菜单
Diagnose server-state issues using running queries, slow-query history, and server info.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | system-troubleshooting |
| description | Diagnose server-state issues using running queries, slow-query history, and server info. |
The user is asking about the live state of the ClickHouse server: it feels slow, queries are stuck, memory looks high, or they want to know what's running now or what ran heavy recently. Do NOT use this to optimize a specific SQL they gave you (use query-optimization), or for storage/schema questions (use schema-diagnosis).
get_server_info — version + uptime context first.get_running_queries — what's executing NOW (elapsed seconds, memory). Use for "stuck"/"slow right now".get_slow_queries — recently FINISHED heavy/slow queries from system.query_log. Use for "what was slow earlier".run_select_query against system.metrics / system.asynchronous_metrics — only when you need a specific live metric value.system-table-reference reference skill — BEFORE writing any raw system.* SELECT, so you use exact column names (e.g. system.errors has last_error_time, not event_time).FORMAT clause.Rules and strategies for exploring databases, tables, schemas, and searching metadata.
Complex rules for calling render_chart, inferring axes, and choosing chart types based on user intent.
Diagnose a ClickHouse server error (code/name/exception) and give a concrete, actionable fix.
Diagnose part/partition health of a MergeTree table — too many parts, slow/stuck merges, bad partition key.
Rules for using explain/analyze/optimize tools to help the user tune their queries.
Diagnose a column/table schema issue — Nullable overhead, oversized integer, or weak compression — and propose an ALTER.