KEP-1645: Fix typo in Upgrade / Downgrade Strategy section#5973
KEP-1645: Fix typo in Upgrade / Downgrade Strategy section#5973AnupamGhosh wants to merge 2 commits intokubernetes:masterfrom
Conversation
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: AnupamGhosh The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
Welcome @AnupamGhosh! |
|
Hi @AnupamGhosh. Thanks for your PR. I'm waiting for a kubernetes member to verify that this patch is reasonable to test. If it is, they should reply with Regular contributors should join the org to skip this step. Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. DetailsInstructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
There was a problem hiding this comment.
Thanks for the fix! I think those are mostly from a time where MCS was supposed to be natively implemented by kube-proxy. So I think those paragraph could be either removed or potentially replaced by something explaining that this an out-of-tree API and this should be addressed by each implementations individually.
|
/ok-to-test |
|
I agree with @MrFreezeex, we can take out the entire paragraph in this PRR, except the one sentence that says
|
Clarify upgrade/downgrade and version skew strategies for MCS services.
|
Thanks @MrFreezeex & @lauralorenz for the feedback! I have removed kube-proxy as a requirement for upgrading MCS API and made the change to have a DNS provider that resolves multicluster service and a mcs-controller that implements the MCS API in the upgrade/downgrade & version skew section. |
|
TBH I am not sure we should even mention the DNS provider explicitly here. It's the responsibility of the implementation to handle DNS that and we never talk about a "DNS provider" in the rest of the KEP so this come in a weird way here. I think in both cases if we want to explain something we could just mention that we provide some out of tree CRDs which follows Kubernetes best practices (as you also mentioned in the current PR) and that implementation should defined their own strategy and compatibility matrix with the CRD versions that we provide ~ |
|
Yeah, I agree that the wording "DNS provider" falls out of line with the context of the KEP. However, there are two distinct implementations here:
I think being explicit about these two distinct implementations makes it clearer about which components need to be updated. What do you think? |
I think the missing word is "DNS" here as: