File kubernetes1.23.changes of Package kubernetes1.23
------------------------------------------------------------------- Wed Mar 16 12:29:58 UTC 2022 - rbrown@suse.com - Update to version 1.23.4: * Update Go to 1.17.7 * Use serializable struct for x-kubernetes-validations in openapi * Make JSON schema round tripping test more strict * ignore CRI PodSandboxNetworkStatus for host network pods * set secondary address on host-network pods * Deeply copy JSONSchemaProps.XValidations. * Ensure the execHostnameTest() compares hostnames * Revert "Fix comparison between FQDN and hostname" * service REST: Call Decorator(old) on update path * add namespace in azurefile volumeid * fix: azurefile volumeid conflict in csi migration * Mark device as uncertain if unmount device succeeds * Update CHANGELOG/CHANGELOG-1.23.md for v1.23.3 * kubelet: fix podstatus not containing pod full name * Fix bug with node restriction blocking pvc.status.resizestatus change * Fix regression pruning array fields with x-kubernetes-preserve-unknown-fields: true * Set max results if its not set * Update CHANGELOG/CHANGELOG-1.23.md for v1.23.2 * Update k/utils to v0.0.0-20211116205334-6203023598ed * [go] update to Go 1.17.6 * fix: remove outdated ipv4 route when the corresponding node is deleted * fix: delete non existing disk issue * Revert "Automated cherry pick of #107554: Correct the feature gate string for RBD migration." * fix containers order after applying * generated: ./hack/update-vendor.sh * upgrade sigs.k8s.io/structured-merge-diff/v4 to v4.2.1 * Execute sync before taking the snapshot * Correct the feature gate string for RBD migration. * fix: azuredisk parameter lowercase translation issue * removed unnecessary log line * kubectl: add integration test for result reporting * cli: let kubectl handle error printing * cli: avoid logging command line errors in more cases * Fix header mutation race in timeout filter * clear pod's .status.nominatedNodeName when necessary * use node informer to check volumes attachment status before backoff * When volume is not marked in-use, do not backoff * kubeadm: remove the restriction that the ca.crt can only contain one certificate * flake fix: remove the error handler for cronjob integration test * Fix the leak of vSphere client sessions * fix nil pointer in create secret commands * Fix order of commands in the snapshot tests for persistent volumes * client-go: Clear the ResourceVersionMatch on paged list calls * Improving performance of EndpointSlice controller metrics cache * fix the error when cleaning up jobs for cronjob * Update CHANGELOG to add missing release notes. * apf: ensure exempt request notes the classification * Enabling kube-proxy metrics on windows kernel mode * Update CHANGELOG/CHANGELOG-1.23.md for v1.23.1 * add gce loadbalancer no-op finalizer and existingFwdRule tests * disable gce service handling if has rbs forwarding rule * add ELBRbsFinalizer * add gce elb rbs opt-in annotation * cherry pick of knp 0.0.27 * Remove JSON logging performance regression * Re-introduce removed kubectl --dry-run values. * Point flowcontrol users at v1beta2 * [go1.17] Update to go1.17.5 * dependencies: Update golang.org/x/net to v0.0.0-20211209124913-491a49abca63 * mount-utils: Detect potential stale file handle * Skip creating HNS loadbalancer with empty endpoints * Add regression test for CPUManager distribute NUMA algorithm * Add unit test for CPUManager distribute NUMA algorithm verifying fixes * Fix accounting bug in CPUManager distribute NUMA policy * Fix error handling in CPUManager distribute NUMA tests * Add a sum() helper to the CPUManager cpuassignment logic * Allow the map.Values() function in the CPUManager to take a set of keys * Fix CPUManager algo to calculate min NUMA nodes needed for distribution * Fix unit tests following bug fix in CPUManager for map functions (2/2) * Fix unit tests following bug fix in CPUManager for map functions (1/2) * Fix bug in CPUManager map.Keys() and map.Values() implementations * Ensure we balance across *all* NUMA nodes in NUMA distribution algo * Short-circuit CPUManager distribute NUMA algo for unusable cpuGroupSize * Round the CPUManager mean and stddev calculations to the nearest 1000th * updated deprecation messages from 1.23 to 1.24 * kubelet: set failed phase during graceful shutdown * kubeadm: avoid requiring a CA key during kubeconfig expiration checks * kubeadm: print the CA of kubeconfig files in "check expiration" * kubeadm: validate local etcd certficates during expiration checks * publishing-bot/doc: add component-helpers to the readme * publishing-bot/rules: remove non existing component-helpers branch 1.19 from the rules * Changelog: mention kube-scheduler bits deprication * rbd: initialize ceph monitors slice with an empty value. * Direct v2betaX users to migrate to HPA v2 * DelegateFSGroupToCSIDriver e2e: skip tests with chgrp * Update CHANGELOG/CHANGELOG-1.23.md for v1.23.0 * [go1.17] Update to go1.17.4 ------------------------------------------------------------------- Mon Feb 7 16:21:21 UTC 2022 - Dirk Müller <dmueller@suse.com> - avoid bashism in client-common postinstall script (bsc#1195391) ------------------------------------------------------------------- Thu Jan 13 12:26:35 UTC 2022 - Richard Brown <rbrown@suse.com> - Increase _constraints to 13GB ------------------------------------------------------------------- Thu Dec 16 09:10:32 UTC 2021 - Richard Brown <rbrown@suse.com> - Restore & rebase revert-coredns-image-renaming.patch from kubernetes1.22. Looks like it's still needed until all supported k8s versions allow us to change how we publish coredns containers ------------------------------------------------------------------- Wed Dec 8 14:51:07 UTC 2021 - Richard Brown <rbrown@suse.com> - Initial Package
