The address is displayed incorrectly when expanding a TiKV in TiDB

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

Original topic: tidb扩容一个tikv有个地址显示不准确

| username: 舞动梦灵

Is the address information here normal? I see that for other stores, the status_address and addressIP correspond to each other. This one is different. Both places only have the IP 0.0.0.0 as the address information. Is there a configuration missing somewhere?

| username: onlyacat | Original post link

Take a look at your configuration file.

| username: 舞动梦灵 | Original post link

This new node is for scaling. The scale file:
tikv_servers:

  • host: 172.16.2.28
    ssh_port: 22444
    port: 20160
    status_port: 20180
    deploy_dir: /server/deploy
    data_dir: /server/deploy/data
    log_dir: /server/deploy/log
| username: Jack-li | Original post link

The configuration file might not have taken effect.

| username: 舞动梦灵 | Original post link

This configuration is directly copied from the previous expansion file with only one IP address changed, and it is exactly the same as the official one with no differences. I also restarted the server three or four times. It shouldn’t be like this. If the configuration didn’t take effect, the 20180 port shouldn’t be usable either. Why is the port started but the corresponding IP not used?

| username: DBAER | Original post link

Is it possible that the host has multiple IPs? The default value is 0.0.0.0, but you specified an IP. How could this still happen? Could it be that the format is not effective?

| username: 舞动梦灵 | Original post link

There is only one IP.

| username: 像风一样的男子 | Original post link

Port 20180 is used for monitoring. Can you check if the monitoring is functioning properly? Also, take a look at the configuration file /data/tidb-deploy/prometheus-8249/conf/prometheus.yml to see if xxx:20180 is correct.

| username: 舞动梦灵 | Original post link

Is this to be checked on the TiDB deployment server or on the configuration of the corresponding problematic server?

| username: 舞动梦灵 | Original post link

I found this file. The addresses inside are all correct, checked on the TiDB server segment.

| username: 舞动梦灵 | Original post link

The monitoring graph shows that there are 2 problematic stores, but when I check the store status in pd-ctl, they are all in the “up” state. However, the problematic server 2.28 has an “up” time of only a few seconds to a few minutes, and it seems to restart every one or two minutes. Its process appears to be continuously restarting.

| username: WalterWj | Original post link

:sweat: Could it be a display bug in the lower version?

| username: 舞动梦灵 | Original post link

It could be an issue with the new expansion. We are not sure if it’s a memory problem, but the TiKV service restarts every few minutes.

| username: zhaokede | Original post link

Go through the scaling process and see which configuration hasn’t been loaded.

| username: WalterWj | Original post link

Upgrade it. This version is too old… I almost forgot how to use Ansible…

| username: 健康的腰间盘 | Original post link

Is there an issue with the Putuo file? Was it copied directly from Windows, causing unrecognized characters that led to it not working?

| username: 小于同学 | Original post link

Check the configuration file again.

| username: 舞动梦灵 | Original post link

No, before using it, the previous DBA directly copied the expansion file and modified the IP address. Nothing else changed.

| username: 舞动梦灵 | Original post link

There doesn’t seem to be any problem with this scale file for expansion.
image