Wired Intelligent Edge

 Hi All,

I had two 8320 switch configured in a VSX setup.

This was working wel during the pre-staging fase.

After we deployd the two switches at the customer site, the ISL link won't come up anymore and we see the following message on the

Secondary switch on the ISL ports.

Interface 1/1/52 is down
Admin state is up
State information: Disabled by feature <---------- !!!!!

!

The switches are above each other. We connected the two switches with two DAC cables of 40G. The cable works on other ports so is good.

The Keepalive link is working. i can ping the other 8320 via this link.

!

Its a straight forward config so i'm rather confused .

Can someone help me out with this?

Below the VSX configuration.

!!

********* Switch 1: *********

interface lag 128
no shutdown
description ISL-Link
no routing
vlan trunk native 1 tag
vlan trunk allowed all
lacp mode active
!
interface 1/1/52
no shutdown
lag 128
interface 1/1/53
no shutdown
lag 128
!
vsx
inter-switch-link lag 128
role primary
keepalive peer 10.255.255.2 source 10.255.255.1
!
Interface 1/1/52 is down
Admin state is up
State information: Waiting for link <---------- !!!!!
Description:
Hardware: Ethernet, MAC Address: d0:67:26:e2:59:3d
MTU 1500
Type QSFP+DA1
qos trust none
Speed 40000 Mb/s
Auto-Negotiation is on
Input flow-control is off, output flow-control is off
!
Interface 1/1/53 is down
Admin state is up
State information: Waiting for link <---------- !!!!!
Description: ISL-Link
Hardware: Ethernet, MAC Address: d0:67:26:e2:59:35
MTU 1500
Type QSFP+DA1
qos trust none
Speed 40000 Mb/s
Auto-Negotiation is on
Input flow-control is off, output flow-control is off

******** Switch 2: *********
interface lag 128
no shutdown
description ISL-Link
no routing
vlan trunk native 1 tag
vlan trunk allowed all
lacp mode active
exit
!
interface 1/1/52
no shutdown
lag 128
interface 1/1/53
no shutdown
lag 128
!
vsx
inter-switch-link lag 128
role secondary
keepalive peer 10.255.255.1 source 10.255.255.2
!
Interface 1/1/52 is down
Admin state is up
State information: Disabled by feature <---------- !!!!!
Description:
Hardware: Ethernet, MAC Address: d0:67:26:e2:09:53
MTU 1500
Type QSFP+DA1
qos trust none
Speed 0 Mb/s
Auto-Negotiation is off
Input flow-control is off, output flow-control is off
Rx
!
!
Interface 1/1/52 is down
Admin state is up
State information: Disabled by feature <---------- !!!!!
Description:
Hardware: Ethernet, MAC Address: d0:67:26:e2:09:53
MTU 1500
Type QSFP+DA1
qos trust none
Speed 0 Mb/s
Auto-Negotiation is off
Input flow-control is off, output flow-control is off

It seems like SW-2 is your issue. What happens if you do a show interface transciever? Do both switches see the DAC?

What about a show vsx status vsx-peer?

The other thing I would try is to remove the DAC, re insert it, and see if the log messages say anything

Hi

Both switches see the DAC cable.

2018-10-24:08:47:11.426599|pmd|3804|LOG_INFO|AMM|-|Transceiver hot-swap insert for interface 1/1/53
2018-10-24:08:47:11.221049|pmd|3804|LOG_INFO|AMM|-|Transceiver hot-swap insert for interface 1/1/52

The switches dont see each other so the vsx peer is down and isl link is down.

Would you mind sharing complete sanitized (AKA no Serial Numbers) running configurations of both Aruba 8320?

Would be nice to see the outputs of commands:

show interface dom

show interface 1/1/52 dom detail

show interface 1/1/52 transceiver detail

show interface 1/1/53 dom detail

show interface 1/1/53 transceiver detail

show vsx status

show vsx status inter-switch-link

show vsx status keepalive

show vsx configuration inter-switch-link

executed on both Aruba 8320 VSX Primary and Secondary.

Well, both switches saw DACs...the date is 2018-10-24...or your switch time isn't actually updated by means of an NTP?

All,

TAC Support suggested to seroize the switches and start from fresh because of a corrupt DB or Upgrade to TL10.02.0010.

After the upgrade the VSX was working like it should.

Thank you all.

We have this issue in TL10.02.0010 as well.

Please try the following if the port is part of the L2 lag and disabled by feature:

- remove the port from LAG

- shut, put it back to routing

- put it back in the L2 LAG, and no shut

Let me know.

It worked in secondary switch and primary switch port didn't come up, same port will work with different LAG.

strange that it works on secondary and not primary.

you may have to go through support for further analysis.

Are you saying that there is an issue with QSFP+DA1 DAC Cables on ArubaOS-CX 10.02.0010?

Personally all looks good to me: I haven't noticed anything bad about those DAC Cables [*] neither on ArubaOS-CX 10.01 nor on ArubaOS-CX 10.02...and I run VSX too.

[*] HPE X242 40G QSFP+ to QSFP+ 1m DAC Cable (JH234A).

I don't think this is related to DAC. I saw once this issue when changing a lot lag port membership (typically for lab/demo/training).