Information About Network Uplink Configurations
The Cisco Nexus Cloud Services Platform product family supports two types of network uplink configurations to connect to the network:
Flexible Network Uplink Configuration
The flexible network configuration offers complete flexibility to connect Cisco Nexus Cloud Services Platform to the network. This configuration type enables appropriate traffic segregation policies like VSB traffic segregation.
This configuration consists of the following features:
- Complete flexibility in terms of port configuration and usage
- Flexible building of ports into a port channel.
- Flexible assignment of a port or port channel to a VSB interface.
- Easy uplink configuration.
- Ability to achieve maximum uplink.
The default flexible network uplink configuration (see Figure 3-1) includes the following features:
- Every physical port individually forms an uplink.
- Each uplink can be configured independently.
- Each uplink can achieve a maximum uplink of 6 Gbps.
- No default redundancy for uplinks.
- You cannot bundle physical ports in a port channel.
- VSB traffic is segregated by default.
- You can manually configure a VSB interface to share a port.
Figure 3-1 Default Flexible Network Uplink Configuration
If you choose flexible configuration during the installation of the Cisco Nexus Cloud Services Platform, the default configuration is used to connect to the network. See the Cisco Nexus Cloud Services Platform Software Installation and Upgrade Guide , for more information.
You can then make changes to the default flexible network uplink configuration to suit your needs:
Figure 3-2 Assigning uplinks to flexible network configuration
Figure 3-3 Adding port channels to flexible network configuration
Static Network Uplink Configuration
In a static network configuration, the Cisco Nexus Cloud Services Platform product family is connected to the network using the following four fixed network uplink configurations:
- In configuration 1, control traffic, management traffic and data traffic share a single uplink.
- In configuration 2, control traffic and management traffic share an uplink and data traffic is a separate uplink.
- In configuration 3, control traffic and data traffic share an uplink and management traffic is a separate uplink.
- In configuration 4, control traffic, management traffic and data traffic are all on separate unlinks.
For more information about uplink configurations, see the Uplinks section.
Guidelines and Limitations
The Cisco Nexus Cloud Services Platform has the following configuration guidelines and limitations:
- A change to the uplink type does not take effect until you reload the software.
- Changing the uplink type is disruptive and leads to a service disruption.
- You can change the uplink type only once before issuing a reboot.
- Use Table 3-1 when modifying the network uplink type.
Table 3-1 Uplink Usage
|
|
1 |
When only the Cisco Nexus 1000V VSM is installed. |
2 |
When only NAM is installed. |
3 |
When the management and data traffic upstream must be separated. |
4 |
When the management and data traffic upstream must be separated and control and data traffic must also be separated. |
5 |
Flexible network uplink |
Configuring Network Uplink Types
This section includes the following topics:
Modifying the Uplink Type
You can modify the uplink type on an operational Cisco Nexus Cloud Services Platform.
BEFORE YOU BEGIN
Before beginning this procedure, you must know or do the following:
- Log in to the CLI in EXEC mode.
- You must reload the Cisco Nexus Cloud Services Platform pair in order to activate the changes made in this procedure. This procedure includes a step for reloading.
Caution To prevent loss of connectivity, you must reconfigure the uplink switches to correspond with the change made in this procedure.
SUMMARY STEPS
1. configure terminal
2. network uplink type number
3. (Optional) show network-uplink type
4. copy running-config startup-config
5. reload
DETAILED STEPS
|
|
|
Step 1 |
configure terminal Example: switch# configure terminal switch(config)# |
Places you in the CLI Global Configuration mode. |
Step 2 |
network uplink type number Example: switch(config)# network uplink type 2 switch(config)# |
Changes the uplink type for the Cisco Nexus Cloud Services Platform. The number argument range is from 1 to 4. |
Step 3 |
show network-uplink type Example: switch(config)# show network uplink type
Administrative topology id: 2
Operational topology id: 1
|
Displays the uplink configuration for verification. |
Step 4 |
copy running-config startup-config Example: switch(config)# copy running-config startup-config |
Saves the running configuration persistently through reboots and restarts by copying it to the startup configuration. |
Step 5 |
reload Example: switch(config)# reload This command will reboot the system. (y/n)? [n] y 2009 Oct 30 21:51:34 s1 %$ VDC-1 %$ %PLATFORM-2-PFM_SYSTEM_RESET: Manual system restart from Command Line Interface switch(config)# |
Migrating from a Static Network Uplink to a Flexible Network Uplink
You can migrate from a static network uplink type to a flexible network uplink type.
BEFORE YOU BEGIN
Before beginning this procedure, you must know or do the following:
- Log in to the CLI in EXEC mode.
- After you change the uplink type from static to flexible, you can configure the port channel and assign uplink assignment to a VSB manually.
- You must reload the Cisco Nexus Cloud Services Platform pair in order to activate the changes made in this procedure.
- Changing the uplink type from static to flexible is disruptive and leads to a service disruption. After you change the uplink type from static to flexible, you must save the configuration and reload for the new configuration to take effect.
- When you change the uplink type from static to flexible, the configuration for all the port channels, native VLANs, and port states is retained in the flexible network type.
SUMMARY STEPS
1. configure terminal
2. network uplink type number
3. network uplink type keyword
4. (Optional) svs-domain
5. (Optional) control uplink interface name
6. (Optional) management uplink interface name
7. copy running-config startup-config
8. reload
9. show network-uplink type
DETAILED STEPS
|
|
|
Step 1 |
configure terminal Example: switch# config terminal switch(config)# |
Enters Global Configuration mode. |
Step 2 |
network uplink type number Example: switch(config)# network uplink type 5 switch(config)# |
Changes the uplink type for the Cisco Nexus Cloud Services Platform. |
Step 3 |
network uplink type keyword Example: switch(config)# network uplink type flexible switch(config)# The command will change network-uplink type and network-uplink type cannot be changed again before reload. Change to [1-4] network-uplink type will lead to loss of native vlan config on all ports. Do you really want to proceed(yes/no)? [no] yes Note: Save the configuration and reload to bring the system with new network_uplink Example: switch(config)#)# network-uplink type flexible force Note: The command will change network-uplink type and network-uplink type cannot be changed again before reload. Change to [1-4] network-uplink type will lead to loss of native vlan config on all ports. Note: Save the configuration and reload to bring the system with new network_uplink |
Changes the uplink type for the Cisco Nexus Cloud Services Platform from static to flexible. Note You can change the network type only once. In order to change the network type again, you must reload and then change the network type. You can use the force option to skip the confirmation step. |
Step 4 |
svs-domain Example: switch(config)# svs-domain switch(config-svs-domain)# |
(Optional) Configures an SVS domain and enters SVS domain configuration mode. |
Step 5 |
control uplink interface name Example: switch(config-svs-domain)# control uplink GigabitEthernet1 switch(config-svs-domain)#
|
(Optional) Changes the default control traffic interface name. Interface names can be Gigabit Ethernet or Portchannel. |
Step 6 |
management uplink interface name Example switch(config-svs-domain)# management uplink GigabitEthernet2
|
(Optional) Changes the default management traffic interface name. Interface names can be Gigabit Ethernet or Portchannel. |
Step 7 |
copy running-config startup-config Example: switch(config)# copy running-config startup-config |
Saves the running configuration persistently through reboots and restarts by copying it to the startup configuration. |
Step 8 |
reload Example: switch(config-svs-domain)# reload This command will reboot the system. (y/n)? [n] y 2011 Oct 27 10:26:30 switch %PLATFORM-2-PFM_SYSTEM_RESET: Manual system restart from Command Line Interface |
|
Step 9 |
show network-uplink type Example: switch(config)# show network uplink type Administrative topology id: flexible Operational topology id: flexible |
Displays the uplink configuration for verification. |
Migrating from a Flexible Network Uplink to a Static Network Uplink
You can migrate from a flexible network uplink to a static network uplink.
BEFORE YOU BEGIN
Before beginning this procedure, you must know or do the following:
- Log in to the CLI in EXEC mode.
- This procedure is disruptive because both the active and standby Cisco Nexus Cloud Services Platforms should be reloaded together.
- You must reload the Cisco Nexus Cloud Services Platform pair to activate the changes made in this procedure. This procedure includes a step for reloading.
Caution To prevent a connectivity loss, you must reconfigure the uplink switches to correspond with the change made in this procedure.
- Table 3-2 lists supported uplink types and the ports that carry each type of VLAN traffic.
Table 3-2 Uplink Types and VLAN Ports
|
|
|
|
1 |
Ports 1 and 2 (HA) |
Ports 1 and 2 (HA) |
Ports 1 and 2 (HA) |
2 |
Ports 1 and 2 (HA) |
Ports 1 and 2 (HA) |
Ports 3–6 (LACP) |
3 |
Ports 1 and 2 (HA) |
Ports 3-6 (LACP) |
Ports 3–6 (LACP) |
4 |
Ports 1 and 2 (HA) |
Ports 3-4 (HA) |
Ports 5–6 (HA) |
Flexible |
There is no traffic segregation based on traffic class. |
For a description of each uplink, see the Uplinks section.
- When you migrate from a flexible network uplink type to a static network uplink type, the configuration for all the port channels, native VLANs, and port states is lost.
- You must ensure that the uplink connectivity is the same for that static network uplink type.
- If the system is not configured for a required uplink type, shut down the system from ILO after you save the configuration for both active and standby and then reload.
SUMMARY STEPS
1. configure terminal
2. network uplink type number
3. (Optional) show network-uplink type
4. copy running-config startup-config
5. reload
DETAILED STEPS
|
|
|
Step 1 |
configure terminal Example: switch# config terminal switch(config)# |
Enters the global configuration mode. |
Step 2 |
network uplink type number Example: switch(config)# network uplink type 2 switch(config)# |
Changes the uplink type for the Cisco Nexus Cloud Services Platform. The number argument range is from 1 to 4. |
Step 3 |
show network-uplink type Example: switch(config)# show network uplink type
Administrative topology id: 2
Operational topology id: 1
|
Displays the uplink configuration for verification. |
Step 4 |
copy running-config startup-config Example: switch(config)# copy running-config startup-config |
Saves the running configuration persistently through reboots and restarts by copying it to the startup configuration. |
Step 5 |
reload Example: switch(config)# reload This command will reboot the system. (y/n)? [n] y 2009 Oct 30 21:51:34 s1 %$ VDC-1 %$ %PLATFORM-2-PFM_SYSTEM_RESET: Manual system restart from Command Line Interface switch(config)# |
Configuring Port Channels
You can configure the port channels in the Cisco Nexus Cloud Services Platform. You can configure the port channels only in the flexible network uplink type configuration.
BEFORE YOU BEGIN
Before beginning this procedure, you must know or do the following:
- Log in to the CLI in EXEC mode.
- You must group the ports together in either HA or LACP mode.
- You must ensure that the ports are not used by any other port channel or by any VSB interface. To delete a port see Deleting Port Channels.
- An Ethernet interface with conflicting native VLAN cannot be part of an existing port channel.
SUMMARY STEPS
1. configure terminal
2. [no] interface name
3. interface ethernet name
4. channel-group id mode {ha | active}
5. (Optional) show network port-channel summary
DETAILED STEPS
|
|
|
Step 1 |
configure terminal Example: switch# configure terminal switch(config)# |
Enters the global configuration mode. |
Step 2 |
[no] interface name Example: switch(config)# [no] interface PortChannel1 switch(config-if)# |
Places you into configuration mode for the port channel or creates a port channel ID if the port channel does not exist. The port channel range is from 1to 6. |
Step 3 |
Example:
switch(config)# interface GigabitEthernet4
|
Places you into configuration mode for the named Ethernet interface. The interface names can be Gigabit Ethernet only. |
Step 4 |
channel-group id mode {ha | active} Example: switch(config-if)# channel-group 1 mode active switch(config-if)# |
Assigns an ethernet interface to a port channel. The mode can be either HA or Active. |
Step 5 |
show network port-channel summary Example: switch(config)# show network port-channel summary ------------------------------------------------------------------------- Group Port-Channel Adm-State Type Member-Ports ------------------------------------------------------------------------- 1 PortChannel1 up ha Gi1 Gi2 |
Deleting Port Channels
You can delete port channels. You must delete ports from the port channel and then delete the port channel.
BEFORE YOU BEGIN
Before beginning this procedure, you must know or do the following:
- Log in to the CLI in EXEC mode.
- Shut down the VSBs using the port or the port channel that contains the port.
SUMMARY STEPS
1. configure terminal
2. interface ethernet name
3. no channel-group
DETAILED STEPS
|
|
|
Step 1 |
configure terminal Example: switch# configure terminal switch(config)# |
Enters the global configuration mode. |
Step 2 |
Example:
switch(config)# interface GigabitEthernet4
|
Places you into the configuration mode for the named interface. The interface names can be Gigabit Ethernet only. |
Step 3 |
no channel-group Example: switch(config-if)# no channel-group switch(config-if)# |
Deletes the port channel. |
Assigning Uplinks to a VSB Interface
You can assign uplinks to a VSB interface. You can assign uplinks to a VSB interface only in the flexible network uplink type configuration.
BEFORE YOU BEGIN
Before beginning this procedure, you must know or do the following:
- You are logged in to the CLI in EXEC mode.
- By default, uplinks are assigned to the first available free port or port channel.
- If both free ports and port channels are available, free ports get preference over port channels.
- If no free ports are available, you must assign the uplink manually.
- A VSB can have different uplinks for every port or port channel.
SUMMARY STEPS
1. configure terminal
2. virtual-service-blade name
3. [no] interface name uplink name
4. (Optional) show network {[uplink] | summary}
DETAILED STEPS
|
|
|
Step 1 |
configure terminal Example: switch# configure terminal switch(config)# |
Enters the global configuration mode. |
Step 2 |
virtual-service-blade name Example: switch(config)# virtual-service-blade vsm-5 switch(config-vsb-config)# |
Places you into the configuration mode for the named virtual service blade. |
Step 3 |
[no] interface name uplink name Example: switch(config-vsb-config)# interface control uplink PortChannel2 |
Assigns a VSB Ethernet interface to an uplink. |
Step 4 |
show network summary |
(Optional) Displays VSB Ethernet interfaces assigned to an uplink. |
|
- Example: switch(config)#show network summary ----------------------------------------------------------------------------- Port State Uplink-Interface Speed RefCnt MTU Nat-Vlan Oper Admin Oper Admin Oper Admin ------------------------------------------------------------------------------ Gi1 up up 1000 0 9000 Gi2 up up 1000 0 9000 Gi3 up up 1000 3 9000 Gi4 down up 1000 0 9000 Gi5 down up 1000 0 9000 Gi6 down up 1000 0 9000 Po1 up up 1000 13 9000 VsbEth6/1 up up Gi3 Gi3 1000 9000 VsbEth6/2 up up Gi3 Gi3 1000 9000 VsbEth6/3 up up Gi3 Gi3 1000 9000 control0 up up Po1 Po1 1000 9000 mgmt0 up up Po1 Po1 1000 9000 |
Verifying the Uplink Configuration
To verify the uplink configuration, use the following commands:
Example 3-1 Network Uplink Type
switch# show network uplink type
Administrative topology id: 2
Operational topology id: 1
Example 3-2 Network
This example shows how to display information about the network:
GigabitEthernet5 is down (not connected)
Hardware: Ethernet, address: 0010.18a5.c524 (bia 0010.18a5.c524)
MTU 9000 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Auto-Negotiation is turned on
0 multicast frames, 0 compressed
0 input errors, 0 frame, 0 overrun, 0 fifo
0 packets output, 0 bytes
0 underrun, 0 output errors, 0 collisions
GigabitEthernet6 is down (Administratively down)
Hardware: Ethernet, address: 0010.18a5.c526 (bia 0010.18a5.c526)
MTU 9000 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Auto-Negotiation is turned on
0 multicast frames, 0 compressed
0 input errors, 0 frame, 0 overrun, 0 fifo
0 packets output, 0 bytes
0 underrun, 0 output errors, 0 collisions
Example 3-3 Network Cdp Neighbors
switch# show network cdp neighbors
Capability Codes: R - Router, T - Trans-Bridge, B - Source-Route-Bridge
S - Switch, H - Host, I - IGMP, r - Repeater,
V - VoIP-Phone, D - Remotely-Managed-Device,
Device-ID Local Intrfce Hldtme Capability Platform Port ID
sfish-cat3k-K5-stack2 GigabitEthernet1 173 S I cisco WS-C375 GigabitEthernet1/0/45
sfish-cat3k-K5-stack1 GigabitEthernet2 133 S I cisco WS-C375 GigabitEthernet1/0/45
sfish-cat3k-K5-stack2 GigabitEthernet3 173 S I cisco WS-C375 GigabitEthernet1/0/46
sfish-cat3k-K5-stack1 GigabitEthernet4 133 S I cisco WS-C375 GigabitEthernet1/0/46
Example 3-4 Network Counters
switch# show network counters
-----------------------------------------------------------------------
Port InOctets InUcastPkts InMcastPkts
-----------------------------------------------------------------------
GigabitEthernet1 146344975 1163124 105444
GigabitEthernet2 128022491 1110953 280235
GigabitEthernet3 28839731 209796 11722
GigabitEthernet4 107951630 907268 269112
PortChannel1 274367466 2274077 385679
VsbEthernet1/1 17208966 81687 0
VsbEthernet1/2 230213 2011 0
-----------------------------------------------------------------------
Port OutOctets OutUcastPkts OutMcastPkts
-----------------------------------------------------------------------
GigabitEthernet1 73351536 339419 105444
GigabitEthernet2 34200 200 280235
GigabitEthernet3 48242 405 11722
GigabitEthernet4 35492 206 269112
PortChannel1 73385736 339619 385679
VsbEthernet1/1 36137879 158796 0
VsbEthernet1/2 35632175 415746 0
VsbEthernet1/3 41904366 148529 0
Example 3-5 Network Summary
switch# show network summary
-----------------------------------------------------------------------------
Port State Uplink-Interface Speed RefCnt MTU Nat-Vlan
Oper Admin Oper Admin Oper Admin
------------------------------------------------------------------------------
VsbEth6/1 up up Gi3 Gi3 1000 9000
VsbEth6/2 up up Gi3 Gi3 1000 9000
VsbEth6/3 up up Gi3 Gi3 1000 9000
control0 up up Po1 Po1 1000 9000
mgmt0 up up Po1 Po1 1000 9000
Example 3-6 Network Port Channel Summary
switch# show network port-channel summary
-------------------------------------------------------------------------
Group Port-Channel Adm-State Type Member-Ports
-------------------------------------------------------------------------
1 PortChannel1 up ha Gi1 Gi2
Example 3-7 Network Uplinks
GigabitEthernet5 is down (not connected)
Hardware: Ethernet, address: 0010.18a5.c524 (bia 0010.18a5.c524)
MTU 9000 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Auto-Negotiation is turned on
0 multicast frames, 0 compressed
0 input errors, 0 frame, 0 overrun, 0 fifo
0 packets output, 0 bytes
0 underrun, 0 output errors, 0 collisions
GigabitEthernet6 is down (Administratively down)
Hardware: Ethernet, address: 0010.18a5.c526 (bia 0010.18a5.c526)
MTU 9000 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Auto-Negotiation is turned on
0 multicast frames, 0 compressed
0 input errors, 0 frame, 0 overrun, 0 fifo
0 packets output, 0 bytes
0 underrun, 0 output errors, 0 collisions
Recommendations for Configuring Uplinks for Passthrough Interfaces
After you create a Cisco Nexus VXLAN Gateway VSB, you can configure it to function as a passthrough interface. The passthrough feature enables the VSB to assign a virtual interface to a dedicated uplink. This uplink can be a Gigabit Ethernet port on the Cisco Nexus Cloud Services Platform or a port channel. For more information about setting up a passthrough interface, see the Setting up a Passthrough Interface section.
Note The passthrough feature is supported only in a flexible topology. You must migrate from a static to flexible topology before you configure your VXLAN Gateway and the Citrix NetScaler 1000V VSBs in the passthrough mode.
These recommendations enable you to free up ports to configure the VSBs as passthrough interfaces. These recommendations are based on the topology types that you may be migrating from. You can use the common procedures in the following sections to free up the ports:
Topology Type 1
In topology type 1, the management, control, and data traffic share the same uplink. To free up the ports in topology 1, do the following:
1. Migrate to Topology 5 by using the instructions in Migrating from a Static Network Uplink to a Flexible Network Uplink.
2. Configure the HA port channel using the first two physical ports.
3. Configure the existing VSBs to use this high-availability (HA) port channel for the management, control and data traffic.
4. Use the remaining four ports to configure the new VSBs in the passthrough mode. To configure your VSBs in the passthrough mode, see the Setting up a Passthrough Interface section.
Topology Type 2
In topology type 2, the management and control traffic share an uplink and the data link traffic is separated. To free up the ports on this topology, do the following:
1. Migrate to Topology 5 by using the instructions in Migrating from a Static Network Uplink to a Flexible Network Uplink.
2. If you want to separate the traffic on data, and the management and control links on your existing VSBs, do the following:
a. Configure the HA port channels using the first two physical ports.
b. Configure the existing VSBs to use this HA port channel for the management and control and data traffic.
c. If your existing VSBs require interface redundancy on the data traffic, do the following:
– Create a port channel using the physical ports 3 and 4.
– Configure the existing VSBs to use this port channel for the data traffic.
– Use the remaining two physical ports to configure the new VSBs in the passthrough mode.
d. If your existing VSBs do not require interface redundancy on the data traffic:
– Make the third physical port a shared interface for all of the existing VSB’s data traffic.
– Use the remaining three physical ports to configure the new VSBs in the passthrough mode.
3. If you do not want to separate data traffic from the management and control traffic, see Topology Type 1.
Topology Type 3
Topology type 3 uses two uplinks where the control and data traffic share an uplink and the management traffic is separated. To free up the ports on this topology do the following:
1. Migrate to Topology 5 by using the instructions in Migrating from a Static Network Uplink to a Flexible Network Uplink.
2. If you want to separate management traffic from the control and data traffic, do the following:
a. Create a port channel using the physical ports 1 and 2.
b. Configure existing VSBs to use this port channel for their management traffic.
c. Create a port channel using physical ports 3 and 4.
d. Configure existing VSBs to use the port channel for their control and data traffic.
e. Use the remaining two physical ports to configure the new VSBs in passthrough mode.
3. If you do not want to the separate management traffic from the control and data traffic, see Topology Type 1.
Topology Type 4
In this topology, the management, control and data traffic are all on separate uplinks. To free up the ports on this topology, do the following:
1. Migrate to Topology 5 by using the instructions in Migrating from a Static Network Uplink to a Flexible Network Uplink.
2. If you want interface redundancy for the management, control and data traffic and also want to separate the traffic for all the links, do the following:
a. Create a port channel using physical ports 1 and port 2.
b. Configure the existing VSBs to use this port channel for the management traffic.
c. Create a second port channel using physical ports 3 and 4.
d. Configure the existing VSB to use this port channel for the control traffic.
e. Create a third port channel using physical ports 5 and 6.
f. Configure the existing VSBs to use this port for the data traffic. Now, there is no physical port available for passthrough interfaces.
3. If you want to separate the management, control and data traffic but do not need HA, do the following:
a. Configure the existing VSBs to use the first interface for the management traffic.
b. Configure the existing VSBs to use the second interface for the control traffic.
c. Configure the existing VSBs to use the third interface for the data traffic.
d. Use physical ports 3,4, and 5 for configuring VSBs in the passthrough mode.
4. If you want to separate the management, control and data, and HA for only management traffic, do the following:
a. Create a port channel using physical ports 1 and 2.
b. Configure the VSBs to use the port channel for management traffic.
c. Configure the VSBs to use the third port for control traffic.
d. Configure the VSBs to use the fourth port for data traffic.
e. Use the physical ports 5 and 6 for configuring the VSBs in the passthrough mode.
5. If you want to separate data from management and control traffic, see the Topology Type 2.
6. If you want to separate the management traffic from the data and control traffic, see the Topology Type 3.