Topology
The topology window shows color encoded nodes and links that correspond to various network elements including switches, links, fabric extenders, port-channel configurations, virtual port-channel, and more. For each of the elements, you can hover over to fetch some more information. Additionally, by clicking on the node for a switch or line for a link, a slide-out panel flies out from the right. This panel shows more detailed information about either the switch or link. Multiple tabs can be opened simultaneously and are intended to function side-by-side for comparison and troubleshooting.
Status
The color encoding of each node and link corresponds to its state. The color states are:
-
Green—Indicates that the element is in good health and functioning as intended.
-
Yellow—Element is in warning state and requires attention to prevent any further problems.
-
Red—Element is in critical state and requires immediate attention.
-
Gray—Indicates lack of information to properly identify the element or the element has just been discovered.
Scope
You can search the topology based on the scope. The default scopes available from the SCOPE drop-down list is: DEFAULT_LAN
The following search options are available for DEFAULT_LAN:
-
Quick Search
-
Tags
-
Host name (VDP)
-
Host name (vCenter)
-
Host IP
-
Host MAC
-
Segment ID
-
Multicast Group
-
VXLAN ID (VNI)
-
VLAN
-
VSAN ID/Name
-
FabricPath
-
VXLAN OAM
Searching
When the node number is large, it quickly becomes hard to find the intended switches and links. You can quickly find switches and links by performing a search. You are also able to search for VM tracker and generic setups. Searching feature enables you to see which leaf the host is connected to.
The following searches are available:
Quick Search
Enables you to search for devices by name, IP address, mode, serial number, and switch role. This search returns immediately and results are highlighted as the user types.
To perform a search for multiple nodes and links, you can separate multiple keywords by a comma.
For example, a search can look like: ABCD12345, N7K, sw-dc4-12345, core, 172.23.45.67. Wildcards are also supported.
If you partially know a serial number or name, you can build a search like so: ABCD*, sw*12345, core, *23.45.*.
If you want to limit your scope to a particular field, you can for example do: name=sw*12345, serialNumber=ABCD12345.
The following fields are available to search on:Name, ipAddress, serialNumber, model, and switchRole.
Host name (vCenter)
Enables you to search for hosts by using vCenter.
Host IP
Search talks to your switches in the scope to find any hosts (VMs or bare metal) that match your given IP address.
The Host IP search supports both IPv4 and IPv6 addresses.
Host MAC
Search talks to your switches in the scope to find any hosts (VMs or bare-metal) that match your given MAC address.
Multicast Group
Search by a given multicast group.
![]() Note |
Multicast group is limited to VXLAN context (VTEP switches) for searching the switches to get VNIs associated with this multicast address. Click the Details link next to the search box to get the detailed multicast address table. The table displays switches which have the searched Multicast address configured on them along with associated VNI, VNI status, and mapped VLAN. |
VXLAN ID (VNI)
Search by a given VNI. Click the Details link next to the search box to get the detailed VNI table. The table displays the switches which have searched VNI configured on them along with the associated multicast address, VNI status, and mapped VLAN.
VLAN
Search by a given VLAN ID. VLAN search provides the search for the VLAN configured on the switch or the links. If STP is enabled, then it provides information that is related to the STP protocol and the STP information for links.
VSAN ID/Name
Search by a given VSAN ID. VSAN search provides the search for VSAN configured on the switch or the links. In order to view the STP details associated with the VSAN, click STP Details link.
This shows the STP details, if STP is enabled. If the link is blocked, it is marked as red, green in case of a forwarding link, and orange if the link is blocked for one VSAN range and forwarding for the other VSAN range.
This search is applicable to both the default LAN and SAN scopes.
FabricPath
Search on the FabricPath topology ID, typically 0 or 1. After search, FabricPath topology links are displayed with purple colored links. You can use the FabricPath Panel below the search box to fetch various FabricPath graphs.
Click the drop-down arrow next to FabricPath Panel, select a below type of graph:
-
Multi destination—Select the switch from the Anchor drop-down list, and select the graph ID from the Graph ID drop-down list to get multi destination or broadcast graph.
-
Reachability—Select the switch in the From drop-down list to get the reachable devices in FabricPath topology from the selected switch.
-
Unicast—Select the switch from the Source and Destination drop-down list to get equal cost multi paths from a source to destination switch in FabricPath topology.
-
Multicast—Select from the Anchor,Ftag ID, and IGMP Addr drop-down list to get the FabricPath multicast tree. The selected anchor switch is the root of the tree.
VXLAN OAM
You can track details such as reachability and actual path of the flows in a VXLAN EVPN based-fabric topology by choosing the VXLAN OAM option from the Search drop-down list or by entering VXLAN OAM in the Search field. This displays the Switch to switch and Host to host tabs. DCNM highlights the route on the topology between the source and destination switch for these two options.
The Switch to switch option provides the VXLAN OAM ping and traceroute test results for the VTEP-to-VTEP use-case. Provide the following values to enable search by using the Switch to switch option:
-
From the Source Switch drop-down list, choose the source switch.
-
From the Destination Switch drop-down list, choose the destination switch.
-
From the VRF drop-down list, choose or enter the VRF details.
-
Check the All Path Included check box to include all the paths in the search results.
The Host to host option provides the VXLAN OAM pathtrace results for the exact path that is taken by a given flow from the VTEP or switch that is connected to the source host to VTEP or switch that is connected to the destination host. For the Host to host use-case, there are two suboptions:
-
VRF or SVI for a network is instantiated on the switches in the VXLAN EVPN fabric. In such a scenario, the IP address information of the end hosts is required.
-
Layer 2 configuration for a given network is instantiated on the switches in the VXLAN EVPN fabric. In such a scenario, both the MAC and IP address information of the end hosts are required.
Provide the following values to enable search using the Host to host option:
-
In the Source IP field, enter the IP address of the source host.
-
In the Destination IP field, enter the IP address of the destination host.
-
In the VRF field, choose VRF from the drop-down list or enter the VRF name that is associated with the hosts.
-
(Optional) In the Source Port field, choose Layer 4 source port number from the drop-down list or enter its value.
-
(Optional) In the Destination Port field, choose destination port number or enter its value.
-
(Optional) In the Protocol field, choose the protocol value from the drop-down list or enter its value. This is the Layer 4 protocol, usually TCP or UDP.
-
Click the Interchange/Swap Source and Destination IPs (and MACs if applicable) icon to interchange the source and destination IP addresses. This interchange allows a quick trace of the reverse path without reentering the host IP addresses or MAC addresses.
-
Check the Layer-2 only check box to search the VXLAN-EVPN fabric that is deployed in Layer 2 only mode for some networks, that is, Layer 2 VNIs. Note that no SVIs or VRFs should be instantiated in the fabric for these networks when you use this search option.
Enter values for the following additional fields:
Show Panel
You can choose to view your topology based on the following options:
-
Auto Refresh—Check the check box to automatically refresh the topology.
-
Switch Health—Check the check box to show the switch health status.
-
FEX—Check the check box to show the Fabric Extender.
FEX feature is available on LAN devices only. Therefore, checking this check box displays only the Cisco Nexus Switches that support FEX.
Note
FEX is also not supported on Cisco Nexus 1000V devices. Therefore, such devices will not be displayed in the topology when you check the FEX check box.
-
Links—Check the check box to show links in the topology. The following options are available:
-
Errors Only—Click this radio button to display only links with errors.
-
All—Click this radio button to display all links in the topology.
-
Bandwidth—Check this check box to show the color coding based on the bandwidth that is consumed on links.
-
-
OTV—Check the check box to show Overlay Transport Virtualization (OTV) topology with cloud icon and dotted links from the OTV edge devices. Hovering the mouse over the cloud and links shows the relevant information for OTV topology such as control group, extended VLANs, and so on. The OTV search box appears below the filter box which can be used to search the shown OTV topology that is based on Overlay ID and Extended VLAN ID. Searched virtual links based on the Overlay ID and Extended VLAN ID are marked green.
A Details link appears after checking the OTV box. Clicking the links shows the OTV topology data. The Overlay Network column shows whether the particular topology is multicast or unicast based. The Edge Device column gives the edge switches in the particular OTV topology. Other columns give the corresponding overlay interface, extended VLANs, join interface and the data group information.
-
UI controls—Check the check box to show or hide the various controls on the topology screen.
-
Compute—Check the check box to enable the compute visibility on the topology screen.
-
Refresh—You can also perform a topology refresh by clicking the refresh icon in the upper right of this panel.
Layouts
The topology supports various different layouts and a Save Layout option which remembers how you positioned your topology.
-
Hierarchical/Hierarchical Left-Right—Provides an architectural view of your topology. Various Switch Roles can be defined that will draw the nodes on how you configure your CLOS topology.
Note
When running a large scale setup, being able to easily view all your switches on a leaf-tier can become difficult. To mitigate this, DCNM splits your leaf-tier every 16 switches.
-
Random—Nodes are placed randomly on the screen. DCNM tries to make a guess and intelligently place nodes that belong together in close-proximity.
-
Circular/Tiered-Circular—Draws nodes in a circular or concentric circular pattern.
-
Custom saved layout—You can drag nodes around to your liking. Once you have the positions as how you like, you can click the save button to remember the positions. Next time you come to the topology, DCNM will draw the nodes based on your last saved layout positions.
Before a layout is chosen, DCNM checks if a custom layout is applied. If a custom layout is applied, DCNM uses it. If a custom layout is not applied, DCNM checks if switches exist at different tiers, and chooses Hierarchical layout or Hierarchical Left-Right layout. Force-directed layout is chosen if all other layouts fail.
Zooming, Panning and Dragging
You can zoom in and zoom out using the controls that are provided at the bottom left of the screen or using the mouse wheel.
To pan, click and hold anywhere in the whitespace and drag up, down, left, or right.
To drag switches, click, hold, and move around in the whitespace region of the topology.
Switch Slide-Out Panel
You can click the switch to display the configured switch name, IP address, switch model, and other summary information like status, serial number, health, last polled CPU utilization, and last polled memory utilization.
Beacon
This button will only by shown for switches that support the beacon command. Once the beaconing starts, the button will show a countdown. By default, the beaconing will stop after 60 seconds but you can stop it immediately by clicking stop beacon.
![]() Note |
The default time can be configured in server.properties. Search for beacon.turnOff.time. The time value here is in milliseconds. Note that this requires a server restart to take effect. |
Tagging
Tagging is a powerful yet easy way to organize your switches. Tags can be virtually any string and an example of tags includes: building 6, floor 2, rack 7, problem switch, and Justin debugging.
You can use the search to perform searches based on tags.
More Details
Click Show more details and the detailed information appears in the Switch Dashboard.
Link Slide-Out Panel
You can click a link to view the status and the port or switches that describe the link.
24 Hour Traffic
This features requires Performance Monitoring be turned ON. When Performance Monitoring is ON, traffic information is collected and aggregative information is displayed along with a graph showing the traffic utilization.
vCenter Compute Visualization
In virtualized environments, any kind of troubleshooting starts with identifying the network attachment point for the virtual machines. That means, means a quick determination of the server, virtual switch, port-group, VLAN, associated network switch, and physical port is critical. This requires multiple touch-points and interactions between the server and network administrator as well as reference to multiple tools (compute orchestrator, compute manager, network manager, network controller and so on).
This feature allows you to visualize the vCenter managed hosts and their leaf switch connections on the topology page. The visualization options include viewing only the attached physical hosts or only the VMs or all. The latter option displays the topology all the way from the leaf switches to the VMs including the virtual switches. The VM Search option highlights the path of the virtual machine. You can hover the mouse cursor over a host or the connected uplink that highlights key information relevant to that entity. Up to four vCenters are supported.
![]() Note |
The vCenter Compute Visualization feature is supported on both the LAN Classic and Easy Fabrics installations for the vCenter-managed computes (vCenter versions 5.5/6.0/6.5). |
Enabling vCenter Compute Visualization
Procedure
Step 1 |
From Cisco DCNM Web Client, choose Configure > Inventory > Virtual Machine Manager. The Inventory > Discovery > Virtual Machine Manger page appears. |
Step 2 |
Click the + button to add a new VMware vSphere vCenter. |
Step 3 |
Add details of the vCenter including its hostname or IP address, and credentials. vCenter version 5.5 or higher is required. After initial discovery, which will take a couple of minutes, the information received from vCenter is appropriately organized and made available on the main topology page. An additional menu item labelled Compute appears on the Show panel. |
Using vCenter Compute Visualization
Procedure
Step 1 |
From Cisco DCNM web interface, choose Topology. |
Step 2 |
Once the Compute check box in the topology view menu is selected, you have enabled the compute visibility. By default, the Host check box is selected, which means that the topology shows the VMWare vSphere ESXi hosts (servers), which are attached to the network switches. The following options are available in the Compute Visualization feature.
In the "All" mode you can see double-arrows that help you to extend a node. If you double-click on this node, you can see all the child nodes that were hidden initially. |
Step 3 |
Once the topology is shown with the additional compute visibility, through clicking a specific ESXi host, additional information is depicted. The expanded topology will display the virtual switches (both vSwitch and Distributed Virtual Switch) that are configured on the specific ESXi host. |
Step 4 |
When changing from the “Host” sub-option to the “All” sub-option, all the compute resources are expanded. This mode provides an expanded view of all the hosts, virtual switches, virtual machines that are part of the topology. If a VM is powered off, it will be shown in red color else it will be green color. |
Step 5 |
Instead of browsing through the large set of available information, to quickly focus on a specific VM, a search option is available. In the top left search bar, select the Host name (vCenter) option. |
Step 6 |
When you start typing characters, the topology is instantaneously updated to highlight the matching objects. |
Searching for vCenter-managed Virtual Machines
![]() Note |
The vCenter search is unavailable when compute visualization is not enabled. Also, this search is available only when you select the "All" option. |
Procedure
Step 1 |
From Cisco DCNM web interface, choose Topology. |
Step 2 |
In the Quick Search drop-down list, choose Host Name(vCenter), enter the name of the vCenter-managed virtual machine in the search text box, and then press the Enter key. When you search for a vCenter-managed virtual machine, the topology displays the path recursively. |
Step 3 |
Press the Enter key. |
Selecting a vCenter in Topology
Procedure
Step 1 |
From Cisco DCNM web interface, choose Topology. |
Step 2 |
In the Topology page, you can select a specific vCenter or all the available vCenters. To select a vCenter, use the Vcenter Selection drop-down list. |
Using the Virtual Machine List
Procedure
Step 1 |
From Cisco DCNM web interface, choose Topology. |
Step 2 |
Click the VM List button in the Topology page. You can click the Export button to export the list of virtual machines into a .CSV file. |
Step 3 |
You can click the name of any Virtual Machine to view additional information about the virtual machine. |
Resyncing Virtual Machines
Procedure
Step 1 |
From Cisco DCNM web interface, choose Topology. |
Step 2 |
Click the Resync button in the Topology page. |
Troubleshooting vCenter Compute Visualization
The following error page appears when there is a time out of the vCenter. This error may occur when the discovery of the vCenter is in progress.
Viewing Topology in Scale Mode
The following page shows how the topology page appears after about 200 devices are available in the topology. The topology graph is trimmed down at scale.