Error When Expanding Nodes

Translated
1

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

Original topic: 扩容节点报错

| username: TiDBer_QHSxuEa1

When I was testing the expansion of TIKV nodes, I encountered an error. I expanded a total of two nodes, one was successful, and the other was not. What could be the reason?
image

image
image1380×108 17.8 KB

image
image1380×278 65.6 KB

Below is the error log
image
image1380×84 30.9 KB

2
| username: 裤衩儿飞上天 | Original post link

Was node 103 previously in another cluster? Was it not completely uninstalled? Is there residual data?

3
| username: TiDBer_QHSxuEa1 | Original post link

Yes, I have also previously expanded TiKV nodes for other clusters.

4
| username: 裤衩儿飞上天 | Original post link

  1. If the node has already been scaled down from other clusters, then clean up all related directories.

  2. If the node has not been scaled down from other clusters, then first perform a normal scale-down (provided that this node is no longer needed in the old cluster), and then join the new cluster.

5
| username: TiDBer_QHSxuEa1 | Original post link

I have cleaned up the deploy_dir, data_dir, and log_dir corresponding to TIKV. However, I am unable to scale down this tikv node from the cluster. Attempting to directly scale up this node also fails.

6
| username: TiDBer_QHSxuEa1 | Original post link

The image is not available for translation. Please provide the text content directly for translation.

7
| username: 裤衩儿飞上天 | Original post link

After executing scale-in, check the status with display.
tiup cluster display XXXX

8
| username: TiDBer_QHSxuEa1 | Original post link

The status doesn’t seem to have changed.

9
| username: 裤衩儿飞上天 | Original post link
  1. Use pd-ctl to check the status and ID of the store.
  2. Then use store delete XXX.
  3. Use pd-ctl to check the status again.
  4. Display the status.
10
| username: TiDBer_QHSxuEa1 | Original post link

pd-ctl cannot find information about the node 192.168.10.103 that cannot be connected :sob:

11
| username: 裤衩儿飞上天 | Original post link
  1. Then delete this node from the tiup configuration file, and check with display. The node should no longer be there.
  2. After the node is no longer displayed, clean up all the TiDB-related directories of this node, and then re-add it.
12
| username: 裤衩儿飞上天 | Original post link

Configuration file directory: /home/tidb/.tiup/storage/cluster/clusters/XXXX/meta.yaml
XXXX is the cluster name

Make a backup before modifying

13
| username: TiDBer_QHSxuEa1 | Original post link

It worked! Thank you very much!

14
| username: system | Original post link

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