schema_version: 1
id: database-time-spike-with-storage-read-wait
title: Database time spike with storage read wait
status: draft
owner: codex
source: jira
environment:
  profile: stable-dedicated
  cluster_id: '10244988415975579886'
  cluster_type: dedicated
  tidb_version: v8.5.3
time_window:
  incident_at: '2026-08-19T02:32:00Z'
  allowed:
  - id: reported_incident_window
    start: '2026-08-19T02:25:00Z'
    end: '2026-08-19T02:40:00Z'
prompt:
  text: 请分析集群 10244988415975579886 在 2026-08-19T02:25:00Z 到 2026-08-19T02:40:00Z 的 database time 短时上升。02:32Z 左右 TiKV disk read wait latency 和 coprocessor disk IO time 同步出现 spike，主要标签是 `copr-select`；同一天还有一些因 TiDB memory quota 被 kill 的查询。请判断这次短时 spike 的直接原因，并区分无关的并发现象。
tags:
- sql
- latency
- storage
case_tier: P1
cost_tier: medium
followup:
  scheduled_max_followups: 0
execution:
  acp_bash_timeout_s: 900