Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.
Original topic: 3个tikv上有个leader数量为0
[TiDB Usage Environment] Production Environment / Testing / Poc
[TiDB Version] 7.5
[Reproduction Path] What operations were performed when the issue occurred
[Encountered Issue: Issue Phenomenon and Impact]
Today, I checked the monitoring and found that 1 out of 3 TiKV nodes has no leader. Let’s see what the problem is.
Take a look at the PD => scheduler panel.
Sorry, I cannot translate images. Please provide the text you need translated.
This is not correct. There is an evict-leader-scheduler. You can remove this scheduler using pd-ctl.
I don’t know what those two schedulers with witnesses are for. Were they added in a higher version? Waiting for an expert to answer.
Have you experienced a node crash?
There were 2 TiKV nodes that went down abnormally before, how can this be recovered?
How to operate specifically?
- Enter pd-ctl interactive mode
tiup ctl:v6.5.5 pd -u http://192.168.1.1:2381 -i
- Check the current scheduler status
scheduler show
- Remove the scheduler with the evict-leader-scheduler keyword
scheduler remove <scheduler with evict-leader keyword from the previous command result>
That means it has been scheduled. If there are few regions, I think it’s fine. It will allocate itself later.
The leader is 0, but there is a region?
Restart this TiKV, and the leader will come back instantly.
Take a look at the scheduler panel.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.