3 Commits

Author	SHA1	Message	Date
ryan	b928928958	fix(observability): restore 24h capacity/network/disk trends via CH hourly agg ... Node and dashboard 24h capacity, network, and disk IO charts only used the latest limited raw snapshots (120/500 rows), so historical hour buckets stayed empty. Prefer ClickHouse hourly aggregates with counter deltas, and fall back to raw snapshots when aggregation is unavailable.	2026-07-10 09:48:45 +08:00
ryan	58624db397	perf(clickhouse): Phase 2 legacy governance — TTL cleanup, unified pool, MV, ops API ... - Replace retention ALTER DELETE with MATERIALIZE TTL; use TRUNCATE for delete-all - Remove GORM ClickHouse pool; migrate user access log reads to ChConn - Drop query-side trim(remote_addr); enable wait_for_async_insert=1 - Add of_node_traffic_hourly MV and dashboard traffic trend fallback - Add GET /admin/status/clickhouse operational metrics endpoint	2026-07-02 15:47:38 +08:00
ryan	578110f615	observation clickhouse	2026-06-19 17:20:39 +08:00