TDSQL MySQL版 DDL 操作指引

操作前检查
1. 确认是否有足够的空间进行 DDL 操作。
查看进行 DDL 操作的表当前占用的空间，并通过当前实例的磁盘使用率判断是否有足够的空间进行 DDL 操作。
SELECT 
    table_name AS '表名',
    ROUND((data_length + index_length) / 1024 / 1024, 2) AS '总空间(MB)',
    ROUND(data_length / 1024 / 1024, 2) AS '数据空间(MB)',
    ROUND(index_length / 1024 / 1024, 2) AS '索引空间(MB)',
    ROUND(data_free / 1024 / 1024, 2) AS '碎片空间(MB)'
FROM information_schema.TABLES
WHERE table_schema = '数据库名'
  AND table_name = '表名';
说明：
部分 DDL 可以使用 instant DDL，不涉及空间占用，详细请参见 OnlineDDL 说明。
2. 查看同名表是否有超长数据的慢查询。
SELECT * FROM information_schema.processlist WHERE INFO LIKE "%表名%" ORDER BY TIME_MS DESC LIMIT 10;
若有，则等慢查询结束后再执行 DDL。
3. 针对大单表，查看数据分布，确认 IO 压力。
SELECT 
    SUM(region_stats_approximate_size) AS size, 
    COUNT(b.rep_group_id) AS region_nums, 
    sql_addr, 
    c.leader_node_name, 
    b.rep_group_id 
FROM 
    information_schema.META_CLUSTER_DATA_OBJECTS a 
    JOIN information_schema.META_CLUSTER_REGIONS b 
    JOIN information_schema.META_CLUSTER_RGS c 
    JOIN information_schema.META_CLUSTER_NODES d 
    ON a.data_obj_id = b.data_obj_id 
    AND b.rep_group_id = c.rep_group_id 
    AND c.leader_node_name = d.node_name 
WHERE 
    a.schema_name = '数据库名' 
    AND a.table_name = '表名' 
GROUP BY 
    rep_group_id 
ORDER BY 
    leader_node_name;
如果数据全部倾斜在某一个节点，建议将 max_parallel_ddl_degree 从默认8设置为4或者2，减少 DDL 并发线程以降低 IO 压力。
4. 通过租户端监控指标确认 CPU/IO 负载。
建议在业务低峰期进行 DDL 操作。
操作中查看进度
-- 查看DDL，是否在执行，LAST_TIMESTAMP 若在更新，则表明 DDL 在执行，如果长时间没有更新，则可能 hang 住
SELECT * FROM information_schema.ddl_jobs WHERE is_history = 0;
﻿
-- 查看进度则关注 information_Schema.ddl_jobs INFO 字段中的"progress"信息
             ID: 13
    SCHEMA_NAME: tdstore
     TABLE_NAME: sbtest1
        VERSION: 13
     DDL_STATUS: SUCCESS
START_TIMESTAMP: 2025-08-08 14:29:35
 LAST_TIMESTAMP: 2025-08-08 14:29:35
        DDL_SQL: alter table tdstore.sbtest1 add index idx(v)
      INFO_TYPE: ALTER TABLE
           INFO: {"tmp_tbl":{"db":"tdstore","table":"#sql-d_1000be_6895994e0000ad_1"},"alt_type":1,"alt_tid_upd":{"tid_from":10039,"tid_to":10039},"cr_idx":[{"id":10040,"ver":4,"stat":0,"tbl_type":1,"idx_type":2}],"rm_idx":[],"init":false,"tmp_tab":false,"online_op":true,"wf_rmed":false,"online_copy_stage":0,"idx_op":true,"row_applied":true,"row_apply_saved":true,"current_schema_name":"tdstore","crt_data_obj_task_id":29437883249066288,"dstr_data_obj_task_id":0,"alt_tbl_pp_stage":0,"alt_tbl_policy_option":0,"data_obj_to_be_dstr_arr":[],"progress":"total: 1, scanned: 1 (100.00%)","fillback_mode":"ThomasWrite","exec_addr":{"ip":"10.10.10.10","port":15035},"recov_addr":{"ip":"10.10.10.10","port":15035}}
     IS_HISTORY: 1
﻿
--通过租户端观察 CPU/IO 负载以及慢查询告警；
INFO 字段中的 progress 即是进度说明： "progress":"total: 1, scanned: 1 (100.00%)"
查看任务结果
--返回empty， 则表明 DDL job 执行结束
SELECT * FROM information_schema.ddl_jobs WHERE is_history = 0;
DDL 操作指引

本页目录：

操作前检查

操作中查看进度

查看任务结果
