Configuring Multi-Subnet DHCP

Configure multi-subnet DHCP in Omnia to enable rack-based network provisioning with per-rack /24 subnets. This procedure covers editing the network_spec.yml file, validating the configuration, updating the CoreSMD container image, and deploying the CoreSMD changes to support multiple subnets via DHCP relay.

Prerequisites

Before configuring multi-subnet DHCP:

  • Omnia Infrastructure Manager (OIM) deployed and operational with omnia_core container up and running (prepare_oim.yml playbook executed successfully)

  • Network switches configured with VLANs and DHCP relay helper-address pointing to the OIM CoreSMD server

  • Network topology documented with rack IDs, subnet allocations, gateway IPs, and VLAN assignments

  • DHCP pool ranges planned and validated to avoid conflicts with static IPs and OIM admin IP

  • PXE mapping file configured and validated for your deployment scenario. Ensure that the pxe_mapping_file.csv is aligned with your network topology. For sample configurations, see `Sample Files`_.

Important

Multi-Subnet DHCP requires DHCP relay agents configured on each subnet’s gateway/router. Without proper DHCP relay configuration, DHCP requests from remote subnets will not reach the CoreSMD server.

Single-Subnet Baseline

Before converting to multi-subnet, verify your single-subnet deployment is working correctly.

network_spec.yml (inside omnia_core container)

Use SSH to connect to the omnia_core container and view the current configuration:

ssh omnia_core
cat /opt/omnia/input/project_default/network_spec.yml

A single-subnet configuration looks like this:

Networks:
- admin_network:
    oim_nic_name: "eno12399np0"
    subnet: "10.40.1.0"
    netmask_bits: "24"
    primary_oim_admin_ip: "10.40.1.111"
    primary_oim_bmc_ip: ""
    router: "10.40.1.1"
    dynamic_range: "10.40.1.201-10.40.1.250"
    dns: []
    ntp_servers: []
    additional_subnets: []

- ib_network:
    subnet: "198.168.0.0"
    netmask_bits: "24"
    dns: []

Note

The router field is mandatory and specifies the gateway IP for the admin network (used as DHCP option 3). If no dedicated router is available, use the primary_oim_admin_ip value.

Generated coredhcp.yaml (on OIM host)

Exit the omnia_core container and view the CoreDHCP configuration on the OIM host:

exit
cat /etc/openchami/configs/coredhcp.yaml

For a single-subnet deployment, the file contains:

server4:
  listen:
    - "%eno12399np0"
  plugins:
    - server_id: 10.40.1.111
    - dns: 10.40.1.111
    - router: 10.40.1.1
    - netmask: 255.255.255.0
    # Single-subnet mode: positional argument format (coresmd v0.4.x)
    - coresmd: https://oimcp.omnia.test:8443 http://10.40.1.111:8081 /root_ca/root_ca.crt 30s 86400s true
    - bootloop: /tmp/coredhcp.db default 5m 10.40.1.201 10.40.1.250

CoreSMD container quadlet files (on OIM host)

The single-subnet deployment uses CoreSMD v0.4.3. Verify the container quadlet files:

cat /etc/containers/systemd/coresmd-coredhcp.container

[Unit]
Description=The CoreSMD CoreDHCP container
Wants=haproxy.service
After=haproxy.service
PartOf=openchami.target

[Container]
ContainerName=coresmd-coredhcp

HostName=coresmd-coredhcp
Image=ghcr.io/openchami/coresmd:v0.4.3

# Capabilities
AddCapability=NET_ADMIN
AddCapability=NET_RAW

# Volumes
Volume=/etc/pki/ca-trust/extracted/pem/tls-ca-bundle.pem:/root_ca/root_ca.crt:ro,Z
Volume=/etc/openchami/configs/coredhcp.yaml:/etc/coredhcp/config.yaml:ro,Z

# Networks for the Container to use
Network=host

# Unsupported by generator options
# Proxy settings
PodmanArgs=--http-proxy=false

[Service]
Restart=always

cat /etc/containers/systemd/coresmd-coredns.container

[Unit]
Description=The CoreSMD CoreDNS container
Wants=haproxy.service
After=haproxy.service
PartOf=openchami.target

[Container]
ContainerName=coresmd-coredns

HostName=coresmd-coredns
Image=ghcr.io/openchami/coresmd:v0.4.3

Exec=/coredns

# Capabilities
AddCapability=NET_ADMIN
AddCapability=NET_RAW

# Volumes
Volume=/etc/pki/ca-trust/extracted/pem/tls-ca-bundle.pem:/root_ca/root_ca.crt:ro,Z
Volume=/etc/openchami/configs/Corefile:/Corefile:ro,Z

# Networks for the Container to use
Network=host

# Unsupported by generator options
# Proxy settings
PodmanArgs=--http-proxy=false

[Service]
Restart=always

Converting Single-Subnet to Multi-Subnet

Follow these steps to convert an existing single-subnet Omnia 2.2 deployment to multi-subnet.

Step 1: Update network_spec.yml

Use SSH to connect to the omnia_core container and edit network_spec.yml to add the additional subnets:

ssh omnia_core
vi /opt/omnia/input/project_default/network_spec.yml

Add the additional_subnets array with subnet entries for each rack. For a complete description of subnet parameters and the multi-subnet DHCP architecture, see `Multi-Subnet DHCP Overview`_.

Example configuration for 2 racks:

Networks:
- admin_network:
    oim_nic_name: "eno12399np0"
    subnet: "10.40.1.0"
    netmask_bits: "24"
    primary_oim_admin_ip: "10.40.1.111"
    primary_oim_bmc_ip: ""
    router: "10.40.1.1"
    dynamic_range: "10.40.1.201-10.40.1.250"
    dns: []
    ntp_servers: []
    additional_subnets:
      - subnet: "10.40.2.0"
        netmask_bits: "24"
        router: "10.40.2.1"
        dynamic_range: "10.40.2.190-10.40.2.200"

      - subnet: "10.40.3.0"
        netmask_bits: "24"
        router: "10.40.3.1"
        dynamic_range: "10.40.3.190-10.40.3.200"

- ib_network:
    subnet: "198.168.0.0"
    netmask_bits: "24"
    dns: []

Note

Leave additional_subnets: [] (empty array) for single-subnet deployments. This maintains backward compatibility with existing configurations.

Step 2: Run prepare_oim.yml

Inside the omnia_core container, run the prepare_oim.yml playbook to regenerate the CoreDHCP configuration:

cd /omnia/prepare_oim
ansible-playbook prepare_oim.yml

Step 3: Verify the generated coredhcp.yaml

Exit the omnia_core container and verify the updated CoreDHCP configuration on the OIM host:

exit
cat /etc/openchami/configs/coredhcp.yaml

The file will contain the single-subnet configuration as active, with the multi-subnet configuration commented out below it:

server4:
  listen:
    - "%eno12399np0"
  plugins:
    - server_id: 10.40.1.111
    - dns: 10.40.1.111
    - router: 10.40.1.1
    - netmask: 255.255.255.0
    # Single-subnet mode: positional argument format (coresmd v0.4.x)
    - coresmd: https://oimcp.omnia.test:8443 http://10.40.1.111:8081 /root_ca/root_ca.crt 30s 86400s true
    - bootloop: /tmp/coredhcp.db default 5m 10.40.1.201 10.40.1.250
    # -------------------------------------------------------------------
    # Multi-subnet configuration (requires coresmd v0.6.x+)
    # To enable multi-subnet DHCP:
    #   1. Pull the new coresmd image: podman pull ghcr.io/openchami/coresmd:v0.6.3
    #   2. Comment out the single-subnet coresmd and bootloop lines above
    #   3. Uncomment the multi-subnet coresmd and bootloop blocks below
    #   4. Replace the new coresmd image version in files: /etc/containers/systemd/coresmd-coredhcp.container /etc/containers/systemd/coresmd-coredns.container with the old version
    #   5. Reload daemon: systemctl daemon-reload
    #   6. Restart services: systemctl restart openchami.target
    # -------------------------------------------------------------------
    # - coresmd: |
    #     svc_base_uri=https://oimcp.omnia.test:8443
    #     ipxe_base_uri=http://10.40.1.111:8081
    #     ca_cert=/root_ca/root_ca.crt
    #     cache_valid=30s
    #     lease_time=86400s
    #     single_port=true
    #     rule=subnet:10.40.1.0/24,type:Node,routers:10.40.1.1,cidr:24
    #     rule=subnet:10.40.2.0/24,type:NodeBMC,routers:10.40.2.1,cidr:24
    #     rule=subnet:10.40.3.0/24,type:Node,routers:10.40.3.1,cidr:24
    #     rule=subnet:10.40.4.0/24,type:NodeBMC,routers:10.40.4.1,cidr:24
    #     rule=type:Node
    #     rule=type:NodeBMC
    #     rule=hostname:unknown-{04d}
    # - bootloop: |
    #     lease_file=/tmp/coredhcp.db
    #     script_path=default
    #     lease_time=5m
    #     subnet_pool=10.40.1.0/24,10.40.1.190,10.40.1.200
    #     subnet_pool=10.40.2.0/24,10.40.2.190,10.40.2.200
    #     subnet_pool=10.40.3.0/24,10.40.3.190,10.40.3.200
    #     subnet_pool=10.40.4.0/24,10.40.4.190,10.40.4.200

Step 4: Stop OpenCHAMI services

On the OIM host, stop all OpenCHAMI services before modifying the container configuration:

systemctl stop openchami.target

Step 5: Pull the new CoreSMD image

Pull the CoreSMD v0.6.3 image which supports multi-subnet DHCP:

podman pull ghcr.io/openchami/coresmd:v0.6.3

Step 6: Enable multi-subnet in coredhcp.yaml

On the OIM host, edit the CoreDHCP configuration to switch from single-subnet to multi-subnet mode:

vi /etc/openchami/configs/coredhcp.yaml

Make the following changes:

  1. Comment out the single-subnet coresmd and bootloop lines (add # prefix)

  2. Uncomment the multi-subnet coresmd and bootloop blocks (remove # prefix)

The resulting file should look like this:

server4:
  listen:
    - "%eno12399np0"
  plugins:
    - server_id: 10.40.1.111
    - dns: 10.40.1.111
    - router: 10.40.1.1
    - netmask: 255.255.255.0
    # Single-subnet mode: positional argument format (coresmd v0.4.x)
    #- coresmd: https://oimcp.omnia.test:8443 http://10.40.1.111:8081 /root_ca/root_ca.crt 30s 86400s true
    #- bootloop: /tmp/coredhcp.db default 5m 10.40.1.201 10.40.1.250
    # -------------------------------------------------------------------
    # Multi-subnet configuration (requires coresmd v0.6.x+)
    # To enable multi-subnet DHCP:
    #   1. Pull the new coresmd image: podman pull ghcr.io/openchami/coresmd:v0.6.3
    #   2. Comment out the single-subnet coresmd and bootloop lines above
    #   3. Uncomment the multi-subnet coresmd and bootloop blocks below
    #   4. Replace the new coresmd image version in files: /etc/containers/systemd/coresmd-coredhcp.container /etc/containers/systemd/coresmd-coredns.container with the old version
    #   5. Reload daemon: systemctl daemon-reload
    #   6. Restart services: systemctl restart openchami.target
    # -------------------------------------------------------------------
    - coresmd: |
        svc_base_uri=https://oimcp.omnia.test:8443
        ipxe_base_uri=http://10.40.1.111:8081
        ca_cert=/root_ca/root_ca.crt
        cache_valid=30s
        lease_time=86400s
        single_port=true
        rule=subnet:10.40.1.0/24,type:Node,routers:10.40.1.1,cidr:24
        rule=subnet:10.40.2.0/24,type:NodeBMC,routers:10.40.2.1,cidr:24
        rule=subnet:10.40.3.0/24,type:Node,routers:10.40.3.1,cidr:24
        rule=subnet:10.40.4.0/24,type:NodeBMC,routers:10.40.4.1,cidr:24
        rule=type:Node
        rule=type:NodeBMC
        rule=hostname:unknown-{04d}
    - bootloop: |
        lease_file=/tmp/coredhcp.db
        script_path=default
        lease_time=5m
        subnet_pool=10.40.1.0/24,10.40.2.190,10.40.1.200
        subnet_pool=10.40.2.0/24,10.40.3.190,10.40.2.200
        subnet_pool=10.40.3.0/24,10.40.2.190,10.40.3.200
        subnet_pool=10.40.4.0/24,10.40.3.190,10.40.4.200

Step 7: Update coresmd-coredhcp.container

Update the CoreSMD CoreDHCP container quadlet file to use the new image version:

vi /etc/containers/systemd/coresmd-coredhcp.container

Change the Image line from v0.4.3 to v0.6.3:

[Unit]
Description=The CoreSMD CoreDHCP container
Wants=haproxy.service
After=haproxy.service
PartOf=openchami.target

[Container]
ContainerName=coresmd-coredhcp

HostName=coresmd-coredhcp
Image=ghcr.io/openchami/coresmd:v0.6.3

# Capabilities
AddCapability=NET_ADMIN
AddCapability=NET_RAW

# Volumes
Volume=/etc/pki/ca-trust/extracted/pem/tls-ca-bundle.pem:/root_ca/root_ca.crt:ro,Z
Volume=/etc/openchami/configs/coredhcp.yaml:/etc/coredhcp/config.yaml:ro,Z

# Networks for the Container to use
Network=host

# Unsupported by generator options
# Proxy settings
PodmanArgs=--http-proxy=false

[Service]
Restart=always

Step 8: Update coresmd-coredns.container

Update the CoreSMD CoreDNS container quadlet file to use the new image version:

vi /etc/containers/systemd/coresmd-coredns.container

Change the Image line from v0.4.3 to v0.6.3:

[Unit]
Description=The CoreSMD CoreDNS container
Wants=haproxy.service
After=haproxy.service
PartOf=openchami.target

[Container]
ContainerName=coresmd-coredns

HostName=coresmd-coredns
Image=ghcr.io/openchami/coresmd:v0.6.3

Exec=/coredns

# Capabilities
AddCapability=NET_ADMIN
AddCapability=NET_RAW

# Volumes
Volume=/etc/pki/ca-trust/extracted/pem/tls-ca-bundle.pem:/root_ca/root_ca.crt:ro,Z
Volume=/etc/openchami/configs/Corefile:/Corefile:ro,Z

# Networks for the Container to use
Network=host

# Unsupported by generator options
# Proxy settings
PodmanArgs=--http-proxy=false

[Service]
Restart=always

Step 9: Reload systemd and restart services

Reload the systemd daemon to pick up the quadlet file changes, then restart the OpenCHAMI services:

systemctl daemon-reload
systemctl restart openchami.target

Verification

After configuring multi-subnet DHCP, verify the following:

  1. Verify all OpenCHAMI services are running:

    systemctl list-dependencies openchami.target

    Ensure that services such as CoreSMD and other dependent services are in an active state.

  2. Verify CoreSMD has registered the additional subnets. Expected output should show subnet= directives for each additional subnet:

    podman logs coresmd-coredhcp | grep "subnet="

  3. Check error logs if any of the core services fail to start:

    journalctl -xeu coresmd-coredhcp
journalctl -xeu coresmd-coredns

Configuration Examples

Ten-Rack Configuration

For a large deployment with 10 racks:

Networks:
- admin_network:
    oim_nic_name: "eno1"
    subnet: "10.40.1.0"
    netmask_bits: "24"
    primary_oim_admin_ip: "10.40.1.111"
    primary_oim_bmc_ip: ""
    router: "10.40.1.1"
    dynamic_range: "10.40.1.201-10.40.1.250"
    dns: []
    ntp_servers: []
    additional_subnets:
      - subnet: "10.40.2.0"
        netmask_bits: "24"
        router: "10.40.2.1"
        dynamic_range: "10.40.2.190-10.40.2.200"
      - subnet: "10.40.3.0"
        netmask_bits: "24"
        router: "10.40.3.1"
        dynamic_range: "10.40.3.190-10.40.3.200"
      - subnet: "10.40.5.0"
        netmask_bits: "24"
        router: "10.40.5.1"
        dynamic_range: "10.40.5.100-10.40.5.200"
      - subnet: "10.40.7.0"
        netmask_bits: "24"
        router: "10.40.7.1"
        dynamic_range: "10.40.7.100-10.40.7.200"
      - subnet: "10.40.9.0"
        netmask_bits: "24"
        router: "10.40.9.1"
        dynamic_range: "10.40.9.100-10.40.9.200"
      - subnet: "10.40.11.0"
        netmask_bits: "24"
        router: "10.40.11.1"
        dynamic_range: "10.40.11.100-10.40.11.200"
      - subnet: "10.40.13.0"
        netmask_bits: "24"
        router: "10.40.13.1"
        dynamic_range: "10.40.13.100-10.40.13.200"
      - subnet: "10.40.15.0"
        netmask_bits: "24"
        router: "10.40.15.1"
        dynamic_range: "10.40.15.100-10.40.15.200"
      - subnet: "10.40.17.0"
        netmask_bits: "24"
        router: "10.40.17.1"
        dynamic_range: "10.40.17.100-10.40.17.200"
      - subnet: "10.40.19.0"
        netmask_bits: "24"
        router: "10.40.19.1"
        dynamic_range: "10.40.19.100-10.40.19.200"

This configuration supports 10 racks with non-overlapping /24 subnets, each with 100 IP addresses available for DHCP allocation.

If you have any feedback about Omnia documentation, please reach out at omnia.readme@dell.com.

If you have any feedback about Omnia documentation, please reach out at omnia.readme@dell.com.