TiKV Node in TiDB Cluster Fails to Start

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

Original topic: tidb集群的tikv节点起不来

| username: TiDBer_vJpITQ5J

[TiDB Usage Environment] Production Environment
If the TiKV node in the TiDB cluster fails to start, can it be directly terminated and restarted?

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

Send over the topology structure and let me take a look :upside_down_face:

| username: TiDBer_vJpITQ5J | Original post link

The image is not visible. Please provide the text you need translated.

| username: TiDBer_vJpITQ5J | Original post link

The image is not visible. Please provide the text you need translated.

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

Premise: The machine is indeed no longer operational.

If there are 3 replicas, you can remove it, and the region replicas will be automatically replenished on other nodes.

After replenishing the replicas, you need to perform a decommissioning operation on the faulty node.

| username: TiDBer_vJpITQ5J | Original post link

The node indeed cannot start~~~ So can I directly use tiup cluster scale-in tidb-prod -N 10.200.78.51:20160 to take the node offline?

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

  1. First, confirm if it is a three-replica setup.

  2. After other nodes have replenished the region replicas, proceed with the normal decommissioning operation.

PS: You can use pd ctl to check the number of regions on that store.

| username: tidb菜鸟一只 | Original post link

If there are 3 replicas, you can directly scale down if one out of five nodes fails.

| username: TiDBer_vJpITQ5J | Original post link

How can I confirm if it is a three-replica setup?

| username: Jiawei | Original post link

Look at this parameter: max-replicas

| username: tidb菜鸟一只 | Original post link

SHOW config WHERE NAME LIKE ‘%max-replicas%’