After truncating the data table in version 7.4, the TiFlash engine for the table becomes unavailable

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

Original topic: 7.4版本数据表truncate之后该表的tiflash引擎不可用

| username: Running

  1. Version 7.4
  2. Issue: After truncating a data table, the TiFlash engine for that table becomes unavailable and can only be restored after a restart.
  3. Reproduction method: On version 7.4, truncate any table and observe that the corresponding TiFlash status remains unavailable.
  4. Expected result: Automatically rebuild the corresponding table’s TiFlash engine after truncation, without affecting usage.
| username: lilinghai | Original post link

Truncate is equivalent to rebuilding a new table (observing that the table_id has changed). Normally, the new table’s TiFlash replica will be available quickly. Could it be that there are many/heavy DDL tasks in the database, affecting it?
How about manually creating a new table, setting the TiFlash replica, and observing when it becomes ready? Also, how large is the table you truncated?

| username: Running | Original post link

A data volume of around 100,000, these operations were fine in version 7.0, but issues appeared in version 7.4.

| username: lilinghai | Original post link

I tried it and couldn’t reproduce the issue. I saw a potential problem. If your cluster was upgraded, you can try the method described in this issue: not able to create table after upgrade TiDB cluster from 7.1.0 to master · Issue #46551 · pingcap/tidb · GitHub

| username: system | Original post link

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.