bit errors in case of BERT test

m4rtin

I have a Cisco WS-C2960-24TC-L(c2960-lanbase-mz.122-50.SE1.bin) switch, which has a JDSU SmartClass Ethernet tester connected to SFP port Gi0/1 and another JDSU SmartClass Ethernet tester connected to SFP port Gi0/2. Those two ports are in VLAN 10:

Gi0/1 -> tester1 connected 10 full 1000 1000BaseLX SFP
Gi0/2 -> tester2 connected 10 full 1000 1000BaseLX SFP

In addition I have following SPAN configuration:

monitor session 1 source vlan 10
monitor session 1 destination interface Fa0/1

..and Fa0/1 in connected to PC NIC. Other than this, switch has default configuration. In the PC, there is a tcpdump(1) running.

The problem is, that if I make L2 traffic test between the JDSU testers, there are no bit errors with Acterna payload, but if I use BERT payload, there is constantly bit errors increasing. I have tried with different BERT patterns(all zero, all ones, 2^23-1 etc) but nothing helps. On the other hand, with Acterna payload, there are no errors. I know, it's not directly Cisco related question, but maybe somebody here has experience with different Ethernet testers and their payloads? In addition, where should one prefer BERT over Acterna?

kryolla

BERT bit error rate test is usually for testing T1

m4rtin

kryolla wrote: »

BERT bit error rate test is usually for testing T1

However, it's suitable for Ethernet testing as well, isn't it? In addition, I quite don't understand, what is the main difference for example between Acterna payload filled with zeros and BERT payload filler with zeros?

kryolla

BERT is a stress test for T1 testing, not sure what you are trying to accomplish with testing ethernet, we only test when we run have a patch panel ran between rooms and have to test the punch downs and making sure its wired correctly but we never stress tested an ethernet port on a switch.