Patch a Host Transport Node

Transport nodes are hypervisor hosts that will participate
in an NSX-T overlay. For a hypervisor host, this means that it hosts
VMs that will communicate over NSX-T logical switches.

This API creates transport node for a host node (hypervisor) in the transport network.

When you run this command for a host, NSX Manager attempts to install the
NSX kernel modules, which are packaged as VIB, RPM, or DEB files. For the
installation to succeed, you must provide the host login credentials and the
host thumbprint.

To get the ESXi host thumbprint, SSH to the host and run the
openssl x509 -in /etc/vmware/ssl/rui.crt -fingerprint -sha256 -noout
command.

To generate host key thumbprint using SHA-256 algorithm please follow the
steps below.

Log into the host, making sure that the connection is not vulnerable to a
man in the middle attack. Check whether a public key already exists.
Host public key is generally located at '/etc/ssh/ssh_host_rsa_key.pub'.
If the key is not present then generate a new key by running the following
command and follow the instructions.

ssh-keygen -t rsa

Now generate a SHA256 hash of the key using the following command. Please
make sure to pass the appropriate file name if the public key is stored with
a different file name other than the default 'id_rsa.pub'.

awk '{print $2}' id_rsa.pub | base64 -d | sha256sum -b | sed 's/ .*$//' | xxd -r -p | base64

Additional documentation on creating a transport node can be found
in the NSX-T Installation Guide.

In order for the transport node to forward packets,
the host_switch_spec property must be specified.

Host switches (called bridges in OVS on KVM hypervisors) are the
individual switches within the host virtual switch. Virtual machines
are connected to the host switches.

When creating a transport node, you need to specify if the host switches
are already manually preconfigured on the node, or if NSX should create
and manage the host switches. You specify this choice by the type
of host switches you pass in the host_switch_spec property of the
TransportNode request payload.

For a KVM host, you can preconfigure the host switch, or you can have
NSX Manager perform the configuration. For an ESXi host NSX Manager always
configures the host switch.

To preconfigure the host switches on a KVM host, pass an array
of PreconfiguredHostSwitchSpec objects that describes those host
switches. In the current NSX-T release, only one prefonfigured host
switch can be specified. See the PreconfiguredHostSwitchSpec schema
definition for documentation on the properties that must be provided.
Preconfigured host switches are only supported on KVM hosts, not on
ESXi hosts.

To allow NSX to manage the host switch configuration on KVM hosts,
ESXi hosts, pass an array of StandardHostSwitchSpec objects in the
host_switch_spec property, and NSX will automatically
create host switches with the properties you provide. In the current
NSX-T release, up to 16 host switches can be automatically managed.
See the StandardHostSwitchSpec schema definition for documentation on
the properties that must be provided.

The request should provide node_deployement_info.
This API is available when using VMware Cloud (AWS, Dell-EMC, Outpost, Hyperscalers) or VMware NSX.

Request:

Method:

PATCH

URI Path(s):

/policy/api/v1/infra/sites/{site-id}/enforcement-points/{enforcementpoint-id}/host-transport-nodes/{host-transport-node-id}

Request Headers:

n/a

Query Parameters:

TransportNodeUpdateParameters+

TransportNodeUpdateParameters (schema)

(Deprecated)

Name	Description	Type	Notes
esx_mgmt_if_migration_dest	The network ids to which the ESX vmk interfaces will be migrated A comma separated list of network ids. When migrating vmks into logical switches, the ids are the logical switches's ids. When migrating out of logical switches, the ids are vSphere Standard Switch portgroup names in a single vSphere Standard Switch, or distributed virtual portgroup names in a single distributed virtual switch (DVS). This property can only used together with 'if_id'.	string
if_id	The ESX vmk interfaces to migrate A comma separated list of vmk interfaces (for example, vmk0,vmk1). This property can only used along with 'esx_mgmt_if_migration_dest'. If all vmk interfaces will be migrated into the same logical switch or DV portgroup, the 'esx_mgmt_if_migration_dest' can be just one logical switch id or DV portgroup name. Otherwise the number of vmks in this list must equal the number of ids in 'esx_mgmt_if_migration_dest' list, and the orders of the two lists are important because the vmks match the network ids one by one in the same order.	string
override_nsx_ownership	Override NSX Ownership Flag indicating whether the NSX ownership constraints (on Managed Objects like Host/Cluster/DVS) should be overridden/bypassed. Note: Overriding/bypassing NSX ownership constraints is not recommended at all. This indicates, you want to use/configure/own certain Managed Objects (like Cluster, Host or DVS) which seem to be already in use/configured/owned by some other NSX instance. This option should be used with caution. It should only be used to come out of situations where: a. The other NSX instance no longer intends to use the Managed Objects (and has already unconfigured NSX configurations) but the ownership still lies with it (incorrectly) and you want those Managed Objects to be used/configured/owned by this NSX instance. b. The other NSX instance has crashed or decommisioned but the ownership still lies with it and you want those Managed Objects to be used/configured/owned by this NSX instance. Enabling this option, while the Managed Objects affected by this operation are actively used by other NSX, can lead to problematic states on both the NSX instances. For example, if a TN is forcefully reconfigured by this NSX instance (using override_nsx_ownership=true), while it was already configured and in use by the other NSX instance, it could corrupt the HostSwitch configurations pushed down by the other NSX instance.	boolean	Default: "False"
ping_ip	IP Addresses to ping right after ESX vmk interfaces were migrated. A comma separated list of IP addresses that match the vmk interfaces given in property 'if_id" or 'vnic' one-by-one in the same order. '0.0.0.0' is a special IP that indicates the pre-migration gateway of the vmk will be pinged post-migration. If a VMK does not need the ping ip or a VM NIC is given inside 'vnic', the ping ip must be skipped but the comma has to stay. For example, '0.0.0.0,,10.1.1.1' indicates the vmk or VM NIC at the 2nd position does not need ping post-migration. Right after all ESX vmk interfaces are migrated, ping packets will be sent through each vmk to its given ping_ip to check if the migraton will break the network connectivity or not. If any vmk_ping fails, the whole migration of all vmks will be rolled back and transport-node will be in failed state.	string
skip_validation	Whether to skip front-end validation for vmk/vnic/pnic migration If this property is set true, all front-end validation for vmk, vnic, and/or pnic migration will be skipped. This is useful when the remote host becomes unreachable as a result of a migration; in which case the front-end validation will always fail because data from the remote host is no longer available. Skipping the validation will allow user to undo the migration by updating the transport node first and then restoring the host network connectivity.	boolean	Default: "False"
vnic	The ESX vmk interfaces and/or VM NIC to migrate A comma separated list of vmk interfaces and/or one VM NIC. Only one VM NIC is allowed in the list; the format must be vmInstanceUuid:DeviceId like '50ca5f2d-1fa2-432d-991e-f01e0e16d182:4000'. An example list is 'vmk0,vmk1,50ca5f2d-1fa2-432d-991e-f01e0e16d182:4000'. The property can only be used along with 'vnic_migration_dest'.	string
vnic_migration_dest	The migration destinations of ESX vmk interfaces and/or VM NIC A comma separated list of vif ids, or port group names. When migrating into logical switches, the ids are vif ids in the logical ports created in the logical switches. When migrating out of logical switches, the ids are vSphere Standard Switch portgroup names in a single vSphere Standard Switch, or distributed virtual portgroup names in a single distributed virtual switch (DVS). The property can only be used in combination with property 'vnic'. The number of vnic interfaces in 'vnic' must equal the number of vif ids or port-group names in this list. The items in the two lists match by the the order.	string

Request Body:

HostTransportNode+

HostTransportNode (schema)

Name	Description	Type	Notes
_create_time	Timestamp of resource creation	EpochMsTimestamp	Readonly Sortable
_create_user	ID of the user who created this resource	string	Readonly
_last_modified_time	Timestamp of last modification	EpochMsTimestamp	Readonly Sortable
_last_modified_user	ID of the user who last modified this resource	string	Readonly
_links	References related to this resource The server will populate this field when returing the resource. Ignored on PUT and POST.	array of ResourceLink	Readonly
_protection	Indicates protection status of this resource Protection status is one of the following: PROTECTED - the client who retrieved the entity is not allowed to modify it. NOT_PROTECTED - the client who retrieved the entity is allowed to modify it REQUIRE_OVERRIDE - the client who retrieved the entity is a super user and can modify it, but only when providing the request header X-Allow-Overwrite=true. UNKNOWN - the _protection field could not be determined for this entity.	string	Readonly
_revision	Generation of this resource config The _revision property describes the current revision of the resource. To prevent clients from overwriting each other's changes, PUT operations must include the current _revision of the resource, which clients should obtain by issuing a GET operation. If the _revision provided in a PUT request is missing or stale, the operation will be rejected.	int
_schema	Schema for this resource	string	Readonly
_self	Link to this resource	SelfResourceLink	Readonly
_system_owned	Indicates system owned resource	boolean	Readonly
children	Subtree for this type within policy tree Subtree for this type within policy tree containing nested elements. Note that this type is applicable to be used in Hierarchical API only.	array of ChildPolicyConfigResource Children are not allowed for this type
description	Description of this resource	string	Maximum length: 1024 Sortable
discovered_node_id_for_create	Discovered node id to create Host Transport Node Specify discovered node id to create Host Transport Node for Discovered Node. This field is required during Host Transport Node create from vCenter server managing the ESXi type HostNode. No need to provide node_deployment_info while creating Host Transport Node for Discovered Node. If both node_deployment_info and discovered_node_id_for_create provided during Host TransportNode create payload then it will create Host Transport Node from the discovered node id provided in this field.	string
display_name	Identifier to use when displaying entity in logs or GUI Defaults to ID if not set	string	Maximum length: 255 Sortable
host_switch_spec	Transport node host switch specification This property is used to either create standard host switches or to inform NSX about preconfigured host switches that already exist on the transport node. Pass an array of either StandardHostSwitchSpec objects or PreconfiguredHostSwitchSpec objects. It is an error to pass an array containing different types of HostSwitchSpec objects.	HostSwitchSpec (Abstract type: pass one of the following concrete types) PreconfiguredHostSwitchSpec StandardHostSwitchSpec
id	Unique identifier of this resource	string	Sortable
is_overridden	Indicates if config is different than compute collection. This flag is relevant to only those hosts which are part of a compute collection which has transport node profile (TNP) applied on it. If through any update, the transport node configuration is found to be different than TNP configuration, then this field is set to true. If through any update, the transport node configuration is found to be same as TNP configuration, then this field is set to false. Updates here include all kinds of updates to the TNP, updates to transport node collection (TNC) or updates to transport node (TN)	boolean	Readonly
maintenance_mode	transport node maintenance mode desired state The property is read-only, used for querying result. User could update transport node maintenance mode by UpdateTransportNodeMaintenanceMode call.	string	Readonly Enum: ENABLED, FORCE_ENABLED, DISABLED
marked_for_delete	Indicates whether the intent object is marked for deletion Intent objects are not directly deleted from the system when a delete is invoked on them. They are marked for deletion and only when all the realized entities for that intent object gets deleted, the intent object is deleted. Objects that are marked for deletion are not returned in GET call. One can use the search API to get these objects.	boolean	Readonly Default: "False"
node_deployment_info		FabricHostNode
origin_site_id	A unique identifier assigned by the system for knowing which site owns an object This is a UUID generated by the system for knowing which site owns an object. This is used in Pmaas	string	Readonly
overridden	Indicates whether this object is the overridden intent object Global intent objects cannot be modified by the user. However, certain global intent objects can be overridden locally by use of this property. In such cases, the overridden local values take precedence over the globally defined values for the properties.	boolean	Readonly Default: "False"
owner_id	A unique identifier assigned by the system for the ownership of an object This is a UUID generated by the system for knowing whoes owns this object. This is used in Pmaas	string	Readonly
parent_path	Path of its parent Path of its parent	string	Readonly
path	Absolute path of this object Absolute path of this object	string	Readonly
realization_id	A unique identifier assigned by the system for realizing intent This is a UUID generated by the system for realizing the entity object. In most cases this should be same as 'unique_id' of the entity. However, in some cases this can be different because of entities have migrated thier unique identifier to NSX Policy intent objects later in the timeline and did not use unique_id for realization. Realization id is helpful for users to debug data path to correlate the configuration with corresponding intent.	string	Readonly
relative_path	Relative path of this object Path relative from its parent	string	Readonly
remote_path	Path of the object on the remote end. This is the path of the object on the local managers when queried on the PMaaS service, and path of the object on PMaaS service when queried from the local managers.	string	Readonly
resource_type	Must be set to the value HostTransportNode	string
tags	Opaque identifiers meaningful to the API user	array of Tag	Maximum items: 30
unique_id	A unique identifier assigned by the system This is a UUID generated by the GM/LM to uniquely identify entites in a federated environment. For entities that are stretched across multiple sites, the same ID will be used on all the stretched sites.	string	Readonly

Example Request:

PATCH https://<nsx-mgr>/policy/api/v1/infra/sites/<site-id>/enforcement-points/<enforcementpoint-id>/host-transport-nodes/ { "display_name": "VDS-TransportNode" "description": "NSX configured VDS Transport Node", "resource_type": "HostTransportNode", "host_switch_spec": { "host_switches": [ { "host_switch_id": "50 24 3c 5b a4 42 56 9e-c1 7e 5d 1d bd 52 61 14", "host_switch_type": "VDS", "host_switch_mode": "STANDARD", "host_switch_profile_ids": [ { "key": "UplinkHostSwitchProfile", "value": "0a26d126-7116-11e5-9d70-feff819cdc9f" } ], "uplinks": [ { "vds_uplink_name": "Uplink 1", "uplink_name": "uplink-1" } ], "ip_assignment_spec": { "resource_type": "AssignedByDhcp" }, "transport_zone_endpoints": [ { "transport_zone_id": "3c930f4e-1817-418d-aed6-d73f6b309e98", "transport_zone_profile_ids": [ { "resource_type": "BfdHealthMonitoringProfile", "profile_id": "52035bb3-ab02-4a08-9884-18631312e50a" } ] } ] } ], "resource_type": "StandardHostSwitchSpec" }, "maintenance_mode": "DISABLED", "node_deployment_info": { "os_type": "ESXI", "os_version": "7.0.0", "managed_by_server": "10.184.158.114", "discovered_node_id":"c59b9674-ce9b-4e52-9e8c-6bd1c6301ad3:host-13", "ip_addresses": [ "10.193.58.122" ] } }

Successful Response:

Response Code:

200 OK

Response Headers:

n/a

Response Body:

n/a

Required Permissions:

crud

Feature:

nodes_transport_nodes

Patch a Host Transport Node

Request:

TransportNodeUpdateParameters (schema)

HostTransportNode (schema)

Example Request:

Successful Response:

Required Permissions:

Feature:

Additional Errors: