Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.
Original topic: 表无法删除任何DDL排队中无响应
[TiDB Usage Environment] Production Environment
[TiDB Version] 6.2
[Encountered Problem] Unable to perform any DDL operations on the table, tasks are queued
[Reproduction Path] Possible cause: During large-scale data synchronization, TiDB Node’s CPU was overloaded and interrupted abnormally
[Problem Phenomenon and Impact]
Phenomenon: The table is stuck at drop index and cannot perform any DDL operations. Executing ADMIN CANCEL DDL JOBS results in a lot of errors: [ddl:8226] This job is almost finished, can’t be cancelled now, and it cannot be cancelled.
Impact: This table has become a dead table and cannot be deleted or rebuilt.
[Attachments]
When performing large-scale data synchronization, it is best to set a limit, otherwise, inexplicable issues may arise.
The current issue is figuring out how to drop this table.
Try restarting the tidb-server.
Just wait for the DDL to finish. I have already answered your question in the WeChat group. Use an SSD hard drive, don’t use a virtual machine, use a physical machine.
Are there any error logs? It seems like it only restarted.
Restarting won’t work; it doesn’t affect the DDL queue.
It just doesn’t end. It’s also impossible to rename or delete the table. It’s not related to the machine; this is still a bug.
Start canceling from the job with the largest ID. The drop index and create index are for the same table, right?
Yes. I just want to know how to delete this table from the TiKV and TiDB layers now.
Recently encountered the same issue.
Version: v6.5.0.
Actions taken that did not resolve the issue: Restarted the entire cluster, admin cancel ddl jobs, individually restarted all TiDB, individually restarted PD, individually restarted TiKV.
Action taken that resolved the issue: Stopped all TiDB, started one TiDB, and then it returned to normal.
How did the original poster solve it?
6.5.3 According to your method, I didn’t succeed. Are there any other tips in the forum?
I am using version 6.5.1 and experiencing the same issue. I don’t know what to do anymore; I’ve restarted everything, but it didn’t help.
The DDL job is stuck in rolling back and there is no progress.
Using your method, it was restored. Stop all TiDB nodes, wait for 1 minute, and start one node.
|
username: ShawU
|
username: 
|
username: 
|
username: 
|
username: 
|
username: 
|
username: 
|
username: 
|
username: 
|
username: 
|
username: