에어 갭이 적용된 Google Distributed Cloud 1.9.3이 출시되었습니다.
Google Distributed Cloud(에어 갭 적용형)의 기능에 대해 알아보려면 제품 개요를 참고하세요.
Google Distributed Cloud 오프라인 1.9.3 감사 로깅 (AL) 작동 가능 구성요소는 AuditLoggingTargets를 개선합니다. 이제 AuditLoggingTarget CR로 생성된 Kubernetes 객체가 업데이트되거나 삭제되면 자체 복구됩니다.
최신 보안 패치와 중요 업데이트를 적용하기 위해 Google Distributed Cloud 버전을 1.14.4-gke.4로 업데이트했습니다.
Google Distributed Cloud 에어 갭 1.9.3 사용자 인터페이스 (UI) 구성요소에서 UI를 사용하여 VM을 만들면 스냅샷에서 복원된 디스크가 부팅 디스크로 사용될 수 있습니다.
Google Distributed Cloud 에어 갭 1.9.3 ID 및 액세스 관리 (IAM) 구성요소에서 업그레이드에 사용 설정된 사전 정의된 역할은 더 이상 수동 변경의 적용을 받지 않습니다. 사전 정의된 역할 매니페스트는 수동 변경사항을 재정의합니다.
Google Distributed Cloud 에어 갭 1.9.3에서는 방화벽 (FW) 구성요소에서 관리자 사용자 인증 정보를 순환하는 동안 발생하는 방화벽 관리자 계정 잠금 문제를 해결합니다.
Google Distributed Cloud 에어 갭 1.9.3에서는 LOG 구성요소에서 Splunk로의 운영 로그 내보내기를 수정합니다.
Google Distributed Cloud 오프라인 1.9.3에는 VM 관리자의 역할 기반 액세스 제어 (RBAC) 및 스키마 설정으로 인해 사용자가 VM 백업 및 복원 프로세스를 시작할 수 없는 알려진 문제가 있습니다.
Google Distributed Cloud 에어 갭 1.9.3에는 kubevm-gpu-driver-daemonset 포드가 CrashLoopBackOff 상태에 있기 때문에 gpu-org-system-cluster을 1.9.1에서 1.9.2로 업그레이드하는 동안 vm-runtime 부가기능이 멈추는 알려진 문제가 있습니다.
Google Distributed Cloud 오프라인 1.9.3에서는 UNET 구성요소의 내부 부하 분산기 (ILB) 서비스 문제가 해결되었습니다. 1.9.0~1.9.2 버전에는 내부 부하 분산기 (ILB) 서비스에 내부 IP 대신 외부 IP가 할당되는 버그가 포함되어 있습니다. 영향은 ILB 서비스가 이 풀에서 주소를 가져오므로 외부 부하 분산기 IP 풀이 더 빨리 사용된다는 것입니다. 하지만 ILB 서비스에 할당된 IP는 조직 외부에서 공지되지 않았으므로 서비스는 조직 내부에 유지되었습니다. 이 버그는 ILB 서비스에 내부 IP가 할당되도록 1.9.3에서 수정되었습니다.
Google Distributed Cloud 오프라인 1.9.3에는 사용자 클러스터가 제때 준비되지 않는 알려진 문제가 있습니다.
Google Distributed Cloud 에어갭 1.9.3에는 애드온 설치가 실패하는 알려진 문제가 있습니다.
Google Distributed Cloud 에어 갭 적용형 1.9.3에는 OrganizationUpgrade 상태가 업데이트되지 않는 알려진 문제가 있습니다.
Google Distributed Cloud 오프라인 1.9.3에는 사용자 클러스터 업그레이드가 웹훅을 호출하지 못하는 알려진 문제가 있습니다.
Google Distributed Cloud 오프라인 1.9.3에는 로그에 Fleet admin controller manager stopped: failed to wait for auditloggingtarget caches to sync: timed out waiting for cache to be synced 오류가 표시되어 플릿 관리자 컨트롤러가 비정상 종료 루프에 갇히는 알려진 문제가 있습니다.
Google Distributed Cloud 에어 갭 1.9.3에는 시스템 클러스터가 제때 준비되지 않는 알려진 문제가 있습니다.
Google Distributed Cloud 오프라인 1.9.3에서는 루트 관리자 클러스터의 AddOn 선택기 라벨을 설정할 수 없습니다.
Google Distributed Cloud 에어 갭 1.9.3에는 GPU와 VM 유형의 호환되지 않는 결합을 선택할 수 있는 UI의 알려진 문제가 있습니다.
Google Distributed Cloud 오프라인 1.9.3에는 메모리가 32GB보다 큰 VM에 잘못된 QEMU 오버헤드 계산으로 인해 메모리 재정의가 필요한 알려진 문제가 있습니다.
Google Distributed Cloud air-gapped 1.9.3에는 kube-state-metrics 배포가 비정상 종료되는 알려진 문제가 있습니다.
Google Distributed Cloud 에어 갭 1.9.3에는 조직 시스템 클러스터의 알림이 티켓팅 시스템에 도달하지 않는 알려진 문제가 있습니다.
[[["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: 2025-09-04(UTC)"],[[["\u003cp\u003eGoogle Distributed Cloud air-gapped 1.9.3 is now available, introducing enhancements such as self-healing Kubernetes objects for AuditLoggingTargets and a UI message indicating the maximum worker node amount during cluster creation.\u003c/p\u003e\n"],["\u003cp\u003eThis version updates the Google Distributed Cloud version to 1.14.4-gke.4 to incorporate the latest security patches and critical updates, while also addressing 46 container image security vulnerabilities.\u003c/p\u003e\n"],["\u003cp\u003eThe new user interface allows disks restored from snapshots to be used as boot disks for VM creation, and it is no longer possible to make manual changes to predefined roles as they will be overriden.\u003c/p\u003e\n"],["\u003cp\u003eSeveral issues have been resolved, including a firewall admin account lock-out problem during credential rotation, the operational log export issue to Splunk, and a bug causing internal load balancer services to be assigned external IPs, along with many known issues that remain.\u003c/p\u003e\n"],["\u003cp\u003eSome of the many known issues include: issues with the VM manager, add-ons, upgrades, the fleet admin controller, system clusters, setting AddOn selector labels, incompatible coupling of GPU to VM types, memory overrides on VMs with 32GB or more, \u003ccode\u003ekube-state-metrics\u003c/code\u003e deployments, and alerts in the organization system clusters not reaching the ticketing system.\u003c/p\u003e\n"]]],[],null,["# Google Distributed Cloud air-gapped 1.9.3 release notes\n\n\u003cbr /\u003e\n\nApril 28, 2023 \\[GDC 1.9.3\\]\n----------------------------\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 is now released.\n\nSee the [product overview](/distributed-cloud/hosted/docs/latest/gdch/overview) to learn about the features of Google Distributed Cloud air-gapped.\n\n*** ** * ** ***\n\n\nThe Google Distributed Cloud air-gapped 1.9.3 audit logging (AL) operable component introduces an enhancement for AuditLoggingTargets. Kubernetes objects created by an AuditLoggingTarget CR are now self-healed if they are updated or deleted.\n\n*** ** * ** ***\n\n\nUpdated Google Distributed Cloud version to 1.14.4-gke.4 to apply the latest security patches and important updates.\n\nSee [the Google Distributed Cloud release notes](https://cloud.google.com/anthos/clusters/docs/bare-metal/latest/release-notes-ver) for the latest information.\n\n*** ** * ** ***\n\n\nThe Google Distributed Cloud air-gapped 1.9.3 user interface (UI) now includes a message to inform you about the maximum worker node amount with the current control plane setting and pod CIDR option when you create a cluster.\n\n*** ** * ** ***\n\n\nThe following container image security vulnerabilities are fixed:\n\n- [CVE-2021-46848](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-46848)\n- [CVE-2022-1354](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-1354)\n- [CVE-2022-1355](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-1355)\n- [CVE-2022-2056](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-2056)\n- [CVE-2022-2057](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-2057)\n- [CVE-2022-2058](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-2058)\n- [CVE-2022-2097](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-2097)\n- [CVE-2022-23521](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-23521)\n- [CVE-2022-24765](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-24765)\n- [CVE-2022-2867](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-2867)\n- [CVE-2022-2868](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-2868)\n- [CVE-2022-2869](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-2869)\n- [CVE-2022-29187](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-29187)\n- [CVE-2022-32221](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-32221)\n- [CVE-2022-34526](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-34526)\n- [CVE-2022-3570](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-3570)\n- [CVE-2022-3597](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-3597)\n- [CVE-2022-3598](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-3598)\n- [CVE-2022-3599](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-3599)\n- [CVE-2022-3626](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-3626)\n- [CVE-2022-3627](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-3627)\n- [CVE-2022-3970](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-3970)\n- [CVE-2022-39253](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-39253)\n- [CVE-2022-39260](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-39260)\n- [CVE-2022-41903](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-41903)\n- [CVE-2022-42898](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-42898)\n- [CVE-2022-4304](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-4304)\n- [CVE-2022-4450](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-4450)\n- [CVE-2022-4645](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-4645)\n- [CVE-2022-48281](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-48281)\n- [CVE-2023-0215](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0215)\n- [CVE-2023-0286](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0286)\n- [CVE-2023-0361](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0361)\n- [CVE-2023-0795](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0795)\n- [CVE-2023-0796](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0796)\n- [CVE-2023-0797](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0797)\n- [CVE-2023-0798](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0798)\n- [CVE-2023-0799](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0799)\n- [CVE-2023-0800](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0800)\n- [CVE-2023-0801](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0801)\n- [CVE-2023-0802](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0802)\n- [CVE-2023-0803](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0803)\n- [CVE-2023-0804](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-0804)\n- [CVE-2023-26604](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2023-26604)\n\n*** ** * ** ***\n\n\nIn the Google Distributed Cloud air-gapped 1.9.3 user interface (UI) component, VM creation using UI allows disks restored from a snapshot to serve as boot disks.\n\n*** ** * ** ***\n\n\nIn the Google Distributed Cloud air-gapped 1.9.3 identity and access management (IAM) component, predefined roles enabled for upgrades are no longer subject to manual changes. Predefined role manifests override any manual changes.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 resolves the Firewall admin account lock out issue while rotating the admin credentials in the firewall (FW) component.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 fixes export of operational logs to Splunk in the LOG component.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where role-based access control (RBAC) and schema settings in the VM manager is stopping users from starting VM backup and restore processes.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where the `vm-runtime` addon is stuck during the upgrade of the `gpu-org-system-cluster` from 1.9.1 to 1.9.2 because the `kubevm-gpu-driver-daemonset` pods are in the `CrashLoopBackOff` state.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 resolves an internal load balancer (ILB) services issue in the UNET component. Releases 1.9.0 - 1.9.2 contained a bug where internal load balancer (ILB) services were assigned an external IP instead of an internal IP. The impact is that the external load balancer IP pool is used more quickly as ILB services take addresses from this pool. However, the IPs assigned to ILB services were not advertised outside of the org, so the service remained internal to the org. This bug is fixed in 1.9.3 so that ILB services are assigned internal IPs.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where a user cluster does not become ready in time.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where an add-on installation fails.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where an `OrganizationUpgrade` status does not get updated.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where a user cluster upgrade fails to call webhooks.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where a fleet admin controller gets stuck in a crash loop with the `Fleet admin controller manager stopped: failed to wait for auditloggingtarget caches to sync: timed out waiting for cache to be synced` error in the logs.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where a system cluster does not become ready in time.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 is unable to set AddOn selector labels for the root admin cluster.\n\n*** ** * ** ***\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue in the UI that lets you select an incompatible coupling of GPU to VM type.\n\n*** ** * ** ***\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where VMs with memory greater than 32 GB require a memory override due to an incorrect QEMU overhead calculation.\n\n*** ** * ** ***\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known\nissue where the `kube-state-metrics` deployment crash loops.\n\n*** ** * ** ***\n\n\nGoogle Distributed Cloud air-gapped 1.9.3 has a known issue where alerts in organization system clusters don't reach the ticketing system."]]