Skip to content

Docs: no production operations / observability guide (signals, alerting, incident runbook) #236

Description

@Arpit-oo

Problem Statement

Several guides cover choosing primitives and building an integration, but none cover RUNNING one in production. HydraDB returns rich operational signals - meta.request_id, meta.latency_ms, meta.deprecation, the async ingestion status pipeline, database readiness flags - but there is no guidance on what to emit, what to alert on, how to trace a request, or what to do during an incident.

Proposed Solution

Add essentials/v2/production-operations: the four signal sources; golden signals to emit; end-to-end request tracing via request_id; monitoring the async pipeline (transient error_code vs terminal errored); readiness gates via /databases/status; deprecation-drift detection; a symptom-to-action incident runbook; and building SLOs from measured latency. Grounded strictly in documented response fields - no invented latency numbers.

Alternatives Considered

  • A generic 'best practices' page: too vague; operators need signal-to-action mapping.
  • Folding into a query-tuning page: tuning is choosing parameters; this is watching them in production.

Additional Context

Submitted for the Mintlify x HydraDB documentation hackathon.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions