3

I'm having a issue with receiving duplicate packets in my network and I'm unsure what's causing it. A simplistic explanation of what's happening is this:

Server1 sends a ping request to Server2

Server2 receives the ping request from Server1, sends ping reply to Server1

Server1 receives two ping replies (DUP)

The network layout is pretty simple - here is a (poorly) drawn diagram:

http://i.imgur.com/jW0LoTU.png

Some things to note that I think is related:

  • The SW1.LAX.AGG switches each have RVI's from different subnets. We have a VLAN called "test" with subnets 10.0.0.0/24, 10.0.1.0/24 and 10.0.2.0/24. The VLAN interface has the first IP of each range assigned. The VLAN "test" has a "l3-interface" pointing to the proper VLAN interface.

  • When you ping an IP from another subnet (say pinging 10.0.1.200 while the primary IP on the machine is 10.0.0.200) it goes over the default gateway of 10.0.0.1 and gets the DUP ping reply. When pinging from a secondary IP on another subnet (10.0.2.200) using the ping -i eth1:0 there is no DUP reply.

  • When receiving the duplicate reply the TTL is not increasing or decreasing.

  • This is not a duplicate IP issue.

Any suggestion on how I can debug this further / is there something silly I'm missing?

Thanks!

2
  • Have a go at recording the ping route; ping Server2 source Server1 count 4 record-route. With any luck, that will expose any anomalies. – Ryan Foley Dec 29 '15 at 23:53
  • Did any answer help you? if so, you should accept the answer so that the question doesn't keep popping up forever, looking for an answer. Alternatively, you could provide and accept your own answer. – Ron Maupin Aug 12 '17 at 22:35
2

Troubleshooting is required. First identify the culprit.

Turn off VRRP (or simply power off one of those routers). Does that fix it? If so look at the VRRP config. Otherwise, power off a VSS switch. Does that fix it? Start by identifying the problem...

Not sure if you are talking FEX's here but ideally those are NOT dual homed (as in your diagram), provided that your servers are. The proper way to do it is shown below.

enter image description here

1

Might be a case of unicast storm flood If pinging from server2 to server1 is fixing the issue, that might be that.

Otherwise, I would check that there is no proxy arp anywhere in your network.

2
  • It's not fixing the issue when pinging - the issue is persistent. It's when pinging out of another virtual interface that has an IP on another subnet is when the issue seizes to exist. There is no proxy arp in the network. I'll see if it is a unicast storm flood and mark this answer if that's what it is. Thanks! – WinkyWolly Nov 25 '15 at 17:25
  • Just a small update - it doesn't appear to be unicast flooding. The traffic seems normal. I've enabled storm control to help verify that nothing is being triggered and it seems to be fine. – WinkyWolly Nov 26 '15 at 20:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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