Operations
Monitor active background jobs and replay dead letters.
Review active background operations, find failed jobs in the dead letter queue, and replay them with a reason.
Quick start
- Open
/settings/operationsfrom a recovery link, support instruction, or direct URL. Church admins only. This advanced page is not in the current Settings navigation. - Review Active operations — the list auto-refreshes every 30 seconds.
- Scroll to Dead-letter recovery for failed jobs.
- Enter a Required replay reason (at least 8 characters).
- Click Reauthorize and replay.
You should see the job leave the dead-letter list after a successful replay, or appear again under Active operations.
Deep dive
What this page is for
Operations monitors church-wide durable background jobs — internal processing and AI work that runs outside the UI you interact with daily.
This is different from Sync operations, which is integration-specific (Planning Center pulls, calendar sync, conflict resolution). Use sync operations for provider sync health; use this page for global job recovery.
| Page | Scope |
|---|---|
| Operations (this page) | All background jobs for the church |
| Sync operations | Integration sync definitions and runs only |
| Operation support | Look up a single operation by ID |
Active operations
The Active operations section lists in-flight jobs with:
- Job kind — internal or AI work type label
- Stage — Queued, Running, Waiting, Needs input, Retry scheduled, Cancel requested, etc.
- Operation ID — unique identifier for support and ledger cross-reference
The server reconciles state every 30 seconds. Refresh the page for an immediate update.
When no jobs are running, you see No active operations.
Dead-letter recovery
Jobs that fail terminally land in Dead-letter recovery. Each entry shows:
- Job kind
- Operation ID
- Attempt count
Before replaying:
- Understand why the job failed (check Operation support with the operation ID)
- Fix the root cause when possible (reauth, quota, missing data)
- Enter a Required replay reason — minimum 8 characters for audit trail
- Read the cost warning: Provider work or cost may already have occurred. Replay can perform additional billable work.
- Click Reauthorize and replay
Replay re-queues the job through the durable replay route. It is not free — repeated replays can incur more usage. Check Usage & limits after large replays.
Operation IDs elsewhere
Operation IDs appear in:
- Error toasts and alerts in the app
- Usage ledger source operation column
- Support emails from WorshipDeck
Use Operation support to inspect timeline, usage attribution, and safe recovery links for a specific ID. Use Support tickets when you need help from the WorshipDeck team.
Relationship to integrations
Integration sync failures may appear here and on Sync operations. Start with sync operations for provider-specific context (conflicts, checkpoints, Run now). Use this page when support directs you to replay a dead-lettered church job.
Tips
- Copy the Operation ID before contacting WorshipDeck support.
- Wait for auto-refresh instead of hammering replay on the same job.
- Fix quota or auth issues before replaying — otherwise the job dead-letters again.
- Write meaningful replay reasons (Reauth completed after PCO token expiry) for your audit log.
- Monitor Usage & limits after replaying AI or connector jobs.
[!warning] Replay can trigger billable provider or AI work. Confirm the underlying issue is resolved before clicking Reauthorize and replay.
Troubleshooting
Operations could not be loaded
- Refresh the page.
- Confirm you are an active church Admin.
Replay button does nothing / error about reason
- Enter at least 8 characters in Required replay reason.
Job replays but fails again
- Look up the operation ID in Operation support.
- Check Usage & limits for exhausted meters.
- For integration sync issues, review Sync operations.
Active operation stuck on Needs input
- Some jobs wait for human review (for example, cloud file import review).
- Complete the required step in the feature UI, then return here.
Cannot find a job from an error message
- Use Operation support with the exact operation ID.
- The job may have already completed — check the usage ledger for final status.