This document describes the features, caveats, and limitations for Cisco NX-OS Release 7.0(3)I1(3) software for use on the Cisco Nexus 9000 Series switches and the Cisco Nexus 3164Q switch. Use this document in combination with documents listed in Related Documentation.
Table 1 shows the online change history for this document.
Table 1. Online History Change
Date |
Description |
August 19, 2015 |
Created the release notes for Release 7.0(3)I1(3) |
September 2, 2015 |
In the “Supported FEX Modules” section, added Cisco Nexus B22IBM. |
January 11, 2016 |
Added a link to the ALE limitations in the Limitations section. |
March 4, 2016 |
■ Updated Limitations. ■ Updated Supported FEX Modules. |
March 23, 2016 |
Removed the bullets stating that private VLANs support PVLAN across switches: ■ Through a regular trunk port-channel ■ Through a regular vPC-port |
April 8, 2016 |
Added the following statement to Limitations: The N9K-X9408PC-CFP2 line card does not support port-channeling. |
May 25, 2016 |
■ Added to FEX limitations: VTEP connected to FEX host interface ports is not supported. ■ Added to the Supported FEX Modules section: Note: For Cisco Nexus 9500 switches, 4x10G breakout for FEX connectivity is not supported. Native 10G or 40G should be used. |
Guidelines and Limitations for Private VLANs
Obtaining Documentation and Submitting a Service Request
Cisco NX-OS software is a data center-class operating system designed for performance, resiliency, scalability, manageability, and programmability at its foundation. The Cisco NX-OS software provides a robust and comprehensive feature set that meets the requirements of virtualization and automation in mission-critical data center environments. The modular design of the Cisco NX-OS operating system makes zero-impact operations a reality and enables exceptional operational flexibility.
The Cisco Nexus 9000 Series uses an enhanced version of Cisco NX-OS software with a single binary image that supports every switch in the series, which simplifies image management.
This section includes the following sections:
Table 2 lists the Cisco Nexus 9000 Series hardware that Cisco NX-OS Release 7.0(3)I1(3) supports. For additional information about the supported hardware, see the Hardware Installation Guide for your Cisco Nexus 9000 Series device.
Table 2. Cisco Nexus 9000 Series Hardware.
Product ID |
Hardware |
Quantity |
N9K-C9516 |
Cisco Nexus 9516 16-slot chassis |
1 |
N9K-C9516-FM |
Cisco Nexus 9500 Series fabric module |
3-6 depending on the line card |
N9K-C9516-FAN |
Cisco Nexus 9516 fan trays |
3 |
N9K-C9508 |
Cisco Nexus 9508 8-slot chassis |
1 |
N9K-C9508-FM |
Cisco Nexus 9508 Series fabric module |
3-6 depending on the line card |
N9K-C9508-FAN |
Cisco Nexus 9508 fan trays |
3 |
N9K-X9564PX |
Cisco Nexus 9500 Series 48-port, 1-/10-Gbps SFP+ plus 4-port QSFP I/O module |
■ Up to 4 in the Cisco Nexus 9504 ■ Up to 8 in the Cisco Nexus 9508 ■ Up to 16 in the Cisco Nexus 9516 |
N9K-X9564TX |
Cisco Nexus 9500 Series 48-port, 1-/10-Gbps BASE-T plus 4-port QSFP I/O module |
■ Up to 8 in the Cisco Nexus 9508 ■ Up to 4 in the Cisco Nexus 9504 ■ Up to 16 in the Cisco Nexus 9516 |
N9K-X9536PQ |
Cisco Nexus 9500 36-port, 40 Gigabit Ethernet QSFP aggregation module |
■ Up to 4 in the Cisco Nexus 9504 ■ Up to 8 in the Cisco Nexus 9508 ■ Up to 16 in the Cisco Nexus 9516 |
N9K-X9636PQ |
Cisco Nexus 9500 Series 36-port 40-Gigabit QSFP I/O module Note: Not supported on the Cisco Nexus 9516 switch (N9K-C9516). |
■ Up to 4 in the Cisco Nexus 9504 ■ Up to 8 in the Cisco Nexus 9508 |
N9K-X9464PX |
Cisco Nexus 9500 Series 48-port 10-Gigabit SFP+ plus 4-port QSFP I/O module |
■ Up to 4 in the Cisco Nexus 9504 ■ Up to 8 in the Cisco Nexus 9508 ■ Up to 16 in the Cisco Nexus 9516 |
N9K-X9464TX |
Cisco Nexus 9500 Series 48-port 10-GBASE-T plus 4-port QSFP I/O module |
■ Up to 4 in the Cisco Nexus 9504 ■ Up to 8 in the Cisco Nexus 9508 ■ Up to 16 in the Cisco Nexus 9516 |
N9K-X9432PQ |
Cisco Nexus 9500 Series 32-port 40-Gigabit QSFP I/O module Note: The Cisco Nexus X9432PQ I/O module supports static breakout. |
■ Up to 4 in the Cisco Nexus 9504 ■ Up to 8 in the Cisco Nexus 9508 ■ Up to 16 in the Cisco Nexus 9516 |
N9K-X9408PC-CFP2 |
Cisco Nexus 9500 Series 8-port 100-Gigabit CFP2 I/O module for the Cisco Nexus 9504, 9508, and 9516 modular switches |
■ Up to 4 in the Cisco Nexus ■ Up to 8 in the Cisco Nexus 9508 ■ Up to 16 in the Cisco Nexus 9516 |
N9K-SC-A |
Cisco Nexus 9500 Series System Controller Module |
2 |
N9K-SUP-A |
Cisco Nexus 9500 Series supervisor module |
2 |
N9K-SUP-B |
Cisco Nexus 9500 Series supervisor B module |
2 |
N9K-PAC-3000W-B |
Cisco Nexus 9500 Series 3000 W AC power supply |
■ Up to 4 in the Cisco Nexus 9504 ■ Up to 8 in the Cisco Nexus 9508 ■ Up to 10 in the Cisco Nexus 9516 |
N9K-C9504 |
Cisco Nexus 9504 4-slot chassis |
1 |
N9K-C9504-FM |
Cisco Nexus 9504 fabric module |
3 to 6 depending on line card |
N9K-C9504-FAN |
Cisco Nexus 9504 fan trays |
3 |
N9K-C9396PX |
Cisco Nexus 9300 48-port, 1/10-Gigabit Ethernet SFP+ and 12-port, 40-Gigabit Ethernet QSPF switch |
1 |
N9K-C9396TX |
Cisco Nexus 9300 48-port, 1/10-Gigabit Ethernet BASE-T and 12-port, 40-Gigabit Ethernet QSFP switch |
1 |
N9K-C9372PX |
Cisco Nexus 9300 48-port, 1/10-Gigabit Ethernet SFP+ and 6-port, 40-Gigabit Ethernet QSFP switch |
1 |
N9K-C9372TX |
Cisco Nexus 9300 48-port, 1/10-Gigabit Ethernet BASE-T and 6-port, 40-Gigabit Ethernet QSFP switch |
1 |
N9K-C9332PQ |
Cisco Nexus 9300 32-port, 40-Gigabit Ethernet QSFP switch with support for 4x10G breakout mode ■ Ports 1 to 26 (except 13 and 14) support 4x10G breakout mode. ■ Ports 27 to 32 (ALE uplink ports) support using QSA for 10G SFP/SFP+ transceivers in QSFP+ ports |
1 |
N9K-C93128TX |
Cisco Nexus 9300 switch with 96 1-/10-Gigabit BASE-T ports and eight 40-Gigabit Ethernet QSPF ports (The 1-/10-Gigabit BASE-T ports also support a speed of 100 Megabits.) |
1 |
N9K-C93120TX |
Cisco Nexus 93120TX switch with 96 1-/10-Gigabit BASE-T ports and 6 QSFP uplink ports |
|
N9K-PAC-650W |
Cisco Nexus 9300 650 W AC power supply, hot air out (red) Note: For use with the Cisco Nexus 9396 switch (N9K-C9396PX). |
2 or less |
N9K-PAC-650W-B3 |
Cisco Nexus 9300 650 W AC power supply, cold air in (blue) Note: For use with the Cisco Nexus 9396 switch (N9K-C9396PX). |
2 or less |
N9K-PAC-1200W |
Cisco Nexus 9300 1200 W AC power supply, hot air out (red) Note: For use with the Cisco Nexus 93128 switch (N9K-C93128TX). |
2 or less |
N9K-PAC-1200W-B |
Cisco Nexus 9300 1200 W AC power supply, cold air in (blue) Note: For use with the Cisco Nexus 93128 switch (N9K-C93128TX). |
2 or less |
N9K-C9300-FAN1 |
Cisco Nexus 9300 fan 1, hot air out (red) Note: For use with the Cisco Nexus 9396 switch (N9K-C9396PX). |
3 |
N9K-C9300-FAN1-B |
Cisco Nexus 9300 fan 1, cold air in (blue) Note: For use with the Cisco Nexus 9396 switch (N9K-C9396PX). |
3 |
N9K-C9300-FAN2 |
Cisco Nexus 9300 fan 2, hot air out (red) Note: For use with the Cisco Nexus 93128 switch (N9K-C93128TX). |
3 |
N9K-C9300-FAN2-B |
Cisco Nexus 9300 fan 2, cold air in (blue) Note: For use with the Cisco Nexus 93128 switch (N9K-C93128TX). |
3 |
NXA-FAN-30CFM-F |
Cisco Nexus 9300 fan, port-side exhaust Note: For use with the Cisco Nexus 9332PQ, 9372PX, and 9372TX switches (N9K-C9332PQ, N9K-C9372PX, and N9K-9372TX). |
4 |
NXA-FAN-30CFM-B |
Cisco Nexus 9300 fan, port-side intake Note: For use with the Cisco Nexus 9332PQ, 9372PX, and 9372TX switches (N9K-C9332PQ, N9K-C9372PX, and N9K-9372TX). |
4 |
N9K-M12PQ |
Cisco Nexus GEM 9300 uplink module, 12-port, 40-Gigabit Ethernet QSPF Note: The front-panel ports on these GEM modules do not support auto negotiation with copper cables. Manually configure the speed on the peer switch. |
1 (required) |
N9K-M6PQ |
Cisco Nexus GEM 6-port 40-Gigabit Ethernet uplink module for the Cisco Nexus 9396PX, 9396TX, and 93128TX switches Note: The front-panel ports on these GEM modules do not support auto negotiation with copper cables. Manually configure the speed on the peer switch. |
1 |
N9K-M4PC-CFP2 |
Cisco Nexus 9300 uplink module for the 93128TX (2 active ports), 9396PX (4 active ports), and 9396TX (4 active ports) Top-of-rack switches |
1 |
Table 3 lists the Cisco Nexus 3164Q switch hardware that Cisco NX-OS Release 7.0(3)I1(3) supports.
Table 3. Cisco Nexus 3164Q Switch Hardware.
Product ID |
Hardware |
Quantity |
N3K-C3164Q-40GE |
Cisco Nexus 3164Q switch |
1 |
N9K-C9300-FAN3 |
Cisco Nexus 3164Q fan module |
3 |
N9K-PAC-1200W |
Cisco Nexus 3164Q 1200W AC power supply |
2 |
For additional information about the supported hardware, see the Cisco Nexus 3000 Series Hardware Installation Guide.
See the Cisco 10-Gigabit Ethernet Transceiver Modules Compatibility Matrix for a list of supported optical components.
Cisco NX-OS Release 7.0(3)I1(3) supports the following FEXes on Cisco Nexus 9372PX, 9396PX and 9500 Series Switches:
■ Cisco Nexus 2224TP
■ Cisco Nexus 2232PP
■ Cisco Nexus 2232TM and 2232TM-E
■ Cisco Nexus 2248PQ
■ Cisco Nexus 2248TP and 2248TP-E
■ Cisco Nexus B22Dell
■ Cisco Nexus B22HP
■ Cisco Nexus B22IBM
Note:
■ The 9408 line card is not supported with the 2300 FEX.
■ On the Nexus 9500 series, X9464PX and X9564 PX line cards are supported for FEX connectivity using the SPF+ ports.
■ For FEX HIF port channels, Cisco recommends that you enable STP port type edge using the spanning tree port type edge [trunk] command.
■ 2248PQ supports 4xQSFP (16x10GE SPF+) as network interfaces. To connect from 2248PQ to Nexus 9300 or Nexus 9500, use the supported QSPF+ to SFP+ breakout cables.
Note: For Cisco Nexus 9500 switches, 4x10G breakout for FEX connectivity is not supported. Native 10G or 40G should be used.
This section lists the following topics:
■ New Hardware Features in Cisco NX-OS Release 7.0(3)I1(3)
■ New Software Features in Cisco NX-OS Release 7.0(3)I1(3)
Cisco NX-OS Release 7.0(3)I1(3) supports no new hardware features.
Cisco NX-OS Release 7.0(3)I1(3) includes no new software features for the Cisco Nexus 9000 Series switches and the Cisco Nexus 3164Q switch.
Only one software image (called nx-os) is required to load the Cisco NX-OS operating system. This image runs on all Cisco Nexus 9000 Series switches and the Cisco Nexus 3164Q switch. For installation instructions, see the Cisco Nexus 9000 Series NX-OS Software Upgrade and Downgrade Guide.
To perform a software upgrade, follow the installation instructions in the Cisco Nexus 9000 Series NX-OS Software Upgrade and Downgrade Guide.
Disable the Guest Shell if you need to downgrade from Cisco NX-OS Release 7.0(3)I1(3) to an earlier release.
Note:
■ Downgrading with PVLANs configured is only supported with 6.1(2)I3(4x) releases.
■ For a boot-variable change and reload to a 7.0(3)I1(1x) release, the PVLAN process is not brought up, and the PVLAN ports are kept down. For a boot-variable change to the 6.1(2)I3(3) release and earlier, an ASCII replay will be tried, but feature PVLANs and other PVLAN configurations will fail.
For information about software maintenance upgrades, see the “Performing Software Maintenance Upgrades” section in the Cisco Nexus 9000 Series NX-OS System Management Configuration Guide.
Note: If you perform a software maintenance upgrade (SMU) and later upgrade your device to a new Cisco NX-OS software release, the new image will overwrite both the previous Cisco NX-OS release and the SMU package file.
This section lists limitations related to Cisco NX-OS Release 7.0(3)I1(3).
■ Generation 1 100G line cards (N9K-X9408PC-CFP2) and generic expansion modules (N9K-M4PC-CFP2) only support 40G flows.
■ N9K-X9408PC-CFP2 line cards do not support port channeling.
■ Layer 3 routed traffic for missing Layer 2 adjacency information is not flooded back onto VLAN members of ingress units when the source MAC address of routed traffic is a non-VDC (Virtual Device Context) MAC address. This limitation is for hardware flood traffic and can occur when the SVI (Switched Virtual Interface) has a user-configured MAC address.
■ neighbor-down fib-accelerate command is supported in a BGP (Border Gateway Protocol)-only environment.
■ Uplink modules should not be removed from a Cisco Nexus 9300 Series switch that is running Cisco NX-OS Release 7.0(3)I1(3). The ports on uplink modules should be used only for uplinks.
■ PortLoopback and BootupPortLoopback tests are not supported.
■ PFC (Priority Flow Control) and LLFC (Link-Level Flow Control) are supported for all Cisco Nexus 9300 and 9500 Series hardware except for the 100G 9408PC line card and the 100G M4PC generic expansion module (GEM).
■ FEXes configured with 100/full-duplex speed, without explicitly configuring the neighboring device with 100/full-duplex speed, will not pass data packet traffic properly. This occurs with or without the link appearing to be “up.”
¯ no speed–Auto negotiates and advertises all speeds (only full duplex).
¯ speed 100–Does not auto negotiate; pause cannot be advertised. The peer must be set to not auto negotiate (only 100 Mbps full duplex is supported).
¯ speed 1000–Auto negotiates and advertises pause (advertises only for 1000 Mbps full duplex).
■ Eight QoS groups are supported only on modular platforms with the Cisco Nexus 9300 N9K-M4PC-CFP2 uplink module, and the following Cisco Nexus 9500 Series line cards:
¯ N9K-X9636PQ
¯ N9K-X9464PX
¯ N9K-X9464TX
¯ N9K-X9432PQ
Note: In a non-BGP eVPN environment, the SVI for L2/L3 boundary is expected to be on a non-VTEP routing block (a router attached to a VTEP).
■ Cisco NX-OS Release 7.0(3)|1(3) supports flooding for Microsoft Network Load Balancing (NLB) unicast mode on Cisco Nexus 9500 Series switches but not on Cisco Nexus 9300 Series switches. NLB is not supported in max-host system routing mode. NLB multicast mode is not supported on Cisco Nexus 9500 or 9300 Series switches.
Note: To work around the situation of Unicast NLB limitation, we can statically hard code the ARP and MAC address pointing to the correct interface. Please refer to bug ID CSCuq03168 in detail in the “Open Caveats—Cisco NX-OS Release 7.0(3)I1(3)” section.
■ TCAM resources are not shared when:
¯ Routed ACL (Access Control List) is applied to multiple SVIs in the egress direction
¯ Applying VACL (VLAN ACL) to multiple VLANs
■ Cisco Nexus 9000 Series switch hardware does not support range checks (layer 4 operators) in egress TCAM. Because of this, ACL/QoS policies with layer 4 operations-based classification need to be expanded to multiple entries in the egress TCAM. Egress TCAM space planning should take this limitation into account.
■ Applying the same QoS policy and ACL on multiple interfaces requires applying the qos-policy with the no-stats option to share the label.
■ Multiple port VLAN mappings configured on an interface during a rollback operation causes the rollback feature to fail.
■ The following switches support QSFP+ with the QSA (QSFP to SFP/SFP+ Adapter) (40G to 10G QSA):
¯ N9K-C9396PX
¯ N9K-C9396TX
¯ N9K-C93128TX
¯ N9K-C9372PX,
¯ N9K-C9372TX
¯ N9K-C9332PQ
Note: The Nexus 9300 support for the QSFP+ with QSA adapter has the following limitations:
■ Only 10G can be supported using QSA on 40G uplink ports on N9300 switches in NX-OS.
■ 1G with QSA is not supported.
■ For the Cisco Nexus 9332PQ switch, all ports except 13-14 and 27-32 can support breakout
■ All ports in the QSA speed group must operate at the same speed (see the configuration guide)
■ The following switches support the breakout cable (40G ports to 4x10G ports):
¯ N9K-C9332PQ
¯ N9K-X9436PQ
¯ N9K-X9536PQ
■ A DHCP relay to a subnet broadcast address does not work. Workaround: Configure ip directed-broadcast.
■ The show license usage command shows the incorrect license if a valid license is installed over an honor license.
■ An ACL or source VLAN on a SPAN session affects traffic on other sessions.
■ Traffic is not spanned if SPAN ACL and policy-based routing are enabled.
■ When the interface is in the STP block state, and DHCP snoop is configured on the VLANs of this interface, the DHCP packets coming from the VLANs cannot be blocked by STP.
■ When a QoS policy with a policer is applied on a FEX HIF port, the actual policing happens at the switch and not at the FEX. As a result, the internal VNTAG header of 6 bytes is also considered by the policer.
■ Limitations for ALE uplink ports are listed at the following URL:
This section provides guidelines and limitations for configuring private VLANs.
■ Secondary and Primary VLAN Configuration
■ Private VLAN Port Configuration
■ Limitations with Other Features
Private VLANs have the following configuration guidelines and limitations:
■ Private VLANs must be enabled before the device can apply the private VLAN functionality.
■ VLAN interface feature must be enabled before the device can apply this functionality.
■ VLAN network interfaces for all VLANs that you plan to configure as secondary VLANs should be shut down before being configured.
■ When a static MAC is created on a regular VLAN, and then that VLAN is converted to a secondary VLAN, the Cisco NX-OS maintains the MAC that was configured on the secondary VLAN as the static MAC.
■ Private VLANs support port modes as follows:
¯ Promiscuous
¯ Promiscuous trunk
¯ Isolated host
¯ Isolated host trunk
¯ Community host
■ When configuring PVLAN promiscuous or PVLAN isolated trunks, it is recommended to allow non-private VLANs in the list specified by the switchport private-vlan trunk allowed id command.
■ Private VLANs are mapped or associated depending on the PVLAN trunk mode.
■ Private VLANs support the following:
¯ PACLs (Port Access Control Lists)
¯ RACLs (Router Access Control Lists)
¯ Layer 2 forwarding
¯ PVLAN across switches through a regular trunk port
■ Private VLANs support SVIs as follows:
¯ SVI allowed only on primary VLANs
¯ Primary and secondary IP's on the SVI
¯ HSRP on the primary SVI
■ Private VLANs support PVLAN and STP as follows:
¯ RSTPs
¯ MSTs
■ Private VLANs do not provide support for Cisco Fabric Extenders (FEXs).
■ Private VLANs port mode is not supported on the following:
¯ 40G interfaces of the Cisco Nexus C9396PX or Cisco Nexus C93128TX
¯ Cisco Nexus 3164Q
■ Private VLANs do not provide port mode support for the following:
¯ Port channels
¯ vPCs (Virtual Port Channels) interfaces
■ Private VLANs do not provide support on breakout.
■ Private VLANs do not provide support for the following:
¯ IP multicast or IGMP snooping
¯ DHCP (Dynamic Host Channel Protocol) snooping
¯ PVLAN QoS
¯ VACLs
¯ VTP (VLAN Trunk Protocol)
¯ Tunnels
¯ VXLANs
¯ SPAN (Switch Port Analyzer) when the source is a PVLAN VLAN
■ Shared interfaces cannot be configured to be part of a private VLAN. For more details, see the Cisco Nexus 9000 Series NX-OS Interfaces Configuration Guide.
■ Configuring multiple isolated VLAN configurations per PVLAN group is allowed by the Cisco NX-OS CLI. However, such a configuration is not supported. A PVLAN group can have at most one isolated VLAN.
Follow these guidelines when configuring secondary or primary VLANs in private VLANs:
■ Default VLANs (VLAN1), or any of the internally allocated VLANs, cannot be configured as primary or secondary VLANs.
■ VLAN configuration (config-vlan) mode must be used to configure private VLANs.
■ Primary VLANs can have multiple isolated and community VLANs associated with it. An isolated or community VLAN can be associated with only one primary VLAN.
■ Private VLANs provide host isolation at Layer 2. However, hosts can communicate with each other at Layer 3.
■ PVLAN groups can have one isolated VLAN at most. Multiple isolated VLAN configurations per primary VLAN configurations are not supported.
■ When a secondary VLAN is associated with the primary VLAN, the STP parameters of the primary VLAN, such as bridge priorities, are propagated to the secondary VLAN. However, STP parameters do not necessarily propagate to other devices. You should manually check the STP configuration to ensure that the spanning tree topologies for the primary, isolated, and community VLANs match exactly so that the VLANs can properly share the same forwarding database.
■ For normal trunk ports, note the following:
¯ There is a separate instance of STP for each VLAN in the private VLAN.
¯ STP parameters for the primary and all secondary VLANs must match.
¯ Primary and all associated secondary VLANs should be in the same MST instance.
■ For non-trunking ports, STP is aware only of the primary VLAN for any private VLAN host port; STP runs only on the primary VLAN for all private VLAN ports.
Note: Cisco recommends that you enable BPDU Guard on all ports that you configure as a host port and not enable this feature on promiscuous ports.
■ Private VLAN promiscuous trunk ports allow you to configure a maximum of 16 private VLAN primary and secondary VLAN pairs on each promiscuous trunk port.
■ For private VLAN isolated trunk ports, note the following:
¯ You can configure a maximum of 16 private VLAN primary and secondary VLAN pairs on each isolated trunk port.
¯ The native VLAN must be either a normal VLAN or a private VLAN secondary VLAN. You cannot configure a private VLAN primary port as the native VLAN for a private VLAN isolated trunk port.
■ Downgrading a system that has private VLAN ports configured requires unconfiguring the ports.
■ Before configuring a VLAN as a secondary VLAN, you must shut down the VLAN network interface for the secondary VLAN.
■ The DHCP relay does not work if the line card only has pvlan-secondary access-hosts.
Follow these guidelines when configuring private VLAN ports:
■ Use only the private VLAN configuration commands to assign ports to primary, isolated, or community VLANs.
■ The Layer 2 access ports that are assigned to the VLANs that you configure as primary, isolated, or community VLANs are inactive while the VLAN is part of the private VLAN configuration. Layer 2 trunk interfaces, which may carry private VLANs, are active and remain part of the STP database.
■ If you delete a VLAN used in the private VLAN configuration, the private VLAN ports (promiscuous ports or host ports, not trunk ports) that are associated with the VLAN become inactive.
Consider these configuration limitations with other features when configuring private VLANs:
Note: In some cases, the configuration is accepted with no error messages, but the commands have no effect.
■ Ensure consistent PVLAN type, states and configuration across vPC peers. There is currently no PVLAN consistency check for vPC. Inconsistent PVLAN configs across vPV peers may end up in incorrect forwarding and impacts.
■ Private VLAN ports can be configured as SPAN source ports.
■ Private VLAN host or promiscuous ports cannot be SPAN destination ports.
■ Destination SPAN ports cannot be isolated ports. However, a source SPAN port can be an isolated port.
■ After configuring the association between the primary and secondary VLANs:
¯ Dynamic MAC addresses that learned the secondary VLANs are aged out.
¯ Static MAC addresses for the secondary VLANs cannot be created.
■ After you configure the association between the primary and secondary VLANs, if you delete the association, all static MAC addresses that were created on the primary VLANs remain on the primary VLAN only.
■ In private VLANs, STP controls only the primary VLAN.
Note: See the Cisco Nexus 9000 Series NX-OS Security Configuration Guide for information on configuring static MAC addresses.
This section lists features that are not supported in the current release.
■ VXLAN
■ DHCP
■ FEX
This section lists VXLAN features that are not supported.
■ TX SPAN (Switched Port Analyzer) for VXLAN traffic is not supported for the access-to-network direction.
■ QoS classification is not supported for VXLAN traffic in the network-to-access direction.
■ QoS buffer-boost is not applicable for VXLAN traffic.
■ ACL and QoS for VXLAN traffic in the network-to-access direction is not supported.
■ Native VLANs for VXLAN are not supported. All traffic on VXLAN Layer 2 trunks needs to be tagged.
■ Consistency checkers are not supported for VXLAN tables.
■ BGP eVPN neighbors are not supported over VPC interfaces.
■ VXLAN routing and VXLAN Bud Nodes features on the 3164Q platform are not supported.
■ DHCP snooping and DAI features are not supported on VXLAN VLANs.
■ IGMP snooping is not supported on VXLAN VLANs.
■ Static MAC pointing to remote VTEP is not supported with BGP EVPN.
■ Rollback is not supported on VXLAN VLANs that are configured with the Port VLAN mapping feature.
■ FEX host interface ports are not supported for VLANs extended with VXLAN.
The following ACL related features are not supported:
■ Ingress RACL that is applied on an uplink Layer 3 interface that matches on the inner or outer payload in the network-to-access direction (decapsulated path)
■ Egress RACL that is applied on an uplink Layer 3 interface that matches on the inner or outer payload in the access-to-network direction (encapsulated path)
■ Egress VACL for decapsulated VXLAN traffic
We recommend that you use a PACL or VACL on the access side to filter out traffic entering the overlay network.
DHCP subnet broadcast is not supported.
■ FEX is supported only on the Cisco Nexus 9372PX and 9396PX and 9500 switches. It is not supported on the other Cisco Nexus 9300 Series.
■ FEX vPC is not supported between Nexus9372PX/9396PX (TOR) and 9500 Switches (EOR) as the parent switches.
■ ASCII replay with FEX needs be done twice for HIF configurations to be applied. The second time should be done after the FEXs have come up.
■ IPSG is not supported on FEX ports.
The following lists other features not supported in the current release:
■ Due to a Poodle vulnerability, SSLv3 is no longer supported.
■ The Cisco Nexus 9300 Series switches do not support the 64-bit ALPM routing mode.
■ IPSG is not supported on the following:
¯ The last 6 40G physical ports on the 9372PX, 9372TX, and 9332PQ switches
¯ All 40G physical ports on the 9396PX, 9396TX, and 93128TX switches
This section includes the following topics:
■ Open Caveats—Cisco NX-OS Release 7.0(3)I1(3)
■ Resolved Caveats—Cisco NX-OS Release 7.0(3)I1(3)
Table 4 lists the open caveats in Cisco NX-OS Release 7.0(3)I1(3). Click the bug ID to access the Bug Search tool and see additional information about the bug.
Table 4 Open Caveats in Cisco NX-OS Release 7.0(3)|1(3)
Bug ID |
Description |
HSRP packet decoding fails with an assertion error. |
|
All VLANs are suspended if one has a QoS policy, but the TCAM is not configured. |
|
Spurious error messages appear when an peer-link port-channel member is shut. |
|
The Cisco Nexus 9000 Series switch accepts untagged packets with a dot1Q tag configuration. If the other end is not configured with a dot1Q configuration, and is sending untagged packets, the untagged packets will be accepted on the Cisco Nexus 9000 Series switch. |
|
The Ethanalyzer does not see packets that egress on a virtual interface on an ALE-asic port. |
|
No drop counters are displayed when a FEX HIF is congested. |
|
Microsoft NLB traffic being routed into the destination VLAN is experiencing packet loss. |
|
The show queuing interface command returns empty output when executed for FEX HIF interfaces. |
|
The show policy-map type queuing command does not show statistics for FEX HIF interfaces. |
|
When QoS Lite TCAM is configured, policer violated statistics shown as part of the show policy-map interface command are reported as 0 instead of NA (Not-Applicable). |
|
When copying the tunnel configuration file to running, the tunnel may flap before stabilizing. |
|
Policer action is not supported when a QoS policy of type qos is applied with the no-stats keyword. |
|
Even though there are no QoS classification policies currently active on any of the FEX HIF interfaces, the show incompatibility command still reports FEX QoS incompatibility during downgrade from 3.2 to earlier versions of software. |
|
Traffic cannot be routed using policy-based routing if the next-hop reachability is across the vPC peer link and the local vPC leg is down. |
|
ERPSAN sessions with a destination on the port-channel sub-interface is not supported. |
|
When a remote end of a vPC port channel member is shut down, the local end takes ~10 seconds to shut down. This only occurs when the port channel is 'active' (i.e., has LACP enabled). |
|
The default interface does not remove all the switchport vlan mapping commands. |
|
Vntag-mgr times out after changing VLANS for a range of 20 vPC port-channels. |
|
Some receivers see duplicate traffic when the peer_link_exclude_flag is set to FALSE for some groups (for some VLANS) that have a source on a VPC VLAN. |
|
When a user reloads the active supervisor, the standby supervisor also reloads. During the reload process, the Service Policy Manager (SPM) cannot send data to the standby supervisor. A syslog is observed, notifying the active supervisor that the SPM has not successfully updated its data base to the standby supervisor. The active supervisor reloads the standby supervisor again, and the standby supervisor eventually reaches a good standby state. |
|
VXLAN:BGP-EVPN sessions fail to come up after removing and re-adding an “nv overlay evpn” configuration in scale scenarios. |
|
A rollback operation may fail after removing the port VLAN mapping and restoring to a previous checkpoint that has the port VLAN mapping. |
|
The source VLAN configuration in a monitor session does not error out when the SPAN TCAM is not carved. SPAN on a source VLAN will not work unless the SPAN TCAM is configured. |
|
The wrong PV mapping configuration will not be detected across vPC links. |
|
With 2 hosts that have the same MAC and different IP addresses, a mac move that will happen 5 times. After that one of the hosts will be rendered "frozen" or unable to communicate. This will require a manual deletion of the other MAC and ARP to enable this host to communicate again. |
|
ERSPAN packets are dropped on the intermediate switches if more than one ERSPAN session resolves over 40 Gig uplinks on a ToR. |
|
An ITD policy is shown in no shut state. However, no policy is actually applied to the ingress policy if an invalid ACL is used for "exclude." |
|
Some show commands are having very slow reaction times and appear to hang with certain ITD configuration scenarios. |
|
When access-list is configured for ITD service, this error is received: "ACL cannot apply when more than one node is active.” |
Table 5 lists the Resolved Caveats in Cisco NX-OS Release 7.0(3)I1(3). Click the bug ID to access the Bug Search tool and see additional information about the bug.
Table 5 Resolved Caveats in Cisco NX-OS Release 7.0(3)|1(3)
Bug ID |
Description |
ACLQOS crashes when configuring the switch PO number the same as the old FEX PO. |
|
To support static breakout configuration, sub-interface range is restricted to a number from 1-511. This enhancement request is to allow users to configure any sub-interface number from 1-4094, but internally re-map the sub-interface to one of the values from 1-511, so that it makes migration of configs from other platforms easier, and let users keep their dot1q VLAN and sub-interface number the same for config readability. |
|
MTS build-up for SAP 199 may lead to IPFIB crash of multiple line cards. |
|
In Cisco Nexus 9300 Series switches, when running show ntp trusted-keys or show ntp authentication-keys, the switch incorrectly says "no trusted keys or no authentication keys" configured. |
|
Tshark crashes and produces a crash core when running the following:: ethanalyzer local interface inband display-filter "arp.hw.type == 6" limit-captured-frames 0 |
|
The ip http source-interface command is shown twice in both the running and startup config. |
|
With hierarchical l3 64b-alpm mode enabled, carving ipv6 lpm-entries to '3840' causes partial hardware programming (default hierarchical lookup entry) in the line card Network Forwarding Engines (NFE/Trident). |
|
Nexus 9000 switches can experience slowness and potentially hold memory in MTS buffers when being polled by an SNMP Interface MIB. Namely, IfTable Counters. |
|
On the Nexus 9000, as-path prepend does not work when remove-private-as is configured |
|
A link does not come up with Port 14 or Port 16 on a 9464PX card with GLC-T Transceivers. |
|
The value returned by cefcFRUPowerOperStatus cannot show the power supply fan status. |
|
The HSRP VIP v6 link-local address of the SVI is missing in the output of show ipv6 interface vlan x. As a result, v6 hosts will not learn the RA messages from the router. |
|
The ARP reply is not forwarded on FEX host interfaces. |
|
In a spine/leaf VXLAN/EVPN setup: when a dual home (Non-vPC) host is reloaded 4-5 times, it becomes unreachable for 25 minutes from the network. The issue clears on its own after 25 minutes, which is the default arp timer. |
|
The OSPFv2/v3 summary address is not advertised after reloading. |
|
A vulnerability in the Message of the Day (MOTD) or banner functionality of the NX-OS operating system could allow an unauthenticated, remote attacker to cause the login process to reset. The vulnerability is due to the MOTD display handling when a certain type of terminal session is requested through telnet. An attacker could exploit this vulnerability by repeatedly issuing a telnet session to the NX-OS device and cause the login process to reset. An exploit could allow the attacker to cause the login process of the NX-OS device to reset repeatedly. |
|
A memory leak can be seen in the Nexus 9000 switch while running OSPF. This issue is related to the presence of the table-map in the configuration. The leak in the OSPF process occurs in librpm_dll.so libraries. |
|
The interfaces on the Nexus 9372TX switches running 6.1(2)I3(4a) intermittently fail with the interfaces sometimes remaining down and not recovering. A reload may or may not recover it. |
|
Some of the control packets are not getting classified in their respective class. This results in control packet drops due to default copp policy. |
|
The Cisco Nexus 9300 and 9500 Series includes a version of OpenSSL that is affected by the vulnerability identified by the Common Vulnerability and Exposures (CVE) IDs: CVE-2015-4000, CVE-2015-1788, CVE-2015-1789, CVE-2015-1790, CVE-2015-1792, CVE-2015-1791, CVE-2014-8176 |
|
ACL stats under show system internal access-list input entries detail is not shown for CoPP ACLs. |
|
With ARP suppression enabled, TCAM is not carved for ARP. When a client boots up and sends a GARP packet, we see this packet looped between the vPC VTEP member switches over the MCT/peer-link. |
|
Cisco Nexus 9000 Series switches in VTP transparent mode loop VTP advertisements within a vPC domain, which causes high CPU utilization. The VTP packet storm can hit adjacent devices causing instability downstream or upstream. |
|
%IPFIB-SLOT1-4-UFIB_ROUTE_CREATE: Unicast route create failed for VRF message will be in syslog |
|
A vulnerability in the Command Line Interface (CLI) parser of the Cisco Nexus Operating System (NX-OS) devices could allow an authenticated local attacker to perform a privilege escalation at the CLI. The vulnerability is due to improper input validation of special characters within filenames. An attacker could exploit this vulnerability by authenticating at the local shell and writing a file to disk with certain special characters. The attacker could then use that file with other CLI commands to obtain a shell prompt at the current privilege level. An exploit could allow the attacker to read/write files and perform other privileged commands. |
|
ifTable and ifXTable counter objects toggle between 0 and the actual value between successive pollings. |
|
All DVIFs hang when: ■ There are more than 2 VLANS in the system ■ There is at least one trunk HIF port with sw trunk allowed vlan XXX-YYY ■ Running no switchport trunk allowed vlan |
|
Depending on the topology, when a module is removed, traffic is disrupted for 5 seconds or STP reconvergence can be triggered. |
|
Nexus 9300 switches may not learn MAC addresses on FEX HIF ports. |
|
The N9K acts as a DHCP relay agent with a link in the Spanning-tree Alternate Blocking State. This causes: ■ High CPU usage in dhcp_snoop and pktmgr ■ Excessive drops in the DHCP CoPP class ■ Control plane instability |
|
If a VRF has 32 chars in its name, it remains in the running config after being deleted. |
The entire Cisco Nexus 9000 Series NX-OS documentation set is available at the following URL:
The Cisco Nexus 3164Q Switch - Read Me First is available at the following URL:
To provide technical feedback on this document, or to report an error or omission, please send your comments to nexus9k-docfeedback@cisco.com. We appreciate your feedback.
For information on obtaining documentation and gathering additional information, see the monthly What’s New in Cisco Product Documentation, which also lists all new and revised Cisco technical documentation, at:
http://www.cisco.com/en/US/docs/general/whatsnew/whatsnew.html
Open a service request online at:
https://tools.cisco.com/ServiceRequestTool/create/launch.do
Subscribe to the What’s New in Cisco Product Documentation as a Really Simple Syndication (RSS) feed and set content to be delivered directly to your desktop using a reader application. The RSS feeds are a free service and Cisco currently supports RSS version 2.0.
This product includes cryptographic software written by Eric Young (eay@cryptsoft.com). This product includes software developed by the OpenSSL Project for use in the OpenSSL Toolkit. (http://www.openssl.org/). This product includes software written by Tim Hudson (tjh@cryptsoft.com).
Cisco and the Cisco logo are trademarks or registered trademarks of Cisco and/or its affiliates in the U.S. and other countries. To view a list of Cisco trademarks, go to this URL: www.cisco.com/go/trademarks. Third-party trademarks mentioned are the property of their respective owners. The use of the word partner does not imply a partnership relationship between Cisco and any other company. (1110R)
Cisco Nexus 9000 Series NX-OS Release Notes, Release 7.0(3)I1(3)
© 2013 - 2015 Cisco Systems, Inc. All rights reserved.