dns/aws: update public zone discovery to search parent domains #117
Conversation
openshift-ci-robot
added
the
do-not-merge/work-in-progress
openshift-ci-robot
added
the
size/L
The installer needs to move towards [1] where the private zone named as `cluster_name.base_domain` ie. `cluster_domain` because of [2], but the public zone still remains `base_domain` as that cannot be created by instaler. This means that cluster-ingress-operator cannot use the public r53 zone with the same name as the `base_domain` from `DNS.config.openshift.io` [3] as it will be set to the `cluster_domain`. This changes the public zone discovery to find a public zone which is the nearest parent domain to `cluster_domain`. [1]: openshift/installer#1169 [2]: openshift/installer#1136 [3]: https://github.com/openshift/api/blob/d67473e7f1907b74d1f27706260eecf0bc9f2a52/config/v1/types_dns.go#L28
abhinavdahiya
force-pushed
the
chenge_r53
branch
from
936c756 to
f6930c9
Compare
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: abhinavdahiya If they are not already assigned, you can assign the PR to them by writing The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
abhinavdahiya
changed the title
openshift-ci-robot
removed
the
do-not-merge/work-in-progress
|
/hold We need to have a discussion about the requirements here |
openshift-ci-robot
added
the
do-not-merge/hold
+1
What kind of requirements? Anything I can include in the commit message for why ? |
Where's the change to the installer which would break our current assumption? Is it now too risky for us to guess hosted zone names to do record CRUD in? Could we get the zone information from cluster config instead? What would that API look like? etc. |
Removing the distinction between "cluster name" and "base domain" in favor of a unified "cluster domain". PR moving in this direction (but not yet completely there) is openshift/installer#1169.
This sounds reasonable to me, and would save having repeated lookup logic here. As a bonus, if the ingress operator ignored the public zone when the setting was unset, it would allow us to address the "no thanks, I don't want public Route 53" use-case without further ingress changes. |
|
Replaced by #121 /close |
|
@ironcladlou: Closed this PR. DetailsIn response to this:
Instructions 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/test-infra repository. |
4 participants
The installer needs to move towards 1 where the private zone named as
cluster_name.base_domainie.cluster_domainbecause of 2, but thepublic zone still remains
base_domainas that cannot be created by instaler.This means that cluster-ingress-operator cannot use the public r53 zone with the same name as the
base_domainfromDNS.config.openshift.io3 as it will be set to thecluster_domain.This changes the public zone discovery to find a public zone which is the nearest parent domain to
cluster_domain./cc @wking @openshift/sig-network-edge