[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-09-03。"],[],[],null,["# Ensure workloads are disruption-ready\n\n[Autopilot](/kubernetes-engine/docs/concepts/autopilot-overview) [Standard](/kubernetes-engine/docs/concepts/choose-cluster-mode)\n\n*** ** * ** ***\n\nApplications running in Google Kubernetes Engine (GKE) clusters must be prepared for\ndisruptions such as node upgrades and other maintenance events. Stateful\napplications, which often need time to cleanly stop I/O and unmount from\nstorage, are especially vulnerable to disruptions. You can use Kubernetes\nfeatures such as [Pod Disruption\nBudgets](https://kubernetes.io/docs/tasks/run-application/configure-pdb/)\n(PDBs) and [Readiness\nProbes](https://cloud.google.com/blog/products/containers-kubernetes/kubernetes-best-practices-setting-up-health-checks-with-readiness-and-liveness-probes)\nto help keep applications available during upgrades.\n\nGKE monitors your clusters and uses the [Recommender\nservice](/recommender/docs/overview) to deliver guidance for how you can\noptimize your usage of the platform. GKE detects opportunities to\nprepare your workloads for disruption and provides guidance on how to update\nyour PDBs or readiness probes to maximize the resiliency of your workloads to\ndisruption. For example, if a StatefulSet is not protected by a PDB, your\ncluster might remove all Pods at once during a node upgrade. To avoid this,\nGKE delivers guidance to create a PDB so that most Pods can stay\nrunning during an upgrade.\n\nTo see the specific conditions where GKE delivers\ndisruption-related guidance, see [When GKE identifies workloads with vulnerability to disruption](#conditions-for-identification).\n\nTo learn more about how to manage insights and recommendations from Recommenders,\nsee [Optimize your usage of GKE with insights and recommendations](/kubernetes-engine/docs/how-to/optimize-with-recommenders).\n\nIdentify workloads with vulnerability to disruption\n---------------------------------------------------\n\nGKE generates insights identifying your cluster's\ndisruption-vulnerable workloads. To get these insights, follow the instructions\nto [view insights and\nrecommendations](/kubernetes-engine/docs/how-to/optimize-with-recommenders#view-insights-recs)\nusing the Google Cloud CLI, or the Recommender API. Use the subtypes listed\nin the following section to filter for specific insights. These insights are not\navailable in the Google Cloud console.\n\nWhen GKE identifies workloads with vulnerability to disruption\n--------------------------------------------------------------\n\nSee the following table for scenarios where GKE delivers an\ninsight and recommendation, and the relevant subtype:\n\nImplement the guidance to improve disruption readiness\n------------------------------------------------------\n\nIf you've received insights and recommendations for workloads in your cluster\nand you want to improve their disruption readiness, implement the instructions\ndescribed in the recommendation and the action for that insight subtype, as seen\nin the previous section.\n\nRecommendations are assessed once daily, so it may take up to 24 hours for them\nto resolve after changes have been implemented.\n\nIf you do not want to implement the recommendation, you can [dismiss\nit](/kubernetes-engine/docs/how-to/optimize-with-recommenders#dismiss-recommendation).\n\nWhat's next\n-----------\n\n- To learn more about ensuring reliability and uptime for your GKE cluster, see [GKE Day 2 Operations Best\n Practices](https://cloud.google.com/blog/products/containers-kubernetes/ensuring-reliability-and-uptime-for-your-gke-cluster).\n- To learn more about possible disruptions to Pods in Kubernetes, see [Disruptions](https://kubernetes.io/docs/concepts/workloads/pods/disruptions/)."]]