Our latest set of patch releases contain a regression, which can cause kubelet crashes in clusters that use DockerShim (and might affect some other clusters using forks of other container runtimes). If you are still on an earlier patch version, SIG-Release advises you not to update at this time. New patch releases with a fixed regression will be available next week. Release Engineering is planning on doing one additional final patch to 1.18 with the other patch releases this month instead.
Eddie Zane of SIG-CLI started a discussion around plans to make it a tad more difficult to accidentally delete your entire cluster. Read the thread for some good discussion and comment if you need to.
Next Deadline: Feature Exceptions Due, June 28th
It’s still feature-hacking period, so if you’re not actively hacking on a feature (or even if you are), now is a good time to fix some flaky tests. This week SIG-API-Machinery, SIG-Storage, and SIG-Cloud-Provider own our new flakes, so if y’all could take a look at those that would be great. With 6 blocking tests flaking, and 13 informing tests not passing, this is gonna be one sloooow release cycle if we don’t get those fixed.
Cherry-picks for patch releases are due June 12th for a June 16th release. This will include an additional release in the 1.18 series, 1.18.20, because of the regression introduced with 1.18.19; that patch will reverse the regression and nothing more. Regardless, 1.18 users should really really plan to upgrade ASAP because there will be no further patches in that version.
kubectl get podshas new “LAST RESTART” column, so you can pick out your cyclically failing pods; congrats to Jordan Jacobelli on his first merged PR!
--sshoptions that are deprecated and only ever worked on GCE anyway
Last Week In Kubernetes Development (LWKD) is a product of some members of the Kubernetes project, but is not an official publication of the Kubernetes project or the CNCF. All original content is licensed Creative Commons Share-Alike, although linked content and images may be differently licensed. LWKD does collect some information on readers, see our privacy notice for details.
You may contribute to LWKD by submitting pull requests or issues on the LWKD github repo.