Api::V1::ReferencesController#create (avg 12080ms, max 12080ms)

I'll run the rca skill on this cluster.

I'll proceed with the RCA. This is a latency cluster (not an error) — `Api::V1::ReferencesController

Let me start by checking status board and exploring the code.

I'll check the status board first.

Status board: `svc:*` scope, no active dep:* incident. Proceed with normal RCA. Two recent resolved

This is a routine Rails controller, not malware. Let me explore ReferenceFactory and search Datadog

Reference factory creates a Reference and then optionally calls `add_reference_sources`. Let me sear

N+1 pattern visible: per-source uniqueness check + create iteratively. Let me search by resource and

Now let me check span/duration metric and look at API logs for the slow trace's specific timeframe (

Empty — try different metric name.

Let me also check `EntityIndexable`/Eventable for hooks that could be slow, and base factory for the

Found a key clue: `_entity_index_document` is called on `after_commit :create` and synchronously cal

No entity index errors — Elasticsearch indexing succeeded but synchronously, contributing to latency

RCA report written. As instructed, I did NOT modify the cluster file frontmatter — the pipeline mana

I'll call the StructuredOutput tool now.

RCA complete. Report written to `content/docs/incidents/44735ce7-36a7-426f-b871-1368b59c8f00/rca.mdx