The VNF deployment is initiated as a service request either originating from the ESC portal or the northbound interfaces.
The service request comprises of XML payloads. ESC supports the following deployment scenarios:
Before you deploy the VNFs, you must ensure that the resources are available on VMware vCenter, or you must create these resources.
See Managing Resources Overview. During a deployment, ESC looks for the deployment details in the deployment data model. For more information on the deployment
data model, see Cisco Elastic Services Controller Deployment Attributes.
Deploying VNFs on multiple VIMs is not supported on VMware vCenter.
A single ESC instance only supports one vCenter Distributed Switch (vDS):
ESC only supports a default resource pool. You cannot add or create resource pools. When you see the error message "Networking
Configuration Operation Is Rolled Back and a Host Is Disconnected from vCenter Server", it is due to a vCenter's limitation.
The auto-select for datastore works as follows:
A vDS contains one or many ESXi hosts that are clustered.
If the ESXi hosts are under one compute cluster, "Automation Level" must be set to "Manual" if DRS is ON.
Clustered Data stores are not supported.
If the hosts are clustered, only flat data stores under the cluster or under the datacenter are supported.
After every redeploy as part of recovery on VMware vCenter, the VM’s interface(s) will have different MAC addresses.
ESC selects a host first. If deployment is cluster targeted, host will be selected based on the ratio of number of VMs against
computing-host’s capacity. Otherwise, host is selected as requested for host targeted deployment.
From the host, datastore is picked based on its free space.
Passing OVF Properties to a VM
As a part of deploying a VNF on VMware vCenter, you can pass the name value pair as OVF property to the VM. To pass these
configurations while deploying a VNF, you must include additional arguments in the deployment data model template.
A sample configuration is as follows:
Deploying VNFs on Multiple Virtual Data Centers (Multi-VDCs)
A Virtual Data Center (VDC) combines virtual resources, operational details, rules, and policies to manage specific group
requirements. A group can manage multiple VDCs, images, templates, and policies. This group can allocate quotas and assign
resource limits for individual groups at the VDC level.
To view the list of VDCs that are available and on the ESC portal, choose Datacenters.
Before you Begin
Before you deploy VNFs on multiple VDCs, ensure that the following conditions are met:
Verify that a standard external network spanning both VDCs is available for the ESC to ping the deployed VMs.
Verify that at least one management interface on the VMs is connected to the external network.
Verify that the VDC is present in the vCenter.
ESC assumes all required resources to be created in VDC are out of band and present in the VDC.
Currently, ESC can deploy in any VDC present in a vCenter. There is no scoping or restriction of VDCs that ESC can deploy
When you deploy a VNF, you must specify the virtual datacenter locator name on which the VNF needs to be provisioned.
A locator element is introduced in deployment request to create and delete resources.
The locator element contains:
a datacenter name tag—to specify the target VDC for the resource (Deployment, Image, Network and Subnets).
switch_name—to specify the target VDS to associate the network with.
Using the locator element,
An image or a template can be created on another VDC by providing the datacenter attribute within the locator. For example,
A network can be created and deleted from a VDC.
If the network is part of unified deployment, then the datacenter attribute is taken from the deployment attribute in deployment
Cisco Elastic Services Controller Portal allows you to choose the VDC on which the VM is provisioned. When you are creating
a service request, you can choose the VDC on which this VM is provisioned. For more information on deploying VNFs on a VDC,
The default_locators container in ESC operational data shows default locators configured in ESC.
The default_locators container is not displayed if there are no locators configured.
Sample operational data is as follows:
/opt/cisco/esc/confd/bin/netconf-console --port=830 --host=172.16.0.1 --user=admin --privKeyFile=/var/confd/homes/admin/.ssh/confd_id_dsa --privKeyType=dsa --get -x "esc_datamodel/opdata"
<?xml version="1.0" encoding="UTF-8"?><rpc-reply xmlns="urn:ietf:params:xml:ns:netconf:base:1.0" message-id="1">