2

This is very odd problem so place stay with me because it's very strange issue.

We have C3850 L3 core switch and we have created VLAN interface on it and we are using it for L3 routing.

I have other switches connected to C3850 core switches, so issue is if i send any traffic to any other server this switch forwarding that traffic to all its interface, just like SPAN port. ( We don't have any mirror session running on this switch)

Does anybody see this kind of issue in network? look like L3 flooding.

Look like this is happening Unicast Flooding: https://en.wikipedia.org/wiki/Unicast_flood

any idea what could be the issue of it?

Update:

After googling i found people saying reduce ARP timeout and increase MAC timeout, (Arp < Mac)

Cisco default Arp timeout is 4 hour, and Mac timeout is 300s (5min).

ARP timeout:

C3850#sh int port-channel 23 | in Timeout
  ARP type: ARPA, ARP Timeout 04:00:00

MAC timeout:

C3850#show mac address-table aging-time
Vlan    Aging Time  Configured Aging Time
----    ----------  ---------------------
Global Vlan Admin Age: 300

Should i change these values? what is the best practice?

Improve this question
3
  • You didn't provide enough information. You should edit your question to include the switch configuration, what you have done to troubleshoot the problem, and any other relevant information, e.g. logs, show commands, etc.
    – Ron Maupin ♦
    Commented Jun 3, 2016 at 0:21
  • What is your spanning tree configuration? Do you have it enabled? What are the attributes of the service? Do you have a layer 2 network loop that might be causing this?
    – Citizen
    Commented Jun 3, 2016 at 8:24
  • We have L3 core switches with HSRP and many L2 switches connected to L3 core switches, we are running rapid+ STP. I think this is not a STP loop issue, it's look like asymmetric path issue I believe, so I just want to know what would be best time out value for arp and Mac
    – Satish
    Commented Jun 3, 2016 at 11:55

1 Answer 1

Reset to default
2

It was HSRP asymmetric routing issue. We has HSRP and i found few VLANs are active on SW1 and few VLANs were active on SW2 and that was creating ARP flooding. as soon as i move all VLANs (HSRP) on single switch flood is gone.

here is the cisco doc: http://www.cisco.com/c/en/us/support/docs/ip/hot-standby-router-protocol-hsrp/10583-62.html#t8

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.