I have a problem with Cisco 4006 switch.
When I do 2000-continuous ping to directly connected servers interface (NIC), I notice about 10 or so, packet loss.
I have no clue to find out, whether this problem within the switch IOS “cat4000-is-mz.121-23.E1.bin”
What are the elements could affect the packet loss on the directly connected interface (i.e. NIC on Servers) to the switch ?
thanks in advance.
Can you check if there are errors on the interfaces of the switch or on the nics of the servers.
Also the CPU load of your 4006 could be the cause of the drops, if it is to high.
Do the servers have multiple NIC's in teaming mode , the traffic they send can sometimes cause drops too.
Do you have redundant 4006 switches? if so try to ping from the physical address of the standby device.
There are no errors on the physical interfaces on the switch. Likewise on the NIC of the servers. One point, I connected a laptop and ping it directly and I see packet loss and ping-time average about 7ms for a directly connected NIC.
CPU load is about 9 percent.
NIC maybe teamed up but I am not sure, how that affect packet loss and 7ms average response time for directly connected interface.
Yes, there are two redundant 4006 switches, trunked between them.
Teaming might cause drops because i have seen it in a network.
We had drops in the ping from the servers to the hsrp address and when we turned of the teaming and put it active/standby mode on the servers , the issue was gone, i cannot really explain it , it is because of the type of traffic they are sending when doing active/active teaming.
If you ping from the physical address of the standby switch to the servers and you do not have drops it is probably the teaming.
I am investigating all possibilities. I have raised the questions to system administrators, with regards to teamed up NIC. I will update you, when I do get a reply from them.
Thanks a million for your input.