Skip to content

EventQueue.getStats() misclassifies retrying failures as "processing" #205

Description

@N-thnI

Description

Give retrying/failing events their own stats bucket instead of counting them as healthy in-flight work.

Problem Statement

engine-bridge/src/event-queue.ts lines 268-276: events in status='failed' with attempts < maxRetries are counted into stats.processing rather than a distinct retry bucket. During an outage where handlers are failing and retrying, operators watching the dashboard's "processing" count won't see the growing failure/retry backlog — it looks like healthy in-flight work.

Proposed Changes

  • Add a distinct retrying (or similar) bucket to getStats() output and stop folding failed-but-retrying events into processing
  • Update any dashboard/consumer of these stats to surface the new bucket

Technical Implementation Scaffolding

  • Target Repository: vero-core-engine
  • Target Path: engine-bridge/src/event-queue.ts
  • Branch Naming: fix/issue--event-queue-stats-retry-bucket
  • Authority Context: Observability

Acceptance Criteria

  • getStats() reports retrying events separately from actively-processing ones
  • Covered by a test with a mix of processing/failed-retrying/failed-exhausted events

Definition of Done

  • Reviewed by lead maintainer
  • Pull request merged via verified status check

Metadata

Metadata

Assignees

Labels

GrantFox OSSIssue tracked in GrantFox OSSMaybe RewardedIssue may be eligible for a GrantFox rewardThird CampaignCampaign: Third Campaign

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions