audit schema¶
The audit schema contains append-heavy logging tables that are operationally important for traceability and abuse review.
Inventory¶
actions_logs: general event log with author, IP, context JSON, and session fieldsposts: request payload capture with route, errors, user agent, and timestamps
Live footprint¶
actions_logs: roughly871kestimated rows and213 MBposts: roughly178kestimated rows and126 MB
Relation to stats¶
These tables do not directly feed stats, but they matter for:
- backfill incident analysis
- abuse or bot investigations affecting traffic-driven metrics
- forensic review when asynchronous scoring or maintenance jobs appear inconsistent with application activity
Operational notes¶
- this schema is log-like and retention should be treated explicitly
- API exposure should be internal-only, if exposed at all
- JSON payload columns are useful for operator tooling but should not be part of public analytics contracts
Operator check¶
- define and review retention windows for
actions_logsandposts - monitor table growth because both tables are already material in size