generic::invalid_argument: Accelerator type v2-512 as preemptible (false) and
reserved (false) is not available in zone us-central1-a, please contact support.
generic::permission_denied: Cloud TPU got permissions denied when trying to
access the customer project. Make sure that the IAM account
'service-[project number]@cloud-tpu.iam.gserviceaccount.com' has the 'Cloud
TPU API Service Agent' role by following https://cloud.google.com/iam/docs/manage-access-service-accounts
解決策
このエラーは、ユーザーが IAM 認可なしでプロジェクト内のノードの作成または一覧取得を試みた場合に発生します。この問題の原因としては、Cloud TPU API サービス アカウントにプロジェクトに必要なロールが付与されていないことが考えられます。アクセス アカウントを管理するドキュメントでは、アクセスを管理する方法の概要を説明しています。単一のロールの付与または取り消しの手順に沿って、アカウントに Cloud TPU API サービス エージェントの「service-PROJECT_NUMBER@cloud-tpu.iam.gserviceaccount.com」ロールを付与します。（PROJECT_NUMBER はプロジェクト番号に置き換えます。プロジェクト番号は Google Cloud コンソールのプロジェクト設定で確認できます）。サービス エージェントの詳細については、サービス エージェントに関するドキュメントをご覧ください。
上限を超えている
エラー メッセージ
You have reached XXXX limit. Please request an increase for the 'YYYY' quota for
Compute Engine API by following https://cloud.google.com/docs/quotas/view-manage#requesting_higher_quota.
[[["わかりやすい","easyToUnderstand","thumb-up"],["問題の解決に役立った","solvedMyProblem","thumb-up"],["その他","otherUp","thumb-up"]],[["わかりにくい","hardToUnderstand","thumb-down"],["情報またはサンプルコードが不正確","incorrectInformationOrSampleCode","thumb-down"],["必要な情報 / サンプルがない","missingTheInformationSamplesINeed","thumb-down"],["翻訳に関する問題","translationIssue","thumb-down"],["その他","otherDown","thumb-down"]],["最終更新日 2025-08-18 UTC。"],[],[],null,["# Cloud TPU Error Glossary\n========================\n\nThis document provides a glossary of common errors with solutions from the Cloud TPU service.\n\n*** ** * ** ***\n\nInvalid Accelerator Type\n------------------------\n\n### **Error Message**\n\n```\ngeneric::invalid_argument: Accelerator type v2-512 as preemptible (false) and\nreserved (false) is not available in zone us-central1-a, please contact support.\n```\n\n### **Solution**\n\nAn invalid parameter has been provided to the create command. The availability\nof an accelerator in a zone depends on the following parameters: the type, the\n`preemptible` flag, the `spot` flag, and the `reserved` flag. The `preemptible`, `spot`,\nand `reserved` flags can be changed by including or excluding them in the create\ncommand.\n\nA TPU created with the `reserved` flag will use reserved\ncapacity. Including the `spot` or `preemptible` flags will allow the TPU to be preempted by\nhigher priority TPUs. If none of these flags are provided, the TPU will be on-demand. It is\nnot a valid configuration to enable more than one of these flags. See the\n[create command documentation](/sdk/gcloud/reference/compute/tpus/create) for more information.\n\nThe accelerator types available in each zone can be found in the [TPU regions and zones documentation](/tpu/docs/regions-zones) or they can be\nqueried using the [accelerator-types list command](/sdk/gcloud/reference/compute/tpus/accelerator-types/list). Change the create command to use one\nof these accelerator types and try again or contact support if the problem persists.\n\nNetwork Not Found\n-----------------\n\n### **Error Message**\n\n```\n Cloud TPU received a bad request. The field \"Network\" cannot be \"xxxx\":\n requested resource not found\n```\n\n### **Solution**\n\nThe Network xxxx was not found. Ensure that the Network was created\nand set up properly. See [Create and manage VPC networks](/vpc/docs/create-modify-vpc-networks) for more information.\n\nService Account Permission Denied\n---------------------------------\n\n### **Error Message**\n\n```\n generic::permission_denied: Cloud TPU got permissions denied when trying to\n access the customer project. Make sure that the IAM account\n 'service-[project number]@cloud-tpu.iam.gserviceaccount.com' has the 'Cloud\n TPU API Service Agent' role by following https://cloud.google.com/iam/docs/manage-access-service-accounts\n```\n\n### **Solution**\n\nThis error occurs when a user attempts to create or list nodes in a project\nwithout IAM authorization. A likely cause of this issue is that the Cloud TPU API service\naccount does not have the required role for the project. The [Manage access accounts documentation](/iam/docs/manage-access-service-accounts)\ngives an overview of how to manage access. Follow the [Grant or revoke a single role](/iam/docs/manage-access-service-accounts#single-role) steps\nand give the account 'service-\u003cvar translate=\"no\"\u003ePROJECT_NUMBER\u003c/var\u003e@cloud-tpu.iam.gserviceaccount.com' the role of 'Cloud TPU API Service Agent' (be sure to replace \u003cvar translate=\"no\"\u003ePROJECT_NUMBER\u003c/var\u003e\nwith your project number, which can be found in the project settings in the Google Cloud console). For more\ninformation on service agents, see the [Service agents documentation](/iam/docs/service-agents).\n\nQuota Exceeded\n--------------\n\n### **Error Message**\n\n```\nYou have reached XXXX limit. Please request an increase for the 'YYYY' quota for\nCompute Engine API by following https://cloud.google.com/docs/quotas/view-manage#requesting_higher_quota.\n```\n\n### **Solution**\n\nYour project has reached a quota limit. To learn more about working with quotas, see the [Cloud Quotas documentation](/docs/quotas/overview). This shouldn't be confused with the [TPU quota](/tpu/docs/quota), which governs the usage of TPU pods.\n\nYou may request an increase to the appropriate limit by following the steps listed at [Request a quota adjustment](/docs/quotas/view-manage#requesting_higher_quota). On the quotas page, you may search for the quota specified by the 'YYYY' part of this message. Some quotas are split across different regions or services. The error message will indicate which one needs to be increased.\n\nThe 'XXXX' and 'YYYY' parts of the message may be one of the following:\n\n- HEALTH_CHECKS - 'Health checks' quota\n- FIREWALLS - 'Firewall rules' quota\n- NETWORK_ENDPOINT_GROUPS - 'Network endpoint groups' quota for this region\n- READ_REQUESTS - 'Read requests per minute' quota for the Compute Engine API service\n- OPERATION_READ_REQUESTS - 'Operation read requests per minute' quota\n\nThis request is typically processed within 2-3 business days. If the request is urgent, reach out to a customer engineer or technical account manager."]]
