# Altinity Expert Clickhouse Part Log

> Diagnose ClickHouse issues by analyzing system.part_log (part creation, merges, mutations, downloads, removals, moves). Use for too many parts / micro-batch inserts, merge backlog or slow merges, mutation storms (ALTER DELETE/UPDATE), unusual replication DownloadPart churn, unexpected RemovePart spikes, or ZooKeeper/Keeper znode growth correlated with part activity.

- Skill: `altinity/altinity-expert-clickhouse-part-log` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add altinity/altinity-expert-clickhouse-part-log`
- Raw SKILL.md: https://api.skillmd.com/api/skills/altinity/altinity-expert-clickhouse-part-log/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Product & Planning
- License: Apache-2.0
- Author: altinity (https://skillmd.com/u/altinity)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/altinity/altinity-expert-clickhouse-part-log

---


# Part Log Based Diagnostics

Run all queries from `checks.sql` in this skill's directory (cluster-wide) and interpret the top offenders by **rate** (events/min), **volume** (rows/bytes), and **errors**.

Notes:
- Default timeframes are relative (e.g., last 1h/6h/24h). Only switch to an explicit time range when the user provides one in the prompt.
- Replace `{cluster}` with your ClickHouse cluster name (DataGrip).
- Keep queries time-bounded (`event_time > now() - INTERVAL ...`) and use `LIMIT`.
- If a query fails due to schema differences, run `DESCRIBE TABLE system.part_log` and drop only missing fields.

Cross-module triggers:
- High `NewPart` rate / micro-batches → load `altinity-expert-clickhouse-ingestion` + `altinity-expert-clickhouse-merges`
- High `MutatePart` rate → load `altinity-expert-clickhouse-mutations`
- Many `DownloadPart` → load `altinity-expert-clickhouse-replication`
- Merge saturation / slow merges → load `altinity-expert-clickhouse-merges` + `altinity-expert-clickhouse-storage`