앱 프로필에서 모든 요청을 단일 클러스터로 라우팅하고 클러스터가 응답하지 않으면 수동 장애 조치를 사용합니다. 클러스터가 응답하지 않는지 확인하기 위해 사용할 수 있는 기준 예시는 수동 장애 조치를 참조하세요. 장애 조치 여부를 결정하기 전에 인스턴스의 복제 지연 시간을 확인합니다. 자세한 내용은 복제 차트를 참조하세요.
수동 장애 조치를 수행하려면 앱 프로필이 응답하지 않는 클러스터 대신 응답하는 클러스터로 요청을 라우팅하도록 업데이트합니다.
CLUSTER_ID: 모든 요청이 라우팅되어야 하는 클러스터 ID입니다. 이 플래그는 단일 클러스터 라우팅을 사용 설정합니다.
오류 메시지가 나타나면 오류 메시지의 모든 경고를 자세히 검토합니다. 오류를 재정의하려면 --force 플래그를 사용하여 명령어를 다시 실행합니다.
앱 프로필을 업데이트하면 앱 프로필을 사용하는 모든 애플리케이션이 선택한 정상 클러스터로 모든 요청을 전달하기 시작합니다. 비정상 클러스터는 CPU를 계속 사용하여 복제 및 기타 유지보수 작업을 처리합니다.
비정상 클러스터가 복구된 후에는 모든 요청을 복구된 클러스터로 라우팅하도록 동일한 단계에 따라 앱 프로필을 업데이트할 수 있습니다.
자동 장애 조치 수행
Bigtable을 사용하면 자동 장애 조치가 자동으로 수행됩니다. 앱 프로필이 멀티 클러스터 라우팅을 사용하고 애플리케이션 서버에 가장 가까운 클러스터가 비정상 상태가 될 경우 별다른 조치를 취할 필요가 없습니다.
Bigtable은 클러스터가 잠시 비정상 상태가 되더라도 자동으로 장애 조치를 수행하며, 비정상 클러스터가 복구될 때까지 가장 가까이 있는 정상 클러스터로 요청을 처리합니다.
일정 기간 동안 자동으로 경로가 변경된 요청 수를 보려면 Google Cloud 콘솔에서 자동 장애 조치 그래프를 봅니다. 인스턴스 목록을 열고 인스턴스 이름을 클릭한 후 모니터링을 클릭하면 됩니다.
[[["이해하기 쉬움","easyToUnderstand","thumb-up"],["문제가 해결됨","solvedMyProblem","thumb-up"],["기타","otherUp","thumb-up"]],[["이해하기 어려움","hardToUnderstand","thumb-down"],["잘못된 정보 또는 샘플 코드","incorrectInformationOrSampleCode","thumb-down"],["필요한 정보/샘플이 없음","missingTheInformationSamplesINeed","thumb-down"],["번역 문제","translationIssue","thumb-down"],["기타","otherDown","thumb-down"]],["최종 업데이트: 2025-08-18(UTC)"],[[["\u003cp\u003eBigtable replication enables traffic failover to another cluster if one becomes unresponsive, either manually or automatically.\u003c/p\u003e\n"],["\u003cp\u003eManual failovers are required when an app profile routes all requests to a single cluster, and involve updating the app profile to direct traffic to a responsive cluster.\u003c/p\u003e\n"],["\u003cp\u003eAutomatic failovers occur when an app profile uses multi-cluster routing, where Bigtable automatically reroutes traffic to the nearest healthy cluster.\u003c/p\u003e\n"],["\u003cp\u003eTo perform a manual failover, one must use the Google Cloud console or the gcloud command-line tool to update the app profile and select a healthy cluster.\u003c/p\u003e\n"],["\u003cp\u003eThe number of requests automatically rerouted can be monitored using the "Automatic Failovers" graph in the Google Cloud console under instance monitoring.\u003c/p\u003e\n"]]],[],null,["# Manage failovers\n================\n\n\nIf a Bigtable cluster becomes unresponsive, replication makes it possible for incoming\ntraffic to fail over to another cluster in the same instance. Failovers can be either manual or\nautomatic, depending on the [app profile](/bigtable/docs/app-profiles) an application\nis using and how the app profile is configured.\n\nThis page describes the steps for performing a failover between clusters.\n\n- If an app profile routes all requests to a single cluster, you can [perform a\n manual failover](#manual).\n- If an app profile uses multi-cluster routing, [failovers are\n automatic](#automatic) and you don't need to take any action.\n\n\nBefore you read this page, you should be familiar with the\n[overview of Bigtable replication](/bigtable/docs/replication-overview).\nYou should also be familiar with the\n[routing options](/bigtable/docs/routing) that are available for\nBigtable.\n\nPerform a manual failover\n-------------------------\n\nUse a manual failover if an app profile routes all requests to a single cluster\nand that cluster becomes unresponsive. For examples of the criteria you could\nuse to determine that a cluster is unresponsive, see\n[Manual failovers](/bigtable/docs/failovers#manual). Check your instance's replication latency\nbefore you decide to fail over. For more information, see [Charts for\nreplication](/bigtable/docs/monitoring-instance).\n\nTo perform a manual failover, update your app profile so that it routes requests\nto a responsive cluster instead of the unresponsive cluster: \n\n### Console\n\n1.\n Open the list of Bigtable instances in the Google Cloud console.\n\n\n [Open the instance list](https://console.cloud.google.com/bigtable/instances)\n2. In the **Application profiles** column, click the app profile that is\n routing traffic to the unresponsive cluster.\n\n If you don't see the app profile you want to edit, you can view a\n complete list by clicking the name of the instance, then clicking\n **Application profiles** in the left pane.\n3. Under **Cluster routing**, select a responsive cluster in your instance.\n\n | **Important:** If **Any cluster (multi-cluster)** is selected, the app profile handles failovers automatically. Click **Cancel** to exit without saving.\n4. Click **Save**. A confirmation dialog appears.\n\n5. Carefully review the warnings in the confirmation dialog, then follow\n the instructions in the dialog and click **Continue**.\n\n### gcloud\n\n1.\n If you don't know the instance ID, use the\n [`bigtable instances\n list` command](/sdk/gcloud/reference/bigtable/instances/list) to view a list of your project's instances:\n\n gcloud bigtable instances list\n\n2.\n If you don't know the instance's cluster IDs, use the\n [`bigtable clusters list`\n command](/sdk/gcloud/reference/bigtable/clusters/list) to view a list of clusters in the instance:\n\n gcloud bigtable clusters list --instances=\u003cvar translate=\"no\"\u003eINSTANCE_ID\u003c/var\u003e\n\n\n Replace \u003cvar translate=\"no\"\u003eINSTANCE_ID\u003c/var\u003e with the permanent identifier for the instance.\n3.\n If you don't know the app profile's ID, use the\n [`bigtable app-profiles\n list` command](/sdk/gcloud/reference/bigtable/app-profiles/list) to view a list of the instance's app profiles:\n\n gcloud bigtable app-profiles list --instance=\u003cvar translate=\"no\"\u003eINSTANCE_ID\u003c/var\u003e\n\n\n Replace \u003cvar translate=\"no\"\u003eINSTANCE_ID\u003c/var\u003e with the permanent identifier for the instance.\n4. Use the [`bigtable app-profiles update` command](/sdk/gcloud/reference/bigtable/app-profiles/update) to\n change the cluster that the app profile uses:\n\n gcloud bigtable app-profiles update APP_PROFILE_ID \\\n --instance=INSTANCE_ID \\\n --route-to=CLUSTER_ID\n\n\n Provide the following:\n - `APP_PROFILE_ID`: The permanent identifier for the app profile.\n - `INSTANCE_ID`: The permanent identifier for the instance.\n - `CLUSTER_ID`: The cluster ID that all requests should be routed to. This flag enables [single-cluster routing](/bigtable/docs/routing#single-cluster).\n\n If you receive an error message, carefully review any warnings in the\n error message. If you want to override the error, run the command again\n with the `--force` flag.\n\nShortly after you update the app profile, any applications that use the app\nprofile will start to direct all of their requests to the healthy cluster that\nyou selected. The unhealthy cluster will continue using CPU to handle\nreplication and other maintenance tasks.\n\nAfter the unhealthy cluster recovers, you can follow the same steps to update\nyour app profile so that it routes all requests to the recovered cluster.\n\nPerform an automatic failover\n-----------------------------\n\nWith Bigtable, automatic failovers truly are automatic. If an app\nprofile uses multi-cluster routing, and the nearest cluster to the application\nserver becomes unhealthy, you don't need to take any action.\nBigtable automatically fails over, even if the cluster is only\nbriefly unhealthy, and uses the nearest healthy cluster to handle requests until\nthe unhealthy cluster has recovered.\n\nTo view the number of requests that were automatically rerouted over a given\nperiod of time, look at the **Automatic Failovers** graph in the\nGoogle Cloud console: open the [list of instances](https://console.cloud.google.com/bigtable/instances), click\nthe instance name, then click **Monitoring**.\n\nWhat's next\n-----------\n\nLearn how to [monitor a Bigtable instance](/bigtable/docs/monitoring-instance)."]]
