Cisco Virtual Topology System (VTS) 2.3 User Guide

Book Contents

Find Matches in This Book

Available Languages

Download Options

Book Title

Cisco Virtual Topology System (VTS) 2.3 User Guide

Chapter Title

Managing Inventory

PDF - Complete Book (2.54 MB) PDF - This Chapter (1.26 MB)
View with Adobe Reader on a variety of devices

Results

Updated:: October 4, 2016

Chapter: Managing Inventory

Creating Authorization Groups
Performing Auto Discovery
Importing Inventory using CSV File
Viewing the Network Topology
Viewing Network Inventory
- Adding Fabric Connection
- Synchronizing Configuration
Viewing Host Inventory
- Installing OpenStack Host Agent and VTF using GUI
Viewing the IOS XRv to VTF Mapping
Migrating from vPC to ESI
Redeploying Device Inventory

Managing Inventory

You can discover the devices using the auto discovery option, and import the details into Cisco VTS. You can also manually create a CSV file with device details, in a prescribed format, and import it into Cisco VTS.

You can create authorization groups and assign devices you import into Cisco VTS, to these groups. Authorization groups are used to group devices with the same credentials (i.e. usernames and passphrases). Once the authorization groups are created, all the devices under these groups may be accessed without specifying the credentials every time they are accessed.

If the same credential are used for accessing all devices, one authorization group can be used. If the credentials are different for different devices, multiple authorization-groups (as many as username/passphrase pairs used by devices) need to be created.

When you do a manual import of devices, the CSV file that is used to import inventory details links the authorization group with a specific device. The applicable authorization group should be used for corresponding device entry in the CSV file.

This chapter has the following sections:

Creating Authorization Groups
Performing Auto Discovery
Importing Inventory using CSV File
Viewing the Network Topology
Viewing Network Inventory
Viewing Host Inventory
Viewing the IOS XRv to VTF Mapping
Migrating from vPC to ESI
Redeploying Device Inventory

Creating Authorization Groups

Authorization Group is used by Cisco VTS to authenticate or to log in to the device.

To create an authorization group:

Step 1

Go to Inventory > Authorization Group. The Inventory / Authorization Group window appears.

Step 2

Click Add (+) icon. The Add Auth Group popup window appears.

Enter the following details, and click Save:

Auth Group Name—The authorization group name.
Controller User Name—This is the VTC administrative user name.
Device User Name—This is the login user name for the device.
Passphrase—This is the login passphrase for the device.

The authorization group gets added to the Groups table.

To edit an authorization group, select the Auth Group Name check box and click the Edit icon.

To delete an authorization group, select the Auth Group Name check box and click the delete (X) icon.

Performing Auto Discovery

In the auto discovery option, Cisco VTS automatically discovers the network topology in the data center . You can modify the device details after discovery is complete and add details to the inventory.

The auto discovery option has the following prerequisites:

Link Layer Discovery Protocol (LLDP) has to be enabled on leafs, spine, DCI, and computes. See documentation for the respective devices for details about how to enable LLDP on these devices.

Note

As part of Topology discovery, once the compute hosts have been discovered using LLDP, you need to add the username and passphrase to each host entry. This update is required for installation of the host-agent (in case of OpenStack) and any subsequent passphrase change via VTS GUI to go through.

A seed device has to be identified, and the IP should be provided. The seed IP is that of one of the leaf or spine devices.
All devices must have a common set of credentials. These credentials will be used during the discovery process. See Managing Inventory for more information. The credentials must be of the appropriate privilege level on the devices.

To perform auto discovery:

Step 1

Go to Inventory > Import and Discovery. The Inventory / Discovery window appears.

Step 2

Click Discovery . The Discover Network popup window appears.

Step 3

Enter the Seed Device IP.

Step 4

Enter the Seed Device User Name.

Step 5

Enter the Seed Device Passphrase.

Step 6

Click Start Discovery.

Step 7

After the auto discovery is complete, the details are displayed in the table. You can use the Bulk Edit option to modify the device details for more than one devices, before you add the devices to inventory, using the Add to Inventory button. To avoid errors, make sure you review the details in the table before you add the devices to Inventory. The Cisco VTS discovery log file is under /var/vts/log. Check for any errors/exceptions in this log file.

If, during auto discovery, Cisco VTS fails to log into any of the nodes, then the information of the nodes along with the error details can be found in /var/vts/log.

When you add devices to the inventory using the Add to Inventory button, all devices in the table are added to inventory.

Note

When two ToRs are configured in VPC and no dual-homed host (connected to those ToRs) is in the VTS inventory, VTS does not correctly identify the VPC. You must add the dual-homed host connected to the ToRs in VPC to the VTS inventory, before provisioning a port on a host connected to the ToRs in VPC.

Note

Different ESI groups/domains must have different ES-id or system MAC. In other words, duplicate ES-id and system MAC are not allowed among ESI groups. This needs to be guaranteed by providing correct Day Zero configurations for ESI on Cisco Nexus 9000 switches.

Importing Inventory using CSV File

The device inventory details need to be uploaded using a CSV file, in the prescribed format. The inventory file is used to define device mappings. If the format is incorrect, Cisco VTS displays an error and provides the details of the error. After a successful import, the topology gets displayed based on the mapping specified in the file.

When adding device to inventory, in the list of devices connected to a switch, the interface field should contain the interface name for network devices and the MAC address for servers.

Note

You should be an admin user to download or upload the CSV file. Also, if you are uploading a CSV file for the first time and there are issues uploading the file, then only the partial information is uploaded. You need to delete all the devices from network inventory and re-upload the file after fixing the issues. Or else, you may encounter problems due to the partial upload.

To download a sample inventory file, click Download Template.

Note

In a VMware environment, each time you add a leaf, you must create a corresponding VMware vSphere Distributed Switch (VDS). See the Important Notes Regarding VMware vSphere Distributed Switch (VDS) section for details.

Step 1	Go to Inventory > Import and Discovery. The Inventory / Discovery window appears.
Step 2	Click Import CSV The Import Devices popup window appears.
Step 3	Click Browse to choose the CSV file. Click Download Template to download a sample CSV file. The CSV file has the following fields: device-name—The device host-name (leaf, spine, DCI) device-ip—IP address for the device (leaf, spine, DCI) device-platform—Can be Cisco Nexus 9000, Cisco Nexus 7000 etc based on the device that is part of the network. device-role—The role that a particular device plays in the data center. leaf—If the device plays the role of a Leaf in the data center. border leaf—If the device plays the role of a Border Leaf in the data center. spine—If the device plays the role of a Spine in the data center. spine-rr—If the Spine plays the role of a Route Reflector in the data center. dci—If the device plays the role of a DCI in the data center. group-tag—Identifier for the group. port-name—Physical port connectivity (local interface) connection-type—server (if connected to compute host); fabric (if connected to another leaf, spine, DCI devices). server-id—Host-name or IP address of the connected device based upon what is configured on the actual host. server-type—virtual-server for computes; baremetal for connections to spine, DCI. interface-name—Physical port connectivity (interface of the connected device) auth-group—Authorization group name, created as part of initialization, with correct credentials.
Step 4	Browse for the CSV file, and click Import.
Step 5	After the import operation complete, the details are displayed in the table. You can use the Bulk Edit option to modify the device details for more than one devices, before you add the devices to inventory, using the Add to Inventory button. To avoid errors, make sure you review the details in the table before you add the devices to Inventory. Click the device name to open the connected devices popup window, which gives details about connected devices for a device.

Viewing the Network Topology

Topology window provides a view of the data center fabric controlled by Cisco VTS. It displays the leafs, spines, border leafs, DCI, hosts, as well as the software VTEPs. You can get a tenant-based topology view using this feature.

To view the network topology:

Step 1

Go to Inventory > Topology. The Inventory / Topology window appears.

Step 2

Select the tenant for which you need to view the topology, from the Select Tenant drop down list.

The topology is displayed in the Topology window. You can use the following buttons to control the display:

Select node mode
Move mode
Zoom in / Zoom out / Zoom Selection
Fit Stage
Full Screen mode

Hover the mouse cursor over the Topology Setting icon to view Topology Setting popup, where you can change the display icon appearance, and display color.

Note

In case of FEX or VPC, if no host is connected, Cisco VTS will not show the VPC or FEX in the Topology. Also, you might encounter errors.

Different types of links will be represented as shown in legends.

Hover the mouse cursor over the link to view the Info popup, which gives the information about the link.

Viewing Network Inventory

The network inventory table displays details about the devices which have been added to the inventory.

To view the network topology:

Go to Inventory > Network Inventory. The Inventory / Network Inventory window appears with the Network Inventory table displayed.

The following details are displayed:

Device Name
Admin State
IP Address
Auth Group
Device Platform
Device Role
Group Tag
BGP-ASN
Loopback Interface Number
Loopback Interface IP

Note

You need to add the BGP-ASN information, loopback interface number, and loopback interface id of every device prior to starting any provisioning.

You can add network devices via the Network Inventory table. To do this, click the Add (+) icon, and provide the details. You can use this option to add devices to the inventory.

To edit network device, select the device you want to edit and click the Edit icon.

To delete network devices from the Network Inventory table, select the device you want to delete and click the Delete (X) icon.

If there is problem in deleting device, you need make sure that fabric link is cleaned up manually. For example, when Device 1 is connected to Device 2, Inventory has two devices and two fabric links (this can be seen in Fabric Connection tab in Network Inventory)—one from Device 1 to Device 2, and the other from Device 2 to Device 1. While deleting Device 1 from network inventory, cleanup is done for Fabric link Device 1 to Device 2 and for the device from the inventory. The link Device 2 to Device 1 has to be cleaned up manually before you delete.

It is important that you remove the resource pool before deleting a device.

You need to discover the devices and add them to the inventory before you bring up the IoS XRv. If you do these tasks simultaneously, you might encounter errors.

To recalculate the inventory topology for a particular device, click the redeploy button. See Redeploying Device Inventory for more details.

Adding Fabric Connection

To add fabric connection:

Step 1	Go to Inventory > Network Inventory. The Inventory / Network Inventory window appears with the Network Inventory table displayed.
Step 2	Click Fabric Connection tab, then click Add (+) icon. The Add Fabric Connection popup window appears.
Step 3	Enter the necessary details and click Save.

Synchronizing Configuration

You can check if the device state is in sync with VTS configuration database.

To check whether the configuration is in sync:

Note

This operation can be done only on a device that has the Admin State as Unlocked. If Admin state is Locked, you must change the Admin State to Unlocked, and then do the check-sync operation. Also, the out-of-sync-commit behavior in System Settings must be set to Reject for this feature to be enabled.

Step 1

Go to Inventory > Network Inventory. The Inventory / Network Inventory page displays the Network Inventory table .

Step 2

Click the Config Sync link under the Sync column, for the device.

If the configuration is in sync with the VTS database, the green In Sync icon is displayed.

If the configuration is not in sync with the device, you will get an option to compare the configuration. You can compare the actual device configuration with the VTS configuration, and can view the diff in compare-config popup. You can also use the copy to clipboard option to copy and paste the configuration in a text editor. To synchronize the configuration, you can use the following options:

sync-from—Synchronize the VTS configuration by pulling configuration from the device.
sync-to—Synchronize the VTS configuration by pushing configuration to the device.

Note

If switchname (switch hostname) is changed in the switch CLI, the sync to option will not work. The switchname has to be the same as the value in the VTS inventory.

Viewing Host Inventory

You can view the details of the hosts connected to the switches.

The Host Inventory has the Install Capabilities button. You can select the host from the list, and click Install Capabilities. If the host is connected to a physical VTEP (ToR), it installs the host agent in case of an OpenStack environment. If the capability is Virtual, clicking the Install Capabilities button installs the VTF VM on that host. See Installing OpenStack Host Agent and VTF using GUI.

To view host inventory details:

Step 1

Go to Inventory > Host Inventory. The Inventory / Host Inventory window appears

Step 2

Select the device from the Select Device drop-down list. The following details are displayed:

Host Name
Host Type—baremetal is a server that is not managed by a VMM. A virtual-server is one which is managed by a VMM.
Host Interface—The interface of the bare metal/virtual server connected to the TOR [Example: eth1 or eth2 etc].
Host IP Address
Device Port Name—The interface of the TOR which is connected to the baremetal/virtual server [Example: ethernet 1/1 etc].
User Name
Capability—The host capability - Physical or Virtual.
Installation Status—Shows the installation status.

Note

If you have converted a Virtual server to Baremetal, before you upgrade, you will not be seeing details like Installation Status and Capabilities, after the upgrade.

To add a host from the table, select the Host Name check box corresponding to the device and click the Add (+) icon. Then provide the necessary details.

To edit a host from the table, select the Host Name check box corresponding to the device and click the Edit icon.

To delete a host from the table, select the Host Name check box corresponding to the device and click the click the Delete (X) icon.

Installing OpenStack Host Agent and VTF using GUI

You can use the Install Capabilities button to install the Host Agent and VTF on a host based on whether it is a virtual or physical server.

Step 1

Go to Inventory > Host Inventory. The Inventory / Host Inventory window appears

Step 2

Click + to add a host. You may also edit a host and modify the parameters to enable installation of physical or virtual capabilities.

If you click + (Add) the Add New Host popup is displayed. Enter the following details.

If you choose to edit an existing host, the following windows are displayed depending upon the host type:
- Host Details:
  - Host Name
  - Host Type
  - Host Interface
  - Host IP Address
  - Device Port Name
  - Capability—Specify whether it is a virtual-switch ir not.
  - Vmm IP Address
  - User
  - Passphrase
- Common Parameters—These are displayed only if the host capability is virtual-switch.
  - VTF IP—The IP address of the VTF.
  - Subnet Mask
  - Gateway
  - Underlay Bridge—Name of the underlay network portgroup/bridge on the binding-host to which VTF is attached.
  - Internal Bridge—Name of the tenant network port group/bridge on the binding-host to which VTF is attached.
  - User
  - Passphrase

Ensure that you review the tooltips for important information about the entries.

Click on the info icon adjacent to the Host Name. Host Summary table popup appears.

If Virtual Capability is virtual-switch, both Additional Host Details and VTF Details can be viewed in the Host Summary table.

If Virtual Capability is no-virtual-switch, only Additional Host Details can be viewed in the Host Summary table.

Step 3

Click Install Capabilities. Based on the host type, it installs the host agent / VTF on the host.

See the Status column for the installation status. The VMM type is also shown once the capabilities get installed.

Note

While installing capabilities for a host, if the selected VMM is openstack-liberty-centos and the mode for VTF deployment is vtf-vhost, VTF installation has to be done manually.

Viewing the IOS XRv to VTF Mapping

Go to Inventory > Virtual Forwarding Groups. The Inventory / Virtual Forwarding Groups window appears.

The canvas displays the number of VTFs that are attached to the IOS XRvs . The table on the right hand side shows the VTFs.

Migrating from vPC to ESI

This section provides details about the generic procedure to migrate from Virtual Port Channel (vPC) to Ethernet Segment Identifier (ESI).

Note

Before you begin, ensure that the following TCAM regions are carved on Cisco Nexus 9000 series switch:

hardware access-list tcam region vpc-convergence 256
hardware access-list tcam region arp-ether 256

To migrate from vPC to ESI:

Step 1

In case of IOS XRv HA, bring down the IOS XRv.

Step 2

Upgrade VTS to a version which supports ESI.

Step 3

Perform a port-detach operation to remove the relevant service configurations provisioned on the TORs.

Step 4

Remove the TORs from Admin Domain in VTS GUI.

Step 5

If the TCAM regions, as mentioned above, are not already carved on Cisco Nexus 9000 series switch, add the lines and save as running config.

hardware access-list tcam region vpc-convergence 256
hardware access-list tcam region arp-ether 256

Note

Do not reboot device (as the TOR will be rebooted in the next step).

Step 6

Upgrade TORs to a new Cisco Nexus 9000 image, which has ESI feature. This will automatically cause device to reboot.

copy run start
install all nxos bootflash:/nxos.7.0.3.ITC5.0.100.bin (or newer n9k image)

Step 7

Upgrade Cisco ASR 9000 series DCIs to an ESI supporting image.

Step 8

Once the setup is up then remove feature VPC and configure ESI on the required TORs that you are planning to convert to ESI.

Remove VPC	no feature vpc
Remove other VPC related configuration under port channel and Ethernet Interfaces
Remove secondary interface from loopback	interface loopback0 no ip address 44.44.44.44/32 secondary
Enable ESI	evpn esi multihoming
Create nve	interface nve1 no shutdown source-interface loopback0 host-reachability protocol bgp
Enable core links	interface Ethernet1/35 Description " Connected with Spine" no switchport evpn multihoming core-tracking <<< Add here ip address 16.1.1.2/24 ip router ospf 100 area 0.0.0.0 ip pim sparse-mode no shutdown
Add Ethernet-segment and system-mac address in the port-channel	interface port-channel220 switchport mode trunk switchport trunk allowed vlan none ethernet-segment 220 system-mac eeee.1111.2222
Apply the channel group to the TORs interface which are connected to compute.	interface Ethernet1/5 switchport trunk allowed vlan none channel-group 220 mode active
Verify whether the ESI is up.	tor1# show nve ethernet-segment ESI Database ---------------------------------------- ESI: 03aa.bbcc.ddee.ee00.002d, Parent interface: port-channel30, ES State: Up Port-channel state: U NVE Interface: nve1 NVE State: Up Host Learning Mode: control-plane Active Vlans: 1001 DF Vlans: 0-4095 Active VNIs: 30001 Number of ES members: 1 My ordinal: 0 DF timer start time: 00:00:00 Config State: config-applied DF List: 1.1.1.1 ES route added to L2RIB: True EAD routes added to L2RIB: True ----------------------------------------

Step 9

On Cisco VTS, perform a sync-from operation for the TORs that have ESI enabled.

Step 10

Redeploy inventory from Cisco VTS only for devices that have new ESI configuration. This is to make sure that Cisco VTS recognizes ESI configuration on Cisco Nexus 9000 series devices. See Redeploying Device Inventory for details.

Step 11

Remove the peer links between previous VPC peer TORs (Inventory > Network Inventory > Fabric Connection).

Step 12

Create a device group (Resource Pools > VLAN Pool), and add the group of ESI enabled devices to the group. Make sure a corresponding VLAN pool/range is also created for the device group.

Step 13

Add the ESI device group to appropriate functional groups in Admin Domain, and also disable ARP suppression in Admin Domain.

Step 14

Upgrade IOS XRv to the latest image.

Step 15

Perform a port-attach from VTS.

Redeploying Device Inventory

You can use the Redeploy feature to recalculate the inventory topology for a particular device. This is important in the context of VPC and ESI.

You need to Redeploy the inventory when device day zero configuration changes for:

VPC or ESI. For example, vpc id for a port-channel is changed
port-channel or ether-channel

Redeploy triggers the inventory for a device again. Since inventory reads the data from the device model in the database it is important to perform sync-from before doing a Redeploy.

Note

Redeploy function is different from the sync-from function. Sync-from gets the configurations from the device and updates it in the device model in the database. However, it does not recalculate the topology. That is, the topology would still show old information/configuration. Redeploy recalculates the inventory topology. After you perform a Redeploy, the topology will be updated with the modified configuration.

To redeploy device inventory:

Step 1

Go to Inventory > Network Inventory, perform a sync-from for the device for which the configuration has changed. See Synchronizing Configuration for more details.

Step 2

Select the device, click Redeploy.

Note

Redeploy just recalculates the inventory. Existing ports/VMs belonging to old device configurations, would not be updated or redeployed. You might need to delete and recreate the existing ports. We recommend that you use redeploy only if there are no existing ports/router/router interfaces.

Was this Document Helpful?

Feedback

Contact Cisco

Open a Support Case
(Requires a Cisco Service Contract)