example release note v1.7.0..v1.7.2

example-relnote-k8s.md

Other notable changes

• Api-Machinery

  • Admission-Control

    • Orphaned objects cannot be deleted once their namespace is removed (#49027)

      • Never prevent deletion of resources as part of namespace lifecycle (#48733, @liggitt)

• Apps

  • NullArea

    • DaemonSet controller doesn't handle NoSchedule taints correctly (#48190)

      • Support NoSchedule taints correctly in DaemonSet controller. (#48189, @mikedanese)

• Auth

  • NullArea

    • NullIssue

      • RBAC role and role-binding reconciliation now ensures namespaces exist when reconciling on startup. (#48480, @liggitt)

• Autoscaling

  • NullArea

    • hpa scales below MinReplicas if desiredReplicas is zero (#49028)

      • hpa: Prevent scaling below MinReplicas if desiredReplicas is zero (#48997, @johanneswuerbach)

• Azure

  • NullArea

    • NullIssue

      • Azure PD (Managed/Blob) (#46360, @khenidak)

• Cluster-Lifecycle

  • NullArea

    • NullIssue

      • kubeadm: Make kube-proxy tolerate the external cloud provider taint so that an external cloud provider can be easily used on top of kubeadm (#49017, @luxas)
      • Added new flag to kubeadm init: --node-name, that lets you specify the name of the Node object that will be created (#48594, @GheRivero)

  • Security

    • NullIssue

      • kubeadm: Expose only the cluster-info ConfigMap in the kube-public ns (#48050, @luxas)

• Instrumentation

  • NullArea

    • Stackdriver event exporter too noisy when watch connection is reset (#47914)

      • Reduce amount of noise in Stackdriver Logging, generated by the event-exporter component in the fluentd-gcp addon. (#48712, @crassirostris)

• Network

  • Kube-Proxy

    • Services (kube-proxy): conntrack for UDP needs to be flushed when set of endpoints goes from empty to non-empty (#48370)

      • Fix udp service blackhole problem when number of backends changes from 0 to non-0 (#48524, @freehan)

  • NullArea

    • Port collision between kube-proxy and node-problem-detector (standalone mode) (#49263)

      • Use port 20256 for node-problem-detector in standalone mode. (#49316, @ajitak)

  • Platform/Gce

    • GCE: Services (LoadBalancer) with static ip causes panic in 1.7 (#48848)

      • Fix service controller crash loop when Service with GCP LoadBalancer uses static IP (#48848, @nicksardo) (#48849, @nicksardo)

• Node

  • NullArea

    • NullIssue

      • Fix pods failing to start when subPath is a dangling symlink from kubelet point of view, which can happen if it is running inside a container (#48555, @redbaron)

• NullSig

  • NullArea

    • NullIssue

      • Add generic NoSchedule toleration to fluentd in gcp config. (#48182, @gmarek)
      • GCE Cloud Provider: New created LoadBalancer type Service will have health checks for nodes by default if all nodes have version >= v1.7.2. (#49330, @MrHohn)
      • kubectl: Fix bug that showed terminated/evicted pods even without --show-all. (#48786, @janetkuo)
      • Add a runtime warning about the kubeadm default token TTL changes. (#48838, @mattmoyer)
      • kubeadm: Fix a bug where kubeadm join would wait 5 seconds without doing anything. Now kubeadm join executes the tasks immediately. (#48737, @mattmoyer)
      • Fix a regression that broke the --config flag for kubeadm init. (#48915, @mattmoyer)
      • Added new flag to kubeadm join: --node-name, that lets you specify the name of the Node object that's gonna be created (#48538, @GheRivero)
      • Fixes issue where you could not mount NFS or glusterFS volumes using hostnames on GCI/GKE with COS images. (#42376, @jingxu97)

• Release

  • Security

    • NullIssue

      • kubeadm: Expose only the cluster-info ConfigMap in the kube-public ns (#48050, @luxas)

• Storage

  • NullArea

    • A pod using a Portworx volume driver PVC gets stuck in ContainerCreating phase (#49034)

      • Fix Pods using Portworx volumes getting stuck in ContainerCreating phase. (#48898, @harsh-px)

    • Kubernetes 1.7.0 local storage teardown failed (#48331)

      • Local storage teardown fix (#48402, @ianchakeres)

    • NullIssue

      • Azure PD (Managed/Blob) (#46360, @khenidak)
      • Fix pods failing to start when subPath is a dangling symlink from kubelet point of view, which can happen if it is running inside a container (#48555, @redbaron)