IBM Support

BareMetal CIDR address configuration error

General Page

During the installation of IBM Fusion HCI version 2.10.0 using the Static IP method, an error occurs when setting up a BareMetal CIDR address with a prefix other than /24. The installer either fails to recognize the specified prefix or defaults to /24, regardless of the input provided.

It leads to incorrect network configuration and can cause problems with cluster communication and provisioning. There are two scenarios to consider:

Before cluster installation - If you have not started the cluster installation process (stage2 install) yet.
After cluster installation - If you have already completed the cluster installation process.
This hotfix only applies to the following conditions:
  • Installing IBM Fusion HCI version 2.10.0 for the first time.
  • Using the Static IP method for installation.
  • If BareMetal CIDR address contains a prefix that is not /24.
    NOTE: This configuration is set during Stage 1 under the field name Provisioner Node CIDR Address.
How to identify the issue:
 
Before cluster installation:
 

Verify whether you see the following sample outputs instead of "Blank" outputs after running the commands on the provisioner node (RU7 or Servicenode) as kni user after Stage 1 completion.

  1. Run the following command.
     cat ~/ipi-installer/ansible-ipi-install/roles/isf_hosts/templates/isf_hosts.j2 | grep -i provisioneripv4

    Sample output:
    provisionerIpV4="{{ user_config.provisionerIpV4 }}"
    brExIpV4Subnet="{{ user_config.provisionerIpV4.split('/')[1] }}"
     
  2. Run the following command.
     cat ~/ipi-installer/ansible-ipi-install/roles/installer/templates/install-config.j2 | grep -i brExIpV4Subnet

    Sample output:
    prefix-length: {{ brExIpV4Subnet }}
    prefix-length: {{ brExIpV4Subnet }}
     

    If you see "Blank" outputs instead of sample outputs after the running commands, then apply the resolution steps.

After cluster installation:
 
  1. After the cluster creation, run the following command to verify the issue by logging into the cluster nodes.
    oc debug node/<node-name>

  2. Once you logged into the debug shell, then run the following command:
    chroot /host
    ip a s | grep br-ex
     

    If the output shows a prefix different from what you entered (for example, /24 instead of /27), then the problem exists in the cluster.

    br-ex: mtu 9000 qdisc noqueue state UNKNOWN group default qlen 1000
    inet 172.20.4.85/24 brd 172.20.5.255 scope global noprefixroute br-ex
    inet 169.254.0.2/17 brd 169.254.127.255 scope global br-ex

    This confirms that the installer ignored the user-defined prefix and applied a default /24.

    In such case, contact IBM Fusion support team for recovery.



Resolution:

To prevent this issue before cluster installtion (Stage 2 - OpenShift installation), follow the steps:
  1. Download the xxxx (link is yet to provide) file.

  2. Copy "cidr_hotfix.zip" on the provisioner node (RU7 or service-node) under directory /home/kni/hotfix/.

    • Run the following to create a directory.
      mkdir /home/kni/hotfix
       
    • Run the following command to go inside the directory. 
      cd /home/kni/hotfix/
       
  3. Run the following command to extract the cidr_hotfix.zip directory.
    unzip cidr_hotfix.zip
     
  4. Copy the install-config.j2 file from the fixed location to the actual code location. Run the following commands one by one:

    cd /home/kni/ipi-installer/ansible-ipi-install/roles/installer/templates/

    cp install-config.j2 install-config.j2_org

    > install-config.j2

    cat /home/kni/hotfix/install-config.j2 >> install-config.j2
     
  5. Copy the isf_hosts.j2 file from the fixed location to the actual code location. Run the following commands one by one:

    cd /home/kni/ipi-installer/ansible-ipi-install/roles/isf_hosts/templates/

    cp isf_hosts.j2 isf_hosts.j2_org

    > isf_hosts.j2

    cat /home/kni/hotfix/ isf_hosts.j2 >> isf_hosts.j2
     
  6. After you run all steps from 1 through 5, proceed with the Stage 2 installation.

    After Stage 2 is completed, follow the steps before you start adding worker nodes.
     
  7. Copy the following images in your artifactory or quay registry for offline installation.

    Artifactory registry:

    skopeo copy --insecure-policy --preserve-digests --all docker://cp.icr.io/cp/fusion-hci/isf-prereq-operator@sha256:12ee7ed72ec034bd7ee82bf3987e73e50e813fd1445577a4b75e2ffe0529c62e docker://$TARGET_PATH/cp/fusion-hci/isf-prereq-operator@sha256:12ee7ed72ec034bd7ee82bf3987e73e50e813fd1445577a4b75e2ffe0529c62e

    Quay registry:

    skopeo copy --insecure-policy --preserve-digests --all docker://cp.icr.io/cp/fusion-hci/isf-prereq-operator@sha256:12ee7ed72ec034bd7ee82bf3987e73e50e813fd1445577a4b75e2ffe0529c62e docker://$TARGET_PATH/cp/fusion-hci/isf-prereq-operator:12ee7ed72ec034bd7ee82bf3987e73e50e813fd1445577a4b75e2ffe0529c62e
     
  8. Run the following commands to get the IBM Fusion operator CSV.
    oc -n ibm-spectrum-fusion-ns get csv
    oc -n ibm-spectrum-fusion-ns get csv <OPERATOR_NAME> isf-<OPERATOR_NAME>.yaml
     

    Example output:

    oc -n ibm-spectrum-fusion-ns get csv
NAME                  DISPLAY             VERSION  REPLACES             PHASE
isf-operator.v2.10.0  IBM Storage Fusion  2.10.0   isf-operator.v2.9.0  Succeeded
oc -n ibm-spectrum-fusion-ns get csv isf-operator.v2.10.0 > isf-operator.v2.10.0.yaml
  9. Update the prereq operator image digest in the CSV as follows:
    1. Log in to the OpenShift Container Platform web console.
    2. Go to Operators > Installed Operators.
    3. In the Project drop-down, select ibm-spectrum-fusion-ns namespace.
    4. Click IBM Storage Fusion operator.
    5. Go to YAML tab.
    6. Update the prereq operator image using the following image.
      cp.icr.io/cp/fusion-hci/isf-prereq-operator@sha256:12ee7ed72ec034bd7ee82bf3987e73e50e813fd1445577a4b75e2ffe0529c62e
    7. Click Save.
    8. Wait for 5 minutes for the isf-prereq-operator-controller-manager operator deployment to update automatically, and then check that the isf-prereq-operator-controller-manager operator pod is recreated successfully.

 

[{"Type":"MASTER","Line of Business":{"code":"LOB69","label":"Storage TPS"},"Business Unit":{"code":"BU048","label":"IBM Software"},"Product":{"code":"SSXEFDS","label":"IBM Fusion HCI Appliance Software"},"ARM Category":[{"code":"a8m3p0000000rX7AAI","label":"HW"}],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"2.10.0"}]

Document Information

Modified date:
29 July 2025

UID

ibm17240826

Page Feedback