Creating a fabric intent for unmanaged nodes

Ensure that:

the participating nodes and their links are already configured as a functioning fabric
an external DHCP server is in place
a manual topology file has been created describing the intended fabric
Note: It is important that this topology file is an accurate representation of the unmanaged fabric. It is your responsibility to ensure that the information in the topology file is complete and accurate.
Note: The "isManaged" flag must be disabled for all nodes in the topology file. The Fabric Services System does not support a fabric that contains a mixture of managed and unmanaged nodes.
an association file has been created containing the serial number and management IP address for each node
Note: The management IP address is mandatory; the serial number is optional but recommended. For any node to be the subject of a future maintenance intent, the serial number is mandatory.

Follow this procedure to create and deploy a fabric intent consisting of unmanaged nodes. Most steps refer you to existing procedures. Some steps include special requirements when performing those other procedures.

Use the Region Selector at the top of the page to select the region in which to create the fabric intent.
Import the manual topology file that describes the existing fabric into the Fabric Services System so that it is available for use when you create the fabric intent.
See Importing a manual topology.
Ensure that certificates are in place for the participating nodes.
Create a Management IP pool that contains one or more CIDR blocks intended for use by unmanaged nodes.

Note: When creating any CIDR block within this pool, you must disable the Is Managed property. This is a unique requirement for management IP pools intended for use with unmanaged nodes.

See Creating IP and Autonomous System pools.
Create a fabric intent that uses a manual topology, including saving the fabric intent and generating the fabric topology. This adds the set of nodes within the topology to the Fabric Services System inventory.

Note: For Imported Topology From File, select the manual fabric topology you previously uploaded.

Note: For the Management Pool Name, select the Management IP Pool you configured containing a CIDR block whose Is Managed property is disabled.

Note: After you generate the fabric, the topology display in the Fabric Design view of the fabric intents page will not display any links between the nodes of an unmanaged fabric if Inter-Switch Link (ISL) data was not included in the manual topology data.

See Creating a fabric intent using a manual topology.
Associate each planned node in the fabric with a node contained in the Fabric Services System inventory by assigning it a management IP address (mandatory) and a serial number (optional, but recommended). Typically, this association is achieved by importing a file with the association data that was prepared ahead of time.

Note: An unmanaged node must be running a supported version of SR Linux to be discovered and become Ready during this step.

See Uploading association data to the fabric intent inventory
Note: Immediately after you complete this step and the node enters a Ready state, the following occurs:
- A set of system GCOs is created: one for each node that is in a Ready state, containing the complete, discovered configuration for that node.
- If the region containing this fabric intent is set to use the Fabric Services System's internal DHCP server, a set of entries is created in the internal DHCP list for the region, one for each Ready node. These are required if you in the future create a maintenance intent for an unmanaged node.
If the node used a Deployment Trigger of greater than 0%, deploy the fabric intent.
See Deploy the fabric intent to hardware

Note: If the fabric is configured with a 0% deployment trigger, deployment proceeds automatically.

Note: This deployment action does not send configuration data to the nodes participating in the fabric. This is purely an internal process that satisfies requirements within the Fabric Services System.

Upon completion of this procedure, the fabric consisting of unmanaged nodes is known to the Fabric Services System. Other capabilities of the Fabric Services System are now active with respect this fabric, including alarms, operational health displays, and so on.