Google 云平臺的開源容器集群管理系統 ?Kubernetes v1.3.0-alpha.5 發布

Kubernetes是Google開源的容器集群管理系統，其提供應用部署、維護、 擴展機制等功能。利用Kubernetes能方便地管理跨機器運行容器化的應用。

   如果你曾經用過Docker容器技術部署容器，那么可以將Docker看成Kubernetes內部使用的低級別組件。Kubernetes不僅僅支持Docker，還支持Rocket，這是另一種容器技術。

   使用Kubernetes可以：

• 自動化容器的部署和復制
• 隨時擴展或收縮容器規模
• 將容器組織成組，并且提供容器間的負載均衡
• 很容易地升級應用程序容器的新版本
• 提供容器彈性，如果容器失效就替換它，等等...

更新日志

Action Required

• Add direct serializer (#26251, @caesarxuchao)
• Add a NodeCondition "NetworkUnavaiable" to prevent scheduling onto a node until the routes have been created (#26415, @wojtek-t)
• Add garbage collector into kube-controller-manager (#26341, @caesarxuchao)
• Add orphaning finalizer logic to GC (#25599, @caesarxuchao)
• GCI-backed masters mount srv/kubernetes and srv/sshproxy in the right place (#26238, @ihmccreery)
• Updaing QoS policy to be at the pod level (#14943, @vishh)
• add CIDR allocator for NodeController (#19242, @mqliang)
• Adding garbage collector controller (#24509, @caesarxuchao)

Other notable changes

• Fix a bug with pluralization of third party resources (#25374, @brendandburns)
• Run l7 controller on master (#26048, @bprashanth)
• AWS: ELB proxy protocol support via annotation service.beta.kubernetes.io/aws-load-balancer-proxy-protocol (#24569,@williamsandrew)
• kubectl run --restart=Never creates pods (#25253, @soltysh)
• Add LabelSelector to PersistentVolumeClaimSpec (#25917, @pmorie)
• Removed metrics api group (#26073, @piosz)
• Fixed check in kubectl autoscale. (#26162, @jszczepkowski)
• Add support for 3rd party objects to kubectl label (#24882, @brendandburns)
• Move shell completion generation into 'kubectl completion' command (#23801, @sttts)
• Setting TLS1.2 minimum because TLS1.0 and TLS1.1 are vulnerable (#26169, @victorgp)
• Kubelet: Periodically reporting image pulling progress in log (#26145, @Random-Liu)
• Federation service controller (#26034, @mfanjie)
• Stabilize map order in kubectl describe (#26046, @timoreimann)
• Google Cloud DNS dnsprovider - replacement for #25389 (#26020, @quinton-hoole)
• Fix system container detection in kubelet on systemd (#25982, @derekwaynecarr)
• Added pods-per-core to kubelet. #25762 (#25813, @rrati)
• promote sourceRange into service spec (#25826, @freehan)
• kube-controller-manager: Add configure-cloud-routes option (#25614, @justinsb)
• kubelet: reading cloudinfo from cadvisor (#21373, @enoodle)
• Disable cAdvisor event storage by default (#24771, @timstclair)
• Remove docker-multinode (#26031, @luxas)
• nodecontroller: Fix log message on successful update (#26207, @zmerlynn)
• remove deprecated generated typed clients (#26336, @caesarxuchao)
• Kubenet host-port support through iptables (#25604, @freehan)
• Add metrics support for a GCE PD, EC2 EBS & Azure File volumes (#25852, @vishh)
• Bump cAdvisor (and dependencies) godeps version (#25914, @timstclair)
• Add RBAC authorization API group and authorizer (#25634, @ericchiang)
• Add Seccomp to Annotations (#25324, @jfrazelle)
• AWS: Fix long-standing bug in stringSetToPointers (#26331, @therc)
• Add dnsmasq as a DNS cache in kube-dns pod (#26114, @ArtfulCoder)
• routecontroller: Add wait.NonSlidingUntil, use it (#26301, @zmerlynn)
• Attempt 2: Bump GCE containerVM to container-v1-3-v20160517 (Docker 1.11.1) again. (#26001, @dchen1107)
• Downward API implementation for resources limits and requests (#24179, @aveshagarwal)
• Replace containervm with GCI as default master image for GCE clusters (#26197, @wonderfly)
• Add a 'kubectl clusterinfo dump' option (#20672, @brendandburns)
• Fixing heapster memory requirements. (#26109, @Q-Lee)
• Handle federated service name lookups in kube-dns. (#25727, @madhusudancs)
• Support sort-by timestamp in kubectl get (#25600, @janetkuo)
• vSphere Volume Plugin Implementation (#24947, @abithap)
• ResourceQuota controller uses rate limiter to prevent hot-loops in error situations (#25748, @derekwaynecarr)
• Fix hyperkube flag parsing (#25512, @colhom)
• Add a kubectl create secret tls command (#24719, @bprashanth)
• Add node problem detector as an addon pod. (#25986, @Random-Liu)
• Handle cAdvisor partial failures (#25933, @timstclair)
• Use SkyDNS as a library for a more integrated kube DNS (#23930, @ArtfulCoder)
• Introduce node memory pressure condition to scheduler (#25531, @ingvagabund)
• Fix detection of docker cgroup on RHEL (#25907, @ncdc)
• Add support for limiting grace period during soft eviction (#25772, @derekwaynecarr)
• Use protobufs by default to communicate with apiserver (still store JSONs in etcd) (#25738, @wojtek-t)
• Add NetworkPolicy API Resource (#25638, @caseydavenport)
• Only expose top N images in NodeStatus (#25328, @resouer)
• Extend secrets volumes with path control (#25285, @ingvagabund)
• Implement OIDC client AuthProvider (#25270, @bobbyrullo)
• Make addon-manager cross-platform and use it with hyperkube (#25631, @luxas)
• kubelet: Optionally, have kubelet exit if lock file contention is observed, using --exit-on-lock-contention flag (#25596,@derekparker)
• Bump up glbc version to 0.6.2 (#25446, @bprashanth)
• Add 'kubectl set image' (#25509, @janetkuo)
• NodeController doesn't evict Pods if no Nodes are Ready (#25571, @gmarek)
• Added enforcing of setting nodes numbers for cluster autoscaler. (#25734, @jszczepkowski)
• systemd node spec proposal (#17688, @derekwaynecarr)
• Bump GCE ContainerVM to container-v1-3-v20160517 (Docker 1.11.1) (#25843, @zmerlynn)
• AWS: Move enforcement of attached AWS device limit from kubelet to scheduler (#23254, @jsafrane)
• Refactor persistent volume controller (#24331, @jsafrane)
• Add support for running GCI on the GCE cloud provider (#25425, @andyzheng0831)
• Implement taints and tolerations (#24134, @kevin-wangzefeng)
• GCI: Ensure that the right version of kubelet is used (#25504, @andyzheng0831)
• Add init containers to pods (#23567, @smarterclayton)
• GCI: Fix the condition for using the default image (#25763, @andyzheng0831)

下載

• kubernetes.tar.gz
• Source code (zip)
• Source code (tar.gz)