“freshly provisioned worker boots with approved role-specific NIC and GPU settings” addresses bCM node category does not reflect the installed hardware role, but the current hardware or infrastructure issue is configuration presence does not demonstrate resilient management operation. The option “existing jobs continue within approved interruption thresholds and a standby assumes management” is more directly tied to the required outcome.

“the new active head retains node categories and provisioning state” addresses hA state replication was not configured or proven, but the current hardware or infrastructure issue is configuration presence does not demonstrate resilient management operation. The option “existing jobs continue within approved interruption thresholds and a standby assumes management” is more directly tied to the required outcome.

“managed nodes receive provisioning identities and register successfully” addresses management discovery paths were confused with provisioning networks, but the current hardware or infrastructure issue is configuration presence does not demonstrate resilient management operation. The option “existing jobs continue within approved interruption thresholds and a standby assumes management” is more directly tied to the required outcome.

Useful for split-brain prevention and HA arbitration are not validated, but it does not remove the specific condition: administrative access controls are insufficient for high-impact provisioning tasks.

“validate provisioning DHCP and node enrollment from the BCM internal network” addresses management discovery paths were confused with provisioning networks, but the current hardware or infrastructure issue is administrative access controls are insufficient for high-impact provisioning tasks. The option “apply least-privilege BCM roles and auditable changes to image assignments” is more directly tied to the required outcome.

“express required configuration through supported BCM managed category or image settings” addresses declarative node category settings are not reconciled with out-of-band manual edits, but the current hardware or infrastructure issue is administrative access controls are insufficient for high-impact provisioning tasks. The option “apply least-privilege BCM roles and auditable changes to image assignments” is more directly tied to the required outcome.

“managed nodes receive provisioning identities and register successfully” addresses management discovery paths were confused with provisioning networks, but the current hardware or infrastructure issue is administrative access controls are insufficient for high-impact provisioning tasks. The option “test unauthorized image edits are denied and valid modifications leave attributable audit trails” is more directly tied to the required outcome.

“cluster nodes register on the intended private and external addresses” addresses bCM provisioning cannot reliably distinguish external management from internal cluster communication, but the current hardware or infrastructure issue is administrative access controls are insufficient for high-impact provisioning tasks. The option “test unauthorized image edits are denied and valid modifications leave attributable audit trails” is more directly tied to the required outcome.

“the new active head retains node categories and provisioning state” addresses hA state replication was not configured or proven, but the current hardware or infrastructure issue is administrative access controls are insufficient for high-impact provisioning tasks. The option “test unauthorized image edits are denied and valid modifications leave attributable audit trails” is more directly tied to the required outcome.

“apply canonical hostname DNS and host resolution configuration during provisioning” addresses system identity and cluster inventory were not synchronized, but the current hardware or infrastructure issue is kernel and NVIDIA driver compatibility were not validated before imaging. The option “select a supported OS and kernel combination for the qualified driver stack” is more directly tied to the required outcome.

Useful for OS platform power policy conflicts with stable AI compute performance, but it does not remove the specific condition: kernel and NVIDIA driver compatibility were not validated before imaging.

“review required IOMMU and device access settings within the certified platform baseline” addresses OS security settings were applied without hardware-driver compatibility testing, but the current hardware or infrastructure issue is kernel and NVIDIA driver compatibility were not validated before imaging. The option “select a supported OS and kernel combination for the qualified driver stack” is more directly tied to the required outcome.

“DPU initialization and isolation tests pass on the hardened image” addresses OS security settings were applied without hardware-driver compatibility testing, but the current hardware or infrastructure issue is kernel and NVIDIA driver compatibility were not validated before imaging. The option “kernel module builds and nvidia-smi operate on the installed image” is more directly tied to the required outcome.

“forward and reverse lookups map to the intended managed nodes” addresses system identity and cluster inventory were not synchronized, but the current hardware or infrastructure issue is kernel and NVIDIA driver compatibility were not validated before imaging. The option “kernel module builds and nvidia-smi operate on the installed image” is more directly tied to the required outcome.

“the correct boot image loads and hardware inventory matches worker role” addresses network boot infrastructure does not bind worker identities to cluster images, but the current hardware or infrastructure issue is kernel and NVIDIA driver compatibility were not validated before imaging. The option “kernel module builds and nvidia-smi operate on the installed image” is more directly tied to the required outcome.

“apply canonical hostname DNS and host resolution configuration during provisioning” addresses system identity and cluster inventory were not synchronized, but the current hardware or infrastructure issue is network boot infrastructure does not bind worker identities to cluster images. The option “scope PXE boot and provisioning sources to the approved GPU-node image” is more directly tied to the required outcome.

“select a supported OS and kernel combination for the qualified driver stack” addresses kernel and NVIDIA driver compatibility were not validated before imaging, but the current hardware or infrastructure issue is network boot infrastructure does not bind worker identities to cluster images. The option “scope PXE boot and provisioning sources to the approved GPU-node image” is more directly tied to the required outcome.

“validate required management networking and container services after restart” addresses boot service dependencies were not part of image acceptance, but the current hardware or infrastructure issue is network boot infrastructure does not bind worker identities to cluster images. The option “scope PXE boot and provisioning sources to the approved GPU-node image” is more directly tied to the required outcome.

“kernel module builds and nvidia-smi operate on the installed image” addresses kernel and NVIDIA driver compatibility were not validated before imaging, but the current hardware or infrastructure issue is network boot infrastructure does not bind worker identities to cluster images. The option “the correct boot image loads and hardware inventory matches worker role” is more directly tied to the required outcome.

“DPU initialization and isolation tests pass on the hardened image” addresses OS security settings were applied without hardware-driver compatibility testing, but the current hardware or infrastructure issue is network boot infrastructure does not bind worker identities to cluster images. The option “the correct boot image loads and hardware inventory matches worker role” is more directly tied to the required outcome.

“boot audit shows prerequisites healthy before scheduler registration” addresses boot service dependencies were not part of image acceptance, but the current hardware or infrastructure issue is network boot infrastructure does not bind worker identities to cluster images. The option “the correct boot image loads and hardware inventory matches worker role” is more directly tied to the required outcome.

“validate required management networking and container services after restart” addresses boot service dependencies were not part of image acceptance, but the current hardware or infrastructure issue is operating system clock synchronization was omitted from bootstrap. The option “configure approved time services before registry and package operations” is more directly tied to the required outcome.

Useful for network boot infrastructure does not bind worker identities to cluster images, but it does not remove the specific condition: operating system clock synchronization was omitted from bootstrap.

“use architecture-aware provisioning and validate PCIe NUMA and interface mapping” addresses hardware-specific OS configuration was assumed portable without validation, but the current hardware or infrastructure issue is operating system clock synchronization was omitted from bootstrap. The option “configure approved time services before registry and package operations” is more directly tied to the required outcome.

“both platform groups enumerate expected devices and pass baseline I/O tests” addresses hardware-specific OS configuration was assumed portable without validation, but the current hardware or infrastructure issue is operating system clock synchronization was omitted from bootstrap. The option “system clock and TLS authenticated package downloads pass controlled tests” is more directly tied to the required outcome.

“DPU initialization and isolation tests pass on the hardened image” addresses OS security settings were applied without hardware-driver compatibility testing, but the current hardware or infrastructure issue is operating system clock synchronization was omitted from bootstrap. The option “system clock and TLS authenticated package downloads pass controlled tests” is more directly tied to the required outcome.

“kernel module builds and nvidia-smi operate on the installed image” addresses kernel and NVIDIA driver compatibility were not validated before imaging, but the current hardware or infrastructure issue is operating system clock synchronization was omitted from bootstrap. The option “system clock and TLS authenticated package downloads pass controlled tests” is more directly tied to the required outcome.

Useful for boot service dependencies were not part of image acceptance, but it does not remove the specific condition: OS disk layout did not reserve capacity for expected images and logs.

“apply canonical hostname DNS and host resolution configuration during provisioning” addresses system identity and cluster inventory were not synchronized, but the current hardware or infrastructure issue is OS disk layout did not reserve capacity for expected images and logs. The option “size and separate required system container and logging storage during imaging” is more directly tied to the required outcome.

“select a supported OS and kernel combination for the qualified driver stack” addresses kernel and NVIDIA driver compatibility were not validated before imaging, but the current hardware or infrastructure issue is OS disk layout did not reserve capacity for expected images and logs. The option “size and separate required system container and logging storage during imaging” is more directly tied to the required outcome.

“forward and reverse lookups map to the intended managed nodes” addresses system identity and cluster inventory were not synchronized, but the current hardware or infrastructure issue is OS disk layout did not reserve capacity for expected images and logs. The option “peak image-pull test finishes without filesystem exhaustion” is more directly tied to the required outcome.

“kernel module builds and nvidia-smi operate on the installed image” addresses kernel and NVIDIA driver compatibility were not validated before imaging, but the current hardware or infrastructure issue is OS disk layout did not reserve capacity for expected images and logs. The option “peak image-pull test finishes without filesystem exhaustion” is more directly tied to the required outcome.

“the correct boot image loads and hardware inventory matches worker role” addresses network boot infrastructure does not bind worker identities to cluster images, but the current hardware or infrastructure issue is OS disk layout did not reserve capacity for expected images and logs. The option “peak image-pull test finishes without filesystem exhaustion” is more directly tied to the required outcome.