Uploaded image for project: 'RHEL'
  1. RHEL
  2. RHEL-105254

RFE Image builder : there is no place for backup gpt table

Linking RHIVOS CVEs to...Migration: Automation ...RHELPRIO AssignedTeam ...SWIFT: POC ConversionSync from "Extern...XMLWordPrintable

    • Icon: Story Story
    • Resolution: Unresolved
    • Icon: Undefined Undefined
    • None
    • rhel-9.6, rhel-10.0
    • osbuild-composer
    • None
    • Low
    • rhel-sst-image-builder
    • None
    • False
    • False
    • Hide

      None

      Show
      None
    • None
    • None
    • None
    • None
    • Unspecified
    • Unspecified
    • Unspecified
    • None

      What were you trying to do that didn't work?

      Trying to create backup gpt table on install device (/dev/sda) on a system generated with image builder (tested with ova and qcow2)

      What is the impact of this issue to you?

      Install are not clean, we need to make some change after provisioning to have everything as expected.

      Please provide the package NVR for which the bug is seen:

      osbuild.noarch 141.2-1.el10_0

      How reproducible is this bug?:

      always

      Steps to reproduce

      On a RHEL10 server install image builder

      create blueprint like this one

       

      name = "RHEL9_test"
      description = ""
      version = "1.0.0"
      modules = []
      groups = []
      distro = "rhel-9"
      [customizations]
      hostname = "myhostname"
      [[customizations.user]]
      name = "root"
      password = "redhat123"
       

      then push it and create ova or qcow2 image and deploy it to the corresponding virtualization tool. 

      Then on the installation device run

      sgdisk -e /dev/<device name>
      

      Expected results

      The operation has completed successfully.

      Actual results

       

      {{Warning! Secondary partition table overlaps the last partition by
      2014 blocks!
      Try reducing the partition table size by 8056 entries.
      (Use the 's' item on the experts' menu.)
      Aborting write of new partition table.}}

       

      More information

      After some investigation it appears that osbuild set First usable sector at 2048 (probably with something like

      sfdisk /dev/<device name> << 'EOF'                             
      label: gpt
      unit: sectors
      first-lba: 2048
      EOF 

      The consequence is that sgdisk -e will copy the whole first 2047 unusable sectors instead of just the 33 first in a standard installation.

      And because osbuild is not adapting the last usable sector to "total number of sector minus 2048", the last partition is extended to the default last usable sector which is "total number of sector minus 33". That why executing sgdisk -e on this disk will generate this error.

       

       

              osbuilders Osbuilders Bot Account
              rh-ee-drenard Dorian Renard
              Osbuilders Bot Account Osbuilders Bot Account
              Release Test Team Release Test Team
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

                Created:
                Updated: