This is coming from a discussion I've had with David where we need to support CSR v1 in 4.5.
I’m adding UpcomingSprint, because I was occupied by fixing bugs with higher priority/severity, developing new features with higher priority, or developing new features to improve stability at a macro level. I will revisit this bug next sprint.
This bug hasn't had any activity in the last 30 days. Maybe the problem got resolved, was a duplicate of something else, or became less pressing for some reason - or maybe it's still relevant but just hasn't been looked at yet. As such, we're marking this bug as "LifecycleStale" and decreasing the severity/priority. If you have further information on the current state of the bug, please update it, otherwise this bug can be closed in about 7 days. The information can be, for example, that the problem still occurs, that you still want the feature, that more information is needed, or that the bug is (for whatever reason) no longer relevant.
The LifecycleStale keyword was removed because the needinfo? flag was reset. The bug assignee was notified.
The Slack discussion before Slack prunes it from archive: David Eads Jun 10th at 16:41 CSR to v1 reviews @maszulik I think kube-controller-manager will need to bump after we get fixes in for 1.19 so that older kube-controller-manager works with newer servers during upgrades Maciej Szulik 3 months ago @deads2k if CSR is being bumped to v1 we'll be getting that with k8s 1.19 bump, are you referring to backporting this to 4.5? David Eads 3 months ago @maszulik there are kcm fixes to work with a v1 api present I think. From jordan and i tagged picks last release Maciej Szulik 3 months ago @deads2k so I'm not quite sure what's the request here David Eads 3 months ago @maszulik after the next kube minor 1.18 release, update 4.5 Maciej Szulik 3 months ago :+1: so my thinking was correct, lemme create a BZ for 4.5.z not to forget Tomáš Nožička 2 months ago Sounds like we need to disable upgrade edges for anything lower then the particular 4.5.z having the bump. How did upstream managed to break +1 compatibility and raise it to particular z stream? (edited)
We talked about this on Tue Arch. David explained that this is for the case when your upgrade stucks for long enough so you need old KCM instance to issue kubelet CSRs. This technically targets 4.5.z but in 4.6.0 timeframe. Maciej said he'll have somebody to do 1.18.z rebase for 4.5 but it's not yet determined who.
Neither in https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.18.md nor in Jordan's PR history for 1.18 I have found anything matching. @deads please link the necessary PRs.
I spoke with David and Jordan and the only necessary fixes are going to merge along with https://bugzilla.redhat.com/show_bug.cgi?id=1860789 in https://github.com/openshift/oc/pull/551 which bumps oc to latest k8s 1.18.8. I'm closing this as not a bug, then.
Removing UpgradeBlocker from this older bug, to remove it from the suspect queue described in [1]. If you feel like this bug still needs to be a suspect, please add keyword again. [1]: https://github.com/openshift/enhancements/pull/475