Consultation on Local Volume Data Persistence Issues

Translated
1

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

Original topic: 关于 local volume 数据持久化问题的咨询

| username: TiShawnwinder

【TiDB Usage Environment】Production
【TiDB Version】-
【Encountered Issues】-
【Reproduction Path】-
【Issue Phenomenon and Impact】-
【TiDB Operator Version】-
【K8s Version】1.23
【Attachments】-

Usage Question: When deploying TiKV volume using local static provisioner, if the k8s node with the mounted local SSD disk goes down, how can the persistent data in the pod on that node be recovered?

2
| username: TiDBer_jYQINSnf | Original post link

If the PVC is not deleted, the TiKV nodes are StatefulSets, for example, xxx-tikv-1. After the pod restarts, it will still be xxx-tikv-1 and will still mount the corresponding PVC, so the data will still be there. If the entire node goes down, it doesn’t matter if the downtime is short; the two replicas will run normally. If the downtime is long, the operator will create a new pod to make up for the replica. This time can be configured, and the specific configuration is in the operator’s tc.

3
| username: TiShawnwinder | Original post link

Thanks a lot!

4
| username: TiDB社区小助手 | Original post link

If the issue is resolved, you can mark the corresponding answer as “Best Answer”~

5
| username: system | Original post link

This topic will be automatically closed 60 days after the last reply. No new replies are allowed.