Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

Our Dell Powerconnect 62xx switches have a very high packetloss on the management interface. I presume this is caused by a new appliance which uses multicast for communication but I am not sure.

Our network setup is following:

servers a -> Dell PC6248 |
servers b -> Dell PC6248 |-> juniper core router
servers c -> Dell PC6248 |

What we see is that the multicast traffic arrives at all servers (but only the servers b use the multicast) and I fear that this multicast traffic floods the switch management interface.

The switches' management interfaces are reachable via vlan101, all other traffic is sent over other vlans. When I tcpdump on one of the 2 servers with a vlan 101 ip address, I only get a few arp requests but almost nothing. When I try to ping between these 2 servers, it works like a charm.

I would like to know what a good way is to troubleshoot this problem and maybe help me understand what is going wrong on that subnet.

share|improve this question

1 Answer 1

up vote 2 down vote accepted

Our Dell Powerconnect 62xx switches have a very high packetloss on the management interface. I presume this is caused by a new appliance which uses multicast for communication but I am not sure.

This sounds like an excellent theory. I would recommend you test it by trying to collect a Wireshark capture either from your workstation connected to the management interface (good) or from your wokrstation's port using a traffic mirroring function (better). You should confirm that you are getting multicast traffic on those interfaces some how before proceeding.

The switches' management interfaces are reachable via vlan101, all other traffic is sent over other vlans.

This is what seems confusing to me. If your application is not sending out multicast traffic on your management VLAN (101) then why would you think its affecting your access to the management interface?

share|improve this answer
    
It seems the problem was indeed the multicast traffic. The switches processed the multicast traffic in software instead of in hardware. which caused the switch lack processing power to process the management traffic. –  Hannes May 8 '13 at 15:50
    
@Hannes - If this answer was helpful you should mark it as accepted. –  kce May 8 '13 at 20:14

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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