[[["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):2025-08-28。"],[],[],null,["# Resolving Istiod scaling issues in Cloud Service Mesh\n=====================================================\n\n| **Note:** This guide only supports Cloud Service Mesh with Istio APIs and does not support Google Cloud APIs. For more information see, [Cloud Service Mesh overview](/service-mesh/docs/overview).\n\nThis section explains common Cloud Service Mesh problems and how to resolve\nthem. If you need additional assistance, see\n[Getting support](/service-mesh/v1.21/docs/getting-support).\n\nScaling factors\n---------------\n\n[Istiod](https://istio.io/v1.21/blog/2020/istiod/)\nsends configuration to each sidecar using a long-lived gRPC stream. It has\nseveral characteristics that affect scaling:\n\n- The size of the configuration to generate:\n - Total number of services/pods \\& Istio resources\n - For large scale, adjust settings for the [Sidecar](https://istio.io/v1.21/docs/reference/config/networking/sidecar/) to reduce the configuration size.\n- The rate of change in the environment:\n - When a new service is created or the Istio configuration is changed, full updates are sent to proxies.\n - Adding new endpoints is inexpensive for performance, because only incremental updates are sent.\n- The number of proxies for which configuration is generated:\n - Affected by the number of gateways and pods with a sidecar.\n\nScaling considerations\n----------------------\n\nIstiod scales well vertically (large requests) and horizontally (more\nreplicas). Ensure that your CPU limits are not too restrictive; if Istiod\nreaches the CPU limit, throttling may occur which will negatively affect\nconfiguration distribution. If you encounter performance issues, consider\nupgrading to the latest version of Cloud Service Mesh, as each version has\nperformance optimizations.\n\nUnbalanced load\n---------------\n\nLarge changes in cluster size might cause a temporarily unbalanced load, due to\nthe long-lived connections. This is mitigated by a 30 minute maximum connection\nage, which might result in error messages in Envoy, such as `gRPC config stream\nclosed: 13`, which allows the load to naturally rebalance.\n\nMitigate this issue by having multiple replicas of Istiod (the default is 2\nreplicas), and pre-scaling if you expect extreme cluster scale-ups."]]