The GitHub Actions job "CI" on rocketmq-dashboard.git/codex/audit-retention has failed. Run started by GitHub user tju-yxq (triggered by tju-yxq).
Head commit for run: 9a5c5e50f6345881badb3a586b35a1fc0b553e79 / Yxq <[email protected]> feat(audit): add an opt-in scheduled retention sweep The audit table is the one high-volume studio table with no automatic retention: sessions, metric snapshots, query history, AI conversations and the notification outbox all have scheduled cleanups, while audit rows accumulate forever unless an administrator remembers to open the audit page and run the manual cleanup — which also has to be repeated after every incident. Mirror the established retention pattern (studio.query-history) with studio.audit.retention-days and an hourly sweep that reuses the same bounded delete the manual endpoint uses (500-row batches, at most 20 per pass, so a large backlog drains over successive sweeps without a long-running statement). The default is deliberately 0 — keep forever — unlike the diagnostic tables that default to 90 days: audit rows are compliance evidence, and an upgrade must never start silently deleting them. Deployments opt in via STUDIO_AUDIT_RETENTION_DAYS. Values above the 365-day maximum the manual endpoint enforces are clamped with a warning instead of failing the sweep every hour, and repository failures are logged and swallowed: retention is hygiene, not an application error. Report URL: https://github.com/apache/rocketmq-dashboard/actions/runs/36900072346 With regards, GitHub Actions via GitBox
