Nexus 7K sub-interface is down with reason "other"

[Drew1815]

Hi All,

Have a weird issue where I have created a new additional /identical sub-interface (L3 Vlan) on 4 x Nexus 7Ks in which on one of the N7K, one sub-interface didn't come up as expected: (i.e. the other 3 x N7K sub-interfaces are UP)

SR002# sh int eth1/2.1026 status

--------------------------------------------------------------------------------
Port             Name            Status    Vlan      Duplex  Speed   Type
--------------------------------------------------------------------------------
Eth1/2.1026      SR001 to down      routed    auto    auto    10Gbase-SR
SR002#

SR002# sh int eth1/2.1026 brief

--------------------------------------------------------------------------------
Ethernet      VLAN    Type Mode   Status  Reason                   Speed     Port
Interface                                                                                               Ch #
--------------------------------------------------------------------------------
Eth1/2.1026   1026    eth  routed down    other                      auto(D) --
SR002#

The primary interface (Eth1/2) already has a number of pre-existing sub-interfaces configured - all up and working with no issues.

Here an example of one of the existing sub-interfaces:

SR002# sh int eth1/2.1021 status

--------------------------------------------------------------------------------
Port             Name            Status    Vlan      Duplex  Speed   Type
--------------------------------------------------------------------------------
Eth1/2.1021      SR001 to connected routed    full    a-10G   10Gbase-SR
SR002#

What appears to be different to the existing ones is the Duplex /Speed status being displayed.

On all the exiting ones:

Duplex : full

Speed: a-10G

On the non-working one:

Duplex : auto

Speed:  auto

From memory only the primary /main interface can set the speed /duplex parameters, which are then inherited by the sub-interfaces, so if all the pre-existing sub-interfaces are functional with full /a-10G, why would the newly configured one display "auto/auto"??? Also for 10G SFPs, the speed is fixed anyway.

Here is the standard sub-interface template:

interface Ethernet1/2
  description <XXX>
  mtu 9216
  no shutdown

!

interface Ethernet1/2.1026
  description <XXX>
  mtu 9100
  encapsulation dot1q 1026
  vrf member <XXX>
  ip address 10.100.10.1/30
  no shutdown

!

interface Ethernet1/2.1021
  description <XXX>
  mtu 9100
  encapsulation dot1q 1021
  vrf member <XXX>
  ip address 10.100.20.1/30
  no shutdown

All thoughts are most welcome!

Thanks

Cheers

Drew

[gaurav.jhawar1]

Is there any result or outcome for this connectivity? I too encounter the same issue. 

[Drew1815 · Accepted solution]

The issue was found to be a "typo" in the VRF Context name - it is "obvious now" :-) that the correctly defined vrf name under the sub-interface did not match the global vrf context name.............resulting in the "other" display.

[Abdulila Alhosaine]

Did you make the configuration for  VRF 1026 ?

Please post the configuration for both sides.