BCAI/easyai-ai-gateway
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

feat(acceptance): 建立同构验收与弹性容量体系

Browse Source 
实现本地三节点 K3s 同构环境、脱敏生产快照、Gemini 图片和多参考图视频协议模拟、统一验收报告及故障注入。\n\n新增 Worker 容量控制器、资源与连接预算、任务恢复保护，并将生产验收拆分为 validation 执行和人工 CAS 放量。\n\n验证包括 Go 全量测试、PostgreSQL HTTP 集成测试、go vet、OpenAPI、ShellCheck、前端检查、迁移及发布脚本测试。
This commit is contained in:
wangbo
2026-07-31 18:02:24 +08:00
parent 93d36d0e55
commit e05922b0f4
94 changed files with 12379 additions and 826 deletions
Download Patch File Download Diff File Expand all files Collapse all files
apps/api/internal/runner/queue_worker.go
+8 -8
View File
@@ -191,7 +191,7 @@ func retryAsyncQueueStartup(
}
func (s *Service) startRiverQueue(ctx context.Context, workerEnabled bool) error {
driver := riverpgxv5.New(s.store.Pool())
driver := riverpgxv5.New(s.riverStore.Pool())
migrator, err := rivermigrate.New(driver, nil)
if err != nil {
return err
@@ -286,7 +286,7 @@ func (s *Service) newRiverAsyncExecutionClient(capacity int) (*river.Client[pgx.
if err := river.AddWorkerSafely(workers, &asyncTaskWorker{service: s}); err != nil {
return nil, err
}
return river.NewClient(riverpgxv5.New(s.store.Pool()), &river.Config{
return river.NewClient(riverpgxv5.New(s.riverStore.Pool()), &river.Config{
ID: fmt.Sprintf("%s-exec-%d-%d", s.workerInstanceID, capacity, time.Now().UnixNano()),
JobTimeout: -1,
Logger: s.logger,
@@ -314,11 +314,11 @@ func (s *Service) loadAsyncWorkerCapacity(ctx context.Context) (store.AsyncWorke
if s.asyncCapacityLoader != nil {
return s.asyncCapacityLoader(ctx, s.cfg.AsyncWorkerHardLimit)
}
snapshot, err := s.store.AsyncWorkerCapacity(ctx, s.cfg.AsyncWorkerHardLimit)
snapshot, err := s.coordinationStore.AsyncWorkerCapacity(ctx, s.cfg.AsyncWorkerHardLimit)
if err != nil {
return store.AsyncWorkerCapacitySnapshot{}, err
}
allocation, err := s.store.RegisterWorkerInstance(ctx, store.WorkerRegistrationInput{
allocation, err := s.coordinationStore.RegisterWorkerInstance(ctx, store.WorkerRegistrationInput{
InstanceID: s.workerInstanceID,
PodUID: strings.TrimSpace(os.Getenv("POD_UID")),
PodName: strings.TrimSpace(os.Getenv("POD_NAME")),
@@ -437,7 +437,7 @@ func (s *Service) stopAsyncWorkersOnShutdown(ctx context.Context) {
<-ctx.Done()
if s.store != nil && strings.TrimSpace(s.workerInstanceID) != "" {
drainCtx, drainCancel := context.WithTimeout(context.Background(), 5*time.Second)
if err := s.store.MarkWorkerDraining(drainCtx, s.workerInstanceID); err != nil && !errors.Is(err, pgx.ErrNoRows) {
if err := s.coordinationStore.MarkWorkerDraining(drainCtx, s.workerInstanceID); err != nil && !errors.Is(err, pgx.ErrNoRows) {
s.logger.Warn("mark async worker draining failed", "error", err)
}
drainCancel()
@@ -567,7 +567,7 @@ func (s *Service) recoverAsyncRiverJobs(ctx context.Context) error {
func (s *Service) recoverOrphanedAsyncRiverJobs(ctx context.Context) {
recoverJobs := func() {
runtimeRecovery, err := s.store.RecoverInterruptedRuntimeState(ctx)
runtimeRecovery, err := s.coordinationStore.RecoverInterruptedRuntimeState(ctx)
if err != nil {
if ctx.Err() == nil {
s.logger.Warn("recover interrupted runtime state from worker failed", "error", err)
@@ -589,7 +589,7 @@ func (s *Service) recoverOrphanedAsyncRiverJobs(ctx context.Context) {
"cleanedTaskAdmissions", runtimeRecovery.CleanedTaskAdmissions,
)
}
yieldedAdmissions, err := s.store.YieldStaleAsyncTaskAdmissions(
yieldedAdmissions, err := s.coordinationStore.YieldStaleAsyncTaskAdmissions(
ctx,
orphanedRiverJobWorkerStaleAfter,
orphanedRiverJobRecoveryBatchSize,
@@ -603,7 +603,7 @@ func (s *Service) recoverOrphanedAsyncRiverJobs(ctx context.Context) {
if yieldedAdmissions > 0 {
s.logger.Warn("stale async task admissions yielded", "count", yieldedAdmissions)
}
recovered, err := s.store.RecoverOrphanedAsyncRiverJobs(
recovered, err := s.coordinationStore.RecoverOrphanedAsyncRiverJobs(
ctx,
orphanedRiverJobWorkerStaleAfter,
orphanedRiverJobRecoveryBatchSize,