrepo.ListNodeObsFrps(ctx, toNodeObservabilityFilter(nodeID, since, limit))
if err != nil {
return nil, err
}
return fromAnalyticsNodeObsFrps(rows), nil
}
func (s *clickhouseLogStore) DeleteAllNodeObservationFrps(ctx context.Context) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteAllNodeObsFrps(ctx)
}
func (s *clickhouseLogStore) DeleteNodeObservationFrpsBefore(ctx context.Context, cutoff time.Time) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteNodeObsFrpsBefore(ctx, cutoff)
}
// BatchInsertNodeObsFrps 是 batchwriter flush 目标:CH 原生批量写入。
func (s *clickhouseLogStore) BatchInsertNodeObsFrps(ctx context.Context, rows []analyticsmodel.NodeObsFrps) error {
if err := s.ensureWritable(ctx); err != nil {
return err
}
return analyticsrepo.BatchInsertNodeObsFrps(ctx, rows)
}
// InsertNodeObservationFrpc 写入入口:冻结检查 + 经 hook 入队(异步),不直接落库。
func (s *clickhouseLogStore) InsertNodeObservationFrpc(ctx context.Context, record *model.OpenFlareNodeObservationFrpc) error {
if record == nil {
return nil
}
if err := s.ensureWritable(ctx); err != nil {
return err
}
if h := currentObservabilityHooks().QueueNodeObsFrpc; h != nil {
h(toAnalyticsNodeObsFrpc(record))
}
return nil
}
func (s *clickhouseLogStore) ListNodeObservationFrpc(ctx context.Context, nodeID string, since time.Time, limit int) ([]*model.OpenFlareNodeObservationFrpc, error) {
rows, err := analyticsrepo.ListNodeObsFrpc(ctx, toNodeObservabilityFilter(nodeID, since, limit))
if err != nil {
return nil, err
}
return fromAnalyticsNodeObsFrpc(rows), nil
}
func (s *clickhouseLogStore) DeleteAllNodeObservationFrpc(ctx context.Context) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteAllNodeObsFrpc(ctx)
}
func (s *clickhouseLogStore) DeleteNodeObservationFrpcBefore(ctx context.Context, cutoff time.Time) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteNodeObsFrpcBefore(ctx, cutoff)
}
// MigrationRange 返回 of_node_access_logs.logged_at 的最小/最大值(空表返回零值)。
func (s *clickhouseLogStore) MigrationRange(ctx context.Context) (time.Time, time.Time, error) {
return chMigrationRange(ctx, analyticsmodel.NodeAccessLog{}.TableName(), "logged_at")
}
// EnsurePartitions 是 CH 分支 no-op(CH 无 PG 式分区)。
func (s *clickhouseLogStore) EnsurePartitions(_ context.Context, _, _ time.Time) error {
return nil
}
// chMigrationRange 查询 CH 表时间列 MIN/MAX;空表(NULL)返回零值。
func chMigrationRange(ctx context.Context, table, column string) (time.Time, time.Time, error) {
if err := chConnErr(); err != nil {
return time.Time{}, time.Time{}, err
}
var minTime, maxTime *time.Time
if err := db.ChConn.QueryRow(ctx,
"SELECT min("+column+"), max("+column+") FROM "+table,
).Scan(&minTime, &maxTime); err != nil {
return time.Time{}, time.Time{}, fmt.Errorf("query migration range %s: %w", table, err)
}
if minTime == nil || maxTime == nil {
return time.Time{}, time.Time{}, nil
}
return minTime.UTC(), maxTime.UTC(), nil
}
// BatchInsertNodeObsFrpc 是 batchwriter flush 目标:CH 原生批量写入。
func (s *clickhouseLogStore) BatchInsertNodeObsFrpc(ctx context.Context, rows []analyticsmodel.NodeObsFrpc) error {
if err := s.ensureWritable(ctx); err != nil {
return err
}
return analyticsrepo.BatchInsertNodeObsFrpc(ctx, rows)
}
// ListMetricSnapshotsForMigration 按 id 升序分页读取(迁移复制用)。
func (s *clickhouseLogStore) ListMetricSnapshotsForMigration(ctx context.Context, afterID uint64, limit int) ([]analyticsmodel.NodeMetricSnapshot, error) {
return chListForMigration(ctx, afterID, limit,
analyticsmodel.NodeMetricSnapshot{}.TableName(),
analyticsmodel.NodeMetricSnapshot{}.InsertColumns(),
func(rows driver.Rows) ([]analyticsmodel.NodeMetricSnapshot, error) {
var result []analyticsmodel.NodeMetricSnapshot
for rows.Next() {
var item analyticsmodel.NodeMetricSnapshot
if err := rows.Scan(
&item.ID,
&item.NodeID,
&item.CapturedAt,
&item.CPUUsagePercent,
&item.MemoryUsedBytes,
&item.MemoryTotalBytes,
&item.StorageUsedBytes,
&item.StorageTotalBytes,
&item.DiskReadBytes,
&item.DiskWriteBytes,
&item.NetworkRxBytes,
&item.NetworkTxBytes,
&item.CreatedAt,
); err != nil {
return nil, fmt.Errorf("scan node metric snapshot row: %w", err)
}
item.CapturedAt = item.CapturedAt.UTC()
item.CreatedAt = item.CreatedAt.UTC()
result = append(result, item)
}
return result, nil
})
}
// chObsRow 迁移读取共用的双字段观测行(字符串状态 + 数值计数):
// edge_health(status/connections)与 obs_frpc(tunnel_status/connected_relays_count)同形状。
type chObsRow struct {
ID uint64
NodeID string
CapturedAt time.Time
Status string
Count int64
CreatedAt time.Time
}
// countToInt32 将观测计数转为 int32(防御溢出;观测计数远小于 int32 上限)。
func countToInt32(v int64) int32 {
if v > math.MaxInt32 {
return math.MaxInt32
}
if v < math.MinInt32 {
return math.MinInt32
}
return int32(v)
}
// scanChObsRow 扫描 chObsRow(含 UTC 归一化)。
func scanChObsRow(rows driver.Rows) ([]chObsRow, error) {
var result []chObsRow
for rows.Next() {
var item chObsRow
if err := rows.Scan(
&item.ID,
&item.NodeID,
&item.CapturedAt,
&item.Status,
&item.Count,
&item.CreatedAt,
); err != nil {
return nil, fmt.Errorf("scan observation row: %w", err)
}
item.CapturedAt = item.CapturedAt.UTC()
item.CreatedAt = item.CreatedAt.UTC()
result = append(result, item)
}
return result, nil
}
// ListEdgeHealthForMigration 按 id 升序分页读取(迁移复制用)。
func (s *clickhouseLogStore) ListEdgeHealthForMigration(ctx context.Context, afterID uint64, limit int) ([]analyticsmodel.NodeEdgeHealth, error) {
rows, err := chListForMigration(ctx, afterID, limit,
analyticsmodel.NodeEdgeHealth{}.TableName(),
analyticsmodel.NodeEdgeHealth{}.InsertColumns(),
scanChObsRow)
if err != nil {
return nil, err
}
out := make([]analyticsmodel.NodeEdgeHealth, len(rows))
for i, r := range rows {
out[i] = analyticsmodel.NodeEdgeHealth{ID: r.ID, NodeID: r.NodeID, CapturedAt: r.CapturedAt, Status: r.Status, Connections: r.Count, CreatedAt: r.CreatedAt}
}
return out, nil
}
// ListNodeObsFrpsForMigration 按 id 升序分页读取(迁移复制用)。
func (s *clickhouseLogStore) ListNodeObsFrpsForMigration(ctx context.Context, afterID uint64, limit int) ([]analyticsmodel.NodeObsFrps, error) {
return chListForMigration(ctx, afterID, limit,
analyticsmodel.NodeObsFrps{}.TableName(),
analyticsmodel.NodeObsFrps{}.InsertColumns(),
func(rows driver.Rows) ([]analyticsmodel.NodeObsFrps, error) {
var result []analyticsmodel.NodeObsFrps
for rows.Next() {
var item analyticsmodel.NodeObsFrps
if err := rows.Scan(
&item.ID,
&item.NodeID,
&item.CapturedAt,
&item.FrpsConnections,
&item.FrpsProxyCount,
&item.FrpsClientCount,
&item.FrpsProxies,
&item.CreatedAt,
); err != nil {
return nil, fmt.Errorf("scan node frps observation row: %w", err)
}
item.CapturedAt = item.CapturedAt.UTC()
item.CreatedAt = item.CreatedAt.UTC()
result = append(result, item)
}
return result, nil
})
}
// ListNodeObsFrpcForMigration 按 id 升序分页读取(迁移复制用)。
func (s *clickhouseLogStore) ListNodeObsFrpcForMigration(ctx context.Context, afterID uint64, limit int) ([]analyticsmodel.NodeObsFrpc, error) {
rows, err := chListForMigration(ctx, afterID, limit,
analyticsmodel.NodeObsFrpc{}.TableName(),
analyticsmodel.NodeObsFrpc{}.InsertColumns(),
scanChObsRow)
if err != nil {
return nil, err
}
out := make([]analyticsmodel.NodeObsFrpc, len(rows))
for i, r := range rows {
out[i] = analyticsmodel.NodeObsFrpc{ID: r.ID, NodeID: r.NodeID, CapturedAt: r.CapturedAt, TunnelStatus: r.Status, ConnectedRelaysCount: countToInt32(r.Count), CreatedAt: r.CreatedAt}
}
return out, nil
}
// chListForMigration 执行按 id 升序分页的 CH 原生表查询,并交给 scanner 扫描。
func chListForMigration[T any](ctx context.Context, afterID uint64, limit int, table, columns string, scanner func(driver.Rows) ([]T, error)) ([]T, error) {
if err := chConnErr(); err != nil {
return nil, err
}
rows, err := db.ChConn.Query(ctx, `
SELECT `+columns+`
FROM `+table+`
WHERE id > ?
ORDER BY id ASC
LIMIT ?`, afterID, limitOr(limit, migrationPageSize))
if err != nil {
return nil, fmt.Errorf("list %s for migration: %w", table, err)
}
defer func() { _ = rows.Close() }()
return scanner(rows)
}
// ---- StatusStore ----
// ActiveDatabase 返回当前日志主库名(CH 分支固定 clickhouse)。
func (s *clickhouseLogStore) ActiveDatabase(_ context.Context) (string, error) {
return dbNameClickHouse, nil
}
// ClickHouseOperationalStats 委托 analyticsrepo 汇总 CH 运行状态。
func (s *clickhouseLogStore) ClickHouseOperationalStats(ctx context.Context) (*analyticsmodel.ClickHouseOperationalStats, error) {
return analyticsrepo.GetClickHouseOperationalStats(ctx)
}
// ---- UserAccessLogStore ----
// clickhouseUserAccessLogStore 实现 UserAccessLogStore。clickhouseLogStore 已占用
// List/Count 方法名(AccessLogStore 接口),Go 不允许同名不同签名方法,故用户访问日志
// 用独立类型嵌入同一 clickhouseLogStore(与 userAccessLogGormStore 同构),复用 ensureWritable。
type clickhouseUserAccessLogStore struct {
*clickhouseLogStore
}
func newClickHouseUserAccessLogStore() *clickhouseUserAccessLogStore {
return &clickhouseUserAccessLogStore{clickhouseLogStore: newClickHouseStore()}
}
// BatchInsert 是 batchwriter flush 目标:CH 原生批量写入;冻结期拒绝写入,空批次直接返回。
func (s *clickhouseUserAccessLogStore) BatchInsert(ctx context.Context, logs []analyticsmodel.UserAccessLog) error {
if len(logs) == 0 {
return nil
}
if err := s.ensureWritable(ctx); err != nil {
return err
}
return analyticsrepo.BatchInsert(ctx, logs)
}
// DeleteAll 清空全部用户访问日志(TRUNCATE 语义,迁移「覆盖目标库已有日志」幂等前提用)。
func (s *clickhouseUserAccessLogStore) DeleteAll(ctx context.Context) (int64, error) {
if err := s.ensureWritable(ctx); err != nil {
return 0, err
}
return analyticsrepo.DeleteAllUserAccessLogs(ctx)
}
// ListForMigration 按 id 升序分页读取(迁移复制用)。
func (s *clickhouseUserAccessLogStore) ListForMigration(ctx context.Context, afterID uint64, limit int) ([]analyticsmodel.UserAccessLog, error) {
return chListForMigration(ctx, afterID, limit,
analyticsmodel.UserAccessLog{}.TableName(),
analyticsmodel.UserAccessLog{}.InsertColumns(),
func(rows driver.Rows) ([]analyticsmodel.UserAccessLog, error) {
var result []analyticsmodel.UserAccessLog
for rows.Next() {
var item analyticsmodel.UserAccessLog
if err := rows.Scan(
&item.ID,
&item.UserID,
&item.Path,
&item.Method,
&item.IP,
&item.UserAgent,
&item.Headers,
&item.Status,
&item.Latency,
&item.CreatedAt,
); err != nil {
return nil, fmt.Errorf("scan user access log row: %w", err)
}
item.CreatedAt = item.CreatedAt.UTC()
result = append(result, item)
}
return result, nil
})
}
// MigrationRange 返回 w_user_access_logs.created_at 的最小/最大值(空表返回零值)。
func (s *clickhouseUserAccessLogStore) MigrationRange(ctx context.Context) (time.Time, time.Time, error) {
return chMigrationRange(ctx, analyticsmodel.UserAccessLog{}.TableName(), "created_at")
}
func (s *clickhouseUserAccessLogStore) Count(ctx context.Context, filter analyticsmodel.AccessLogFilter) (uint64, error) {
return analyticsrepo.CountAccessLogs(ctx, filter)
}
func (s *clickhouseUserAccessLogStore) List(ctx context.Context, filter analyticsmodel.AccessLogFilter, page, pageSize int) ([]analyticsmodel.UserAccessLog, uint64, error) {
return analyticsrepo.ListAccessLogs(ctx, filter, page, pageSize)
}
func (s *clickhouseUserAccessLogStore) GetDailyTrend(ctx context.Context, days int) ([]analyticsmodel.DailyTrend, error) {
return analyticsrepo.GetDailyTrend(ctx, days)
}
func (s *clickhouseUserAccessLogStore) GetBrowserDistribution(ctx context.Context, startTime time.Time) ([]analyticsmodel.BrowserShare, error) {
return analyticsrepo.GetBrowserDistribution(ctx, startTime)
}
func (s *clickhouseUserAccessLogStore) GetTopActiveUsers(ctx context.Context, startTime time.Time, limit int) ([]analyticsmodel.TopUser, error) {
return analyticsrepo.GetTopActiveUsers(ctx, startTime, limit)
}
// toNodeObservabilityFilter 构造 CH 可观测查询过滤器(limit<=0 表示不限制)。
func toNodeObservabilityFilter(nodeID string, since time.Time, limit int) analyticsmodel.NodeObservabilityFilter {
return analyticsmodel.NodeObservabilityFilter{
NodeID: nodeID,
Since: since,
Limit: limit,
}
}
// toNodeObservabilitySince 构造不带 limit 的可观测查询过滤器
// (小时级聚合读无需分页,避免传无意义的 0)。
func toNodeObservabilitySince(nodeID string, since time.Time) analyticsmodel.NodeObservabilityFilter {
return analyticsmodel.NodeObservabilityFilter{NodeID: nodeID, Since: since}
}