Configuration of online-ddl is ineffective in the scenario of merging sharded tables and databases

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: 分库分表合并场景下,配置online-ddl不生效

| username: liujia

[TiDB Usage Environment] Production Environment
[TiDB Version] v5.3.0
[DM Version] v2.0.6
[Reproduction Path] Use DM to synchronize sharded tables from MySQL to TiDB and merge them into one table. When gh-ost is used to perform DDL operations on the upstream MySQL, the synchronization reports an error that the shadow table does not exist.
[Encountered Problem: Phenomenon and Impact]
[Resource Configuration] Go to TiDB Dashboard - Cluster Info - Hosts and take a screenshot of this page
[Attachments: Screenshots/Logs/Monitoring]


The original table is performing an index addition operation: alter table chapter_report add index idx_shardIdx_userId_updatedTime (shardIdx,userId,updatedTime)

| username: Jasper | Original post link

Didn’t you filter out the operation of adding an index?

| username: Kongdom | Original post link

:flushed: Uh… so this was a misunderstanding?

| username: TiDBer_7S8XqKfl-1158 | Original post link

The teacher hit the nail on the head.

| username: TiDBer_3Cusx9uk-0775 | Original post link

Try upgrading the version.

| username: liujia | Original post link

Yes, it is intended to skip adding the index, but in fact, DM did not skip this operation and stopped proceeding after encountering an error.

| username: WalterWj | Original post link

The upstream tool you used added an index, right? It looks like the tool creates a temporary table for the rename operation. Is it possible that the temporary table was filtered out and not synchronized? In other words, the _chapter table does not exist.

If that doesn’t work, you might need to skip this SQL and handle it manually.