Issue with Unsuccessful Modification of PD Parameters

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

Original topic: pd参数修改没成功的问题

| username: zhanggame1

[TiDB Usage Environment] /Test/
[TiDB Version]
[Reproduction Path] What operations were performed when the issue occurred
[Encountered Issue: Issue Phenomenon and Impact]
Used tiup cluster edit-config tidb-test
Added 3 lines log.level: warn to modify the log levels of PD, TiKV, and TiDB
Then tiup cluster reload tidb-test

image
After restarting, it was found that the PD log level was not modified

| username: ShawnYan | Original post link

Check the documentation again:

For PD configuration items that can be modified online, the changes will be persisted to etcd after a successful modification, but will not be persisted to the configuration file. Subsequent configurations will be based on the settings in etcd.

set config pd `log.level`='info'
| username: zhanggame1 | Original post link

It is not deployed on Kubernetes.

| username: ShawnYan | Original post link

Read it again,

| username: redgame | Original post link

Set config tikv log-level=‘warn’;

| username: tidb菜鸟一只 | Original post link

Set config PD log.level = ‘warn’;
PD does have this situation, you can directly modify it with the command.

| username: zhanggame1 | Original post link

In other words, the parameter file cannot be changed?

| username: zhanggame1 | Original post link

Can it be changed through the parameter file?

| username: tidb菜鸟一只 | Original post link

The PD configuration items that can be modified online will be persisted to etcd after successful modification. They will not be persisted to the configuration file, and the subsequent configurations in etcd will take precedence. Modifying the configuration file will be ineffective.

| username: zhanggame1 | Original post link

I know there are configuration items that can be modified online, but it should also support configuration file modifications, right?

| username: zhanggame1 | Original post link

Re-deployed the cluster and set the config pd log.level = 'warn'; it works.

| username: system | Original post link

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.