TiFlash node has been successfully removed, but TiDB node logs still show attempts to connect to the removed TiFlash

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

Original topic: tiflash 某个节点已经正常删除,但是tidb节点日志还是会显示要连接剔除的tiflash

| username: wluckdog

[TiDB Usage Environment] Production Environment / Testing / Poc
[TiDB Version] v6.5.0
[Reproduction Path] A certain TiFlash node was removed half a year ago. TiFlash was removed normally, but the TiDB node logs still show attempts to connect to the removed TiFlash node, resulting in connection refusals, though only as warning logs.
[Encountered Problem: Problem Phenomenon and Impact]

[2024/01/29 09:48:05.574 +08:00] [WARN] [mpp_probe.go:246] [“Store is not ready”] [“store address”=] [“err message”=“rpc error: code = Unavailable desc = connection error: desc = "transport: Error while dialing dial tcp xxxxx:3930: connect: connection refused"”]

[Resource Configuration] Enter TiDB Dashboard - Cluster Info - Hosts and take a screenshot of this page
[Attachments: Screenshots/Logs/Monitoring]

| username: TiDBer_jYQINSnf | Original post link

Check with pd-ctl, is this store a tombstone?

| username: tidb菜鸟一只 | Original post link

If you can no longer see this store in pdctl, you can try restarting the TiDB server in a loop to check.

| username: 这里介绍不了我 | Original post link

If it doesn’t affect the business, you can ignore it for now. If you really have OCD, you can check pd-ctl to see if the information is still there.

| username: wluckdog | Original post link

After deleting a large amount of data, creating an index is extremely slow, and there is this error in the log. Check if it has any impact.

| username: dba远航 | Original post link

Does the TiDB cluster display show that it has been successfully deleted?

| username: wangccsy | Original post link

It wasn’t completely deleted, right?

| username: wluckdog | Original post link

I previously wrote about normal deletion, and it’s done now.