I noticed that there are a handful of runtime flags that aren't exposed as helm values, and there's no way to set them in the underlying deployment yaml without modifying the chart:

--log-level-key
--log-level-encoder
--metrics-addr
--host
--cert-dir
--enable-pprof
--pprof-port
--cert-service-name
--api-cache-dir
--debug-use-fake-pod
--otlp-endpoint
--otlp-metric-interval
--stackdriver-only-when-available
--stackdriver-metric-interval
--readiness-retries
--disable-enforcementaction-validation
--client-ca-name
--client-cn-name

Unless there's a reason they're not exposed, helm chart users should be able to set these either via helm values or a extraArgs section.

@JaydipGabani Cool, thanks for clarifying! So any flag could be passed when running the binary, but I think we should also add those missing flags I mentioned in #3358 (comment) to the chart as part of this unless there's a reason they're not exposed.

My thought is users should be able to set any available flag via the helm chart as well, and it doesn't look like they can at the moment. We could add the missing flags as individual helm values, or add a extraArgs section where they could be passed in and applied to their respective deployments. WDYT?

I can keep this simple for now and add a docs page with a table for each flag.

Originally posted by @ianstanton in #3358