-
Notifications
You must be signed in to change notification settings - Fork 520
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Remove two-deployment mode #157
Comments
I will take a look at this. |
@dmitryax If we are willing to break backwards compatibility for both modes enabled, should we just rip the bandaid off and make |
I think majority of the use cases is enabled one mode or another. So I would say we should not break their upgrade |
The situation I dislike is that since
which relies on
I don't love that the default values.yaml would still be using the agentCollector configs and any new user would have to reference the old config setting even though we are trying to deprecate it :( |
Sorry I don't think I understand your concern. Are you saying that users have to set |
If users want to use the I think the only way to deal with that potentially confusing |
What if we don't set
Later we will just drop |
Yes I like that a lot. I will get the PR updated with that strategy, see how it comes out. |
As agreed in #91, we need remove the two-deployment mode and simplify the configuration.
Instead of
agentCollector
andstandaloneCollector
, we should have another config optionmode
which can have one of the following values:daemonset
ordeployment
. All the nested configuration options fromagentCollector
andstandaloneCollector
should be moved to the root level. Some of them can be no-op based on the selectedmode
.Ideally we need to provide backward compatibility with existing interface for now. It means that we translate
agentCollector
andstandaloneCollector
configs into the new config and show deprecation warning if they are used.If user has both modes enabled (
agentCollector.enabled=true
andstandaloneCollector.enabled=true
) the upgrade should fail. This can be achieved by updatingschema.values.json
. For this use case we need to provide a detailed upgrade guideline on how to replace it with two separate helm installations.The text was updated successfully, but these errors were encountered: