Unable to modify the log level for storaged.

[catchdog007]

Please check the FAQ documentation before raising an issue

Describe the bug (required)

使用kubectl edit nc nebula 修改storaged服务的config，设置minloglevel为1,2,3,4，均不生效，进入容器后，执行curl nebula-storaged-headless:19779/flags 看到的minloglevel还是0
但是，一模一样的操作，对于graphd和metad却是能立马生效的

Your Environments (required)

使用helm3安装的nebula-operator-1.7.6和nebula-cluster-1.7.0组建的nebula集群，k8s的版本是1.23.9

How To Reproduce(required)

使用上面的方式，必现

Expected behavior

Additional context

[kqzh]

hi @catchdog007 ,thank you for reporting the bug, we will fix it in the next version

[MegaByte875]

I did a test found that flag minloglevel reset to 0 ifself seconds later after operator apply the config, may it's a bug. Only storaged can reproduce the issue.

[Sahil333]

I am facing the same issue. I can reproduce the same issue in other components metad and graphd as well. It seems to be happening with "minloglevel" and "memory_tracker_limit_ratio" field in my case. Some fields like "redirect_stdout" and "stderrthreshold" are not overwritten and correct value is used given in the config field of cluster yaml.

[Sahil333]

I was looking at the configMaps in which dynamic flags are not updated instead separate http calls are made in the control loop to update these flags. I am not facing the OP's issue, my mistake.