如果您已升級執行 Oracle VM (OVM) 的 Bare Metal 解決方案伺服器，但無法連線至 OVM Manager，則可能是因為這個問題。升級 Bare Metal 解決方案伺服器的韌體會變更其 SMBIOS UUID。OVM 會使用這個 UUID 向 OVM Manager 識別自己。因此，如果變更 UUID，就可能導致兩者之間的通訊發生問題。為避免發生這種情況，請在升級 Bare Metal 解決方案伺服器的韌體前，先套用 Oracle Doc ID 1534416.1 所述的因應措施。如需這項解決方法的相關協助，請與客戶服務團隊聯絡。
[[["容易理解","easyToUnderstand","thumb-up"],["確實解決了我的問題","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["難以理解","hardToUnderstand","thumb-down"],["資訊或程式碼範例有誤","incorrectInformationOrSampleCode","thumb-down"],["缺少我需要的資訊/範例","missingTheInformationSamplesINeed","thumb-down"],["翻譯問題","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["上次更新時間：2025-09-04 (世界標準時間)。"],[[["\u003cp\u003eBare Metal Solution does not support modifications to BIOS settings, including disabling C-states and P-states at the BIOS level, but OS-level controls can be used as a workaround.\u003c/p\u003e\n"],["\u003cp\u003eVPC Service Controls parameters are not supported in Bare Metal Solution regional extensions, although the Bare Metal Solution API can be added to a secure perimeter.\u003c/p\u003e\n"],["\u003cp\u003eCustom autonomous system numbers (ASN) are not supported with Bare Metal Solution and Partner Interconnect, with the client network ASN fixed at \u003ccode\u003e65500\u003c/code\u003e and Partner Interconnect requiring a local ASN of \u003ccode\u003e16550\u003c/code\u003e.\u003c/p\u003e\n"],["\u003cp\u003eThe maximum number of LUNs that can be attached to a Bare Metal Solution server is 200, inclusive of the boot LUN.\u003c/p\u003e\n"],["\u003cp\u003eA known issue exists where using the \u003ccode\u003eethtool -d\u003c/code\u003e command can cause a buffer overflow, potentially leading to kernel panic, with specific workarounds provided for RHEL, SLES, and Oracle Enterprise Linux distributions, as well as OVM server.\u003c/p\u003e\n"]]],[],null,["# Known issues and limitations in Bare Metal Solution\n===================================================\n\nThis page describes known issues and limitations that you might run into when\nusing Bare Metal Solution.\n\nLimitations\n-----------\n\n### Modification to BIOS settings is not supported\n\nBare Metal Solution does not support modifications to BIOS settings, including\ndisabling C-states and P-states at the BIOS level.\n\nTo work around this issue, you can utilize the OS-level controls of C-states and\nP-states through your OS's power management functionality. To learn how to do\nthis, refer to your OS documentation.\n\n### Bare Metal Solution regional extension doesn't support VPC Service Controls parameters\n\n\u003cbr /\u003e\n\nConnecting a VPC with service controls enabled to your Bare Metal Solution\nenvironment doesn't uphold any service control guarantees.\n\nThe Bare Metal Solution API can be added to a secure perimeter. However, the\nVPC Service Controls perimeters don't extend to the Bare Metal Solution\nenvironment in the regional extensions. For more information, see\n[Supported products and limitations](/vpc-service-controls/docs/supported-products#table_baremetalsolution).\n\nIf you still want to use Bare Metal Solution with VPC Service Controls enabled,\ncontact [Customer Care](https://cloud.google.com/support-hub) to add your\nGoogle Cloud project to allowlist for using this feature.\n\n### Changing ASN is not supported\n\nBare Metal Solution and Partner Interconnect do not support custom\nautonomous system numbers (ASN). The Bare Metal Solution client network ASN is set to\n`65500`. For Partner Interconnect, all\nCloud Routers must have a local ASN of `16550`.\n\nWe recommend that you plan your deployments accordingly.\n\n### Maximum number of LUNs attached to a server\n\nYou can attach a maximum of 200 LUNs (including the boot LUN) to a\nBare Metal Solution server.\n\nKnown issues\n------------\n\n### Cold shutdown after the first reboot through OVM Manager\n\nFor servers provisioned with OVM 3.4.6, the first reboot initiated through the\nOVM Manager, either after the initial provisioning or re-imaging, results in a\ncold shutdown.\n\nTo learn how to power on a server, see\n[Operate your Bare Metal Solution server](/bare-metal/docs/bms-maintenance#restart_a_bare_metal_solution_server).\n\n### Server takes a long time to boot\n\nThe boot time may vary depending on the size of your server and the number of\nshared LUNs.\n\nThe larger the server, the longer it takes to boot.\n\nThe number of shared LUNs on a server also impacts its boot time. For instance,\na `o2-highmem-224-metal` server with \\~100 shared LUNs might take over an hour to boot\ncompared to a `o2-highmem-224-metal` server with fewer LUNs that would take \\~45 minutes.\nThis is because of the time required to run all the checks and is normal.\n\n### Buffer overflow when using ethtool with the debug flag\n\nA bug causing a buffer overflow when using `ethtool -d` was\n[fixed in Linux 5.8 kernel](https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id=da3287111ab43b32cec54d7ca6b48640f210a196).\nThis bug can cause a kernel panic and might impact your Bare Metal Solution server\ndepending on the OS and hypervisor that you are using.\n\nFollowing are our recommendations and workaround for this bug: \n\n### Red Hat Enterprise Linux (RHEL)\n\nIf you are using **RHEL 7.x**, then follow these guidelines:\n\n- Do not run `ethtool` with the debug flag: `ethtool -d`.\n- Prevent the `sosreport` utility from invoking `ethtool -d` by disabling\n the networking plugin.\n\n In the `/etc/sos/sos.conf` file, add the following lines: \n\n ```\n [plugins]\n disable = networking\n ```\n\nIf you are using **RHEL 8.x**, then follow these guidelines:\n\n- Update to kernel version **RHEL 8.3 (kernel 4.18.0-240)** or later.\n- For systems with kernel versions lower than **4.18.0-240** that\n can't be updated, prevent the `sosreport` utility from invoking\n `ethtool -d` by disabling the networking plugin.\n\n In the `/etc/sos/sos.conf` file, add the following lines: \n\n ```\n [report]\n skip-plugins = networking\n ```\n\nFor more information, see the\n[Red Hat Solution](https://access.redhat.com/solutions/5352911).\n\n### SUSE Linux Enterprise Server(SLES)\n\nThis bug was fixed in **SLES 15 SP4 (kernel version 5.14.21-150400.22.1)** .\nUpdate to **SLES 15 SP4 (kernel version 5.14.21-150400.22.1)** or later.\n\n### Oracle Enterprise Linux\n\nIf you are using **Oracle Linux 7.x**, then follow these guidelines:\n\n- Do not run `ethtool` with the debug flag: `ethtool -d`.\n- Prevent the `sosreport` utility from invoking `ethtool -d` by disabling\n the networking plugin.\n\n In the `/etc/sos/sos.conf` file, add the following lines: \n\n ```\n [plugins]\n disable = networking\n ```\n\nIf you are using **Oracle 8.x**, then follow these guidelines:\n\n- This bug was fixed in **OL8U7 (kernel-uek-5.15.0-3.60.5.1)** . You can update to the most recent kernel by following the [update instructions\n from Oracle](https://docs.oracle.com/en/operating-systems/oracle-linux/8/relnotes8.7/ol8-NewFeaturesandChanges.html#ol8-features-install).\n- If you can't update to kernel version **kernel-uek-5.15.0-3.60.5.1** ,\n prevent the `sosreport` utility from invoking `ethtool -d` by\n disabling the networking plugin.\n\n In the `/etc/sos/sos.conf` file, add the following lines: \n\n ```\n [report]\n skip-plugins = networking\n ```\n\n### Oracle VM Server\n\n- Do not run `ethtool` in `dom0` domain with the debug flag: `ethtool -d`.\n- Prevent the `sosreport` utility from invoking `ethtool -d` by disabling\n the networking plugin.\n\n In the `/etc/sos/sos.conf` file, add the following lines: \n\n ```\n [plugins]\n disable = networking\n ```\n\n### OVM server can't connect to OVM Manager after TS54 firmware upgrade\n\nIf you have upgraded your Bare Metal Solution server running Oracle VM (OVM) and it cannot connect to the OVM Manager, then this issue might be the reason. Upgrading firmware of a Bare Metal Solution server changes its SMBIOS UUID. The OVM uses this UUID to identify itself with the OVM Manager. Therefore, the change of UUID can cause issues with the communication between the two. To prevent this, before upgrading the firmware of your Bare Metal Solution server, apply the workaround described in [Oracle Doc ID 1534416.1](https://support.oracle.com/knowledge/Oracle%20Linux%20and%20Virtualization/1534416_1.html). For assistance in implementing this workaround, contact [Customer Care](https://cloud.google.com/support-hub)."]]
