案例1
[問(wèn)題澄清]
TiKV啟動(dòng)過(guò)程中報(bào)錯(cuò)：
[ERROR] [util.rs:326] [“request failed”] [err=“Grpc(RpcFailure(RpcStatus { status: RpcStatusCode(2), details: Some(“duplicated store address: id:297732 address:“192.xxx.1.34:20160” version:“4.0.0-alpha” , already registered by id:1 address:“192.xxx.1.34:20160” version:“3.0.0-beta.1” “) }))”] [2019/10/23 12:10:00.094 +08:00] [FATAL] [server.rs:310] [“failed to start node: Other(”[components/pd_client/src/util.rs:334]: fail to request”)”]
[原因分析]
節(jié)點(diǎn)之前部署過(guò) TiKV，縮容沒(méi)有清理干凈
[解決方案]
可能是之前舊版本的目錄沒(méi)有清理梭伐，清理完畢就可以正常啟動(dòng).
[參考案例]
https://asktug.com/t/tikv/1491
[學(xué)習(xí)引申點(diǎn)]
使用TiDB Ansible擴(kuò)容縮容TiDB集群
https://pingcap.com/docs-cn/stable/how-to/scale/with-ansible/
案例2
[問(wèn)題澄清]
PD啟動(dòng)過(guò)程中報(bào)錯(cuò)：
[FATAL] [main.go:111] ["run server failed"] [error="Etcd cluster ID mismatch, expect 14363702570076372167, got 2436494335309057221"]
[原因分析]
擴(kuò)容過(guò)程中贰拿，使用舊服務(wù)器，沒(méi)有清理干凈
[解決方案]
按照官方文檔擴(kuò)容，如果之前有舊的PD數(shù)據(jù)雪猪，需要清理干凈
[參考案例]
TiDB 3.0.2 PD擴(kuò)容 [error] rafthttp: [request sent was ignored (cluster ID mismatch:? remote
[學(xué)習(xí)引申點(diǎn)]
使用TiDB Ansible擴(kuò)容縮容TiDB集群
https://pingcap.com/docs-cn/stable/how-to/scale/with-ansible/
