To Reproduce Steps to reproduce the behavior:

1. Enter into your favorite RHEL AI image

• • registry.redhat.io/rhelai1/instructlab-nvidia-rhel9:1.3.1-1733951397

1. Try to install a package within the container
   • dnf install bc
2. Errors out since it can't find RHELAI 1.3 RHEL 9 RPMs metadata

Expected behavior

• DNF should not error out within InstructLab containers

Screenshots

Device Info (please complete the following information):

• Hardware Specs: Intel Icelake Xeon, 1.3T Mem, IBM Cloud system
• OS Version: RHEL 9.4, RHELAI Bootc 1.3.1
• InstructLab Version: ilab, version 0.21.2
• Provide the output of these two commands:
  • sudo bootc status --format json | jq .status.booted.image.image.image to print the name and tag of the bootc image, should look like registry.stage.redhat.io/rhelai1/bootc-intel-rhel9:1.3-1732894187
    • registry.redhat.io/rhelai1/bootc-nvidia-rhel9:1.3.1
    • Also observed using RHEL and starting an instructlab container
  • ilab system info to print detailed information about InstructLab version, OS, and hardware – including GPU / AI accelerator hardware
    • Platform:
        sys.version: 3.11.7 (main, Oct  9 2024, 00:00:00) [GCC 11.4.1 20231218 (Red Hat 11.4.1-3)]
        sys.platform: linux
        os.name: posix
        platform.release: 5.14.0-427.48.1.el9_4.x86_64
        platform.machine: x86_64
        platform.node: 8c85982038d8
        platform.python_version: 3.11.7
        os-release.ID: rhel
        os-release.VERSION_ID: 9.4
        os-release.PRETTY_NAME: Red Hat Enterprise Linux 9.4 (Plow)
        memory.total: 1259.37 GB
        memory.available: 1245.61 GB
        memory.used: 3.55 GB

InstructLab:
  instructlab.version: 0.22.1
  instructlab-dolomite.version: 0.2.0
  instructlab-eval.version: 0.4.2
  instructlab-quantize.version: 0.1.0
  instructlab-schema.version: 0.4.1
  instructlab-sdg.version: 0.6.2
  instructlab-training.version: 0.6.1

Torch:
  torch.version: 2.4.1
  torch.backends.cpu.capability: AVX512
  torch.version.cuda: 12.4
  torch.version.hip: None
  torch.cuda.available: False
  torch.backends.cuda.is_built: True
  torch.backends.mps.is_built: False
  torch.backends.mps.is_available: False

llama_cpp_python:
  llama_cpp_python.version: 0.2.79
  llama_cpp_python.supports_gpu_offload: True

Bug impact

• Customers cannot install additional tools in InstructLab container should they desire.  This can cause issues with debugging issues if the needed tooling is not built into the InstructLab image.
• This causes issues with automation should it need to install dependencies within the running container and the system is outside the Red Hat network (such as in public clouds)

Known workaround

• Delete, Move, or Disable /etc/yum.repos.d/rhelai.repo

Additional context

• There is no public DNS for rhsm-pulp.corp.redhat.com, it can only resolve within Red Hat network or VPN.
• I've observed this behavior on all released Instructlab containers including prelease containers