fix(publish): prevent stuck publish queue and duplicate posting
Desktop Client Build / Resolve Build Metadata (push) Successful in 41s
Backend CI / Backend (push) Failing after 7m14s
Desktop Client Build / Build Desktop Client (push) Successful in 23m46s
Desktop Client Build / Publish Client Artifacts to NAS (push) Successful in 56s
Desktop Client Build / Resolve Build Metadata (push) Successful in 41s
Backend CI / Backend (push) Failing after 7m14s
Desktop Client Build / Build Desktop Client (push) Successful in 23m46s
Desktop Client Build / Publish Client Artifacts to NAS (push) Successful in 56s
The desktop publish queue could stall for a long time and end users assumed the software was broken. Root cause: a hung adapter held its execution slot with no wall-clock timeout while auto-renewing its lease forever, so the server never reclaimed it and every queued task behind it stayed 等待发布. Auto-recovery also risked silently re-posting a non-idempotent article. Client (Electron): - per-task wall-clock deadline + abort; progress-gated lease renewal that stops and aborts a stalled task instead of renewing it forever - decouple the concurrency cap from CDP-induced CPU/memory pressure (admission gate instead of self-throttling collapse); 15s watchdog pump - all adapter network I/O now has fetch timeouts and honors context.signal; bounded image-upload concurrency with per-image timeout - surface live adapter progress, elapsed time, queue position and a working-vs-queued distinction in the publish view Server (tenant-api): - publish lease-recovery worker (every 3m) + supporting index: reclaim expired in_progress publish leases, requeue (<3 attempts) or terminal-fail - 3-minute lease TTL with client-presence-gated extension; max 3 attempts Idempotency (production-grade core): - durable publish_submit_started_at marker, set before the irreversible platform submit POST; recovery and abort route a maybe-submitted task to unknown (manual reconcile, kept in the dedup set) instead of re-posting - desktop UI requires explicit confirmation before retrying a possibly- already-published task Verified: go build/vet/test (incl. resolvePublishRecoveryOutcome), vue-tsc, vitest 141/141, gofmt all clean. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
@@ -0,0 +1,114 @@
|
||||
package scheduler
|
||||
|
||||
import (
|
||||
"context"
|
||||
"time"
|
||||
|
||||
"go.uber.org/zap"
|
||||
|
||||
tenantapp "github.com/geo-platform/tenant-api/internal/tenant/app"
|
||||
)
|
||||
|
||||
const (
|
||||
defaultPublishLeaseRecoveryInterval = 3 * time.Minute
|
||||
defaultPublishLeaseRecoveryTimeout = 30 * time.Second
|
||||
defaultPublishLeaseRecoveryLimit = 1000
|
||||
)
|
||||
|
||||
type PublishLeaseRecoveryWorker struct {
|
||||
service *tenantapp.DesktopTaskService
|
||||
logger *zap.Logger
|
||||
interval time.Duration
|
||||
timeout time.Duration
|
||||
limit int
|
||||
}
|
||||
|
||||
func NewPublishLeaseRecoveryWorker(service *tenantapp.DesktopTaskService, logger *zap.Logger) *PublishLeaseRecoveryWorker {
|
||||
return &PublishLeaseRecoveryWorker{
|
||||
service: service,
|
||||
logger: logger,
|
||||
interval: defaultPublishLeaseRecoveryInterval,
|
||||
timeout: defaultPublishLeaseRecoveryTimeout,
|
||||
limit: defaultPublishLeaseRecoveryLimit,
|
||||
}
|
||||
}
|
||||
|
||||
func (w *PublishLeaseRecoveryWorker) Start(ctx context.Context) {
|
||||
go w.Run(ctx)
|
||||
}
|
||||
|
||||
func (w *PublishLeaseRecoveryWorker) Run(ctx context.Context) {
|
||||
if w == nil || w.service == nil {
|
||||
return
|
||||
}
|
||||
w.run(ctx)
|
||||
}
|
||||
|
||||
func (w *PublishLeaseRecoveryWorker) run(ctx context.Context) {
|
||||
w.runOnce(context.Background())
|
||||
|
||||
ticker := time.NewTicker(w.interval)
|
||||
defer ticker.Stop()
|
||||
|
||||
for {
|
||||
select {
|
||||
case <-ctx.Done():
|
||||
return
|
||||
case <-ticker.C:
|
||||
w.runOnce(context.Background())
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func (w *PublishLeaseRecoveryWorker) runOnce(parent context.Context) {
|
||||
ctx, cancel := context.WithTimeout(parent, w.timeout)
|
||||
defer cancel()
|
||||
|
||||
result, err := w.service.RecoverExpiredPublishTasks(ctx, w.limit)
|
||||
if err != nil {
|
||||
if w.logger != nil {
|
||||
w.logger.Warn("publish lease recovery failed", zap.Error(err))
|
||||
}
|
||||
return
|
||||
}
|
||||
|
||||
if (result.Requeued > 0 || result.Failed > 0 || result.Uncertain > 0) && w.logger != nil {
|
||||
w.logger.Info("publish lease recovery completed",
|
||||
zap.Int("requeued_task_count", result.Requeued),
|
||||
zap.Int("failed_task_count", result.Failed),
|
||||
zap.Int("uncertain_task_count", result.Uncertain),
|
||||
)
|
||||
}
|
||||
}
|
||||
|
||||
func (w *PublishLeaseRecoveryWorker) RunOnce(parent context.Context, run JobRunContext) (map[string]any, error) {
|
||||
if w == nil || w.service == nil {
|
||||
return map[string]any{"skipped": true, "reason": "worker_not_ready"}, nil
|
||||
}
|
||||
startedAt := time.Now()
|
||||
timeout := w.timeout
|
||||
if run.Job != nil && run.Job.TimeoutSeconds > 0 {
|
||||
timeout = time.Duration(run.Job.TimeoutSeconds) * time.Second
|
||||
}
|
||||
limit := w.limit
|
||||
if run.Job != nil && run.Job.BatchSize != nil && *run.Job.BatchSize > 0 {
|
||||
limit = *run.Job.BatchSize
|
||||
}
|
||||
if value := AsInt(run.Config, "batch_size", 0); value > 0 && (run.Job == nil || run.Job.BatchSize == nil) {
|
||||
limit = value
|
||||
}
|
||||
ctx, cancel := context.WithTimeout(parent, timeout)
|
||||
defer cancel()
|
||||
|
||||
result, err := w.service.RecoverExpiredPublishTasks(ctx, limit)
|
||||
if err != nil {
|
||||
return map[string]any{"stage": "recover"}, err
|
||||
}
|
||||
return map[string]any{
|
||||
"requeued_task_count": result.Requeued,
|
||||
"failed_task_count": result.Failed,
|
||||
"uncertain_task_count": result.Uncertain,
|
||||
"duration_ms": time.Since(startedAt).Milliseconds(),
|
||||
"batch_size": limit,
|
||||
}, nil
|
||||
}
|
||||
Reference in New Issue
Block a user