[prometheus-node-exporter] extraArgs forces users to duplicate boilerplate

[holmesb]

Describe the bug a clear and concise description of what the bug is.

Hi,

When adding a custom arg to prometheus-node-exporter's extraArgs (ie. --some-extra-arg), the entire extraArgs list is replaced rather than merged with the chart defaults. This forces users to manually reproduce the default args in their own values, which is brittle.
The defaults defined in the chart at the time of writing are:

extraArgs:
  - --collector.filesystem.ignored-mount-points=^/(dev|proc|sys|var/lib/docker/.+|var/lib/kubelet/.+)($|/)
  - --collector.filesystem.ignored-fs-types=^(autofs|binfmt_misc|bpf|cgroup2?|configfs|debugfs|devpts|devtmpfs|fusectl|hugetlbfs|iso9660|mqueue|nsfs|overlay|proc|procfs|pstore|rpc_pipefs|securityfs|selinuxfs|squashfs|sysfs|tracefs)$

If a user wants to add a single flag, they must copy these defaults verbatim into their own values or silently lose them, with no indication that anything is wrong until filesystem metrics disappear.

Expected behaviour
Custom extraArgs entries should be appended to the defaults, not replace them. Perhaps move the above into a new defaultArgs list, and then include both this and any extraArgs under args: in the daemonset.yaml template.

PS. I haven't checked the other charts to see whether this issue also applies to their extraArgs

What's your helm version?

v3.x

What's your kubectl version?

v1.34

Which chart?

prometheus-node-exporter

What's the chart version?

4.53

[zeritti]

Thanks for raising this issue. Indeed, from a user's perspective, such a case is not practical. It occurs in other charts and with other fields as well. There are basically two options: making the field a map or introducing a new field like you suggest. My opinion is that extraArgs should be empty by default (they are extra, after all), regardless of its type. A PR will be appreciated!