topic RE: Tips to find root cause of problem in ExtremeSwitching (EXOS/Switch Engine)

Tips to find root cause of problem

Kar_Del_Moral — Thu, 05 Apr 2018 01:26:00 GMT

latency of some computer in one specific vlan, Not all of the computers are affected, but it is increasing... checking traffic on ports, no messages on Core. If I move the pc to a different vlan the problem is solved for that computer, but still happening on the vlan. Please help

RE: Tips to find root cause of problem

Robert_Cummins — Thu, 05 Apr 2018 02:35:00 GMT

Have you checked for:
- port error counters
- client side error counters
- counters in general

What does a tcpdump from a client show?
How are you measuring the latency and how much of a change are you measuring?

RE: Tips to find root cause of problem

Kar_Del_Moral — Thu, 05 Apr 2018 02:57:00 GMT

Yes... I check.. but nothing.. on the counters all normal, but a did a sh fdb port on all the ports in the affected vlan... and found a switch connected, as soon I disable the port the other pc's normalize... now we need to check if it is the switch or a pc connected to it generating the traffic...

Thanks

RE: Tips to find root cause of problem

Robert_Cummins — Thu, 05 Apr 2018 03:27:00 GMT

 I suspect your network was looped with that other switch and you were encountering a broadcast storm. This would be apparent in a tcpdump from the client; you'll see the same broadcast packet appear in the dump multiple times. Have you enabled erp on the switch? I think it's erp; erp will automatically shutdown ports when it detects a loop.

RE: Tips to find root cause of problem

Dorian_Perry — Thu, 05 Apr 2018 03:27:00 GMT

ELRP - Extreme Loop Resolution Protocol
https://gtacknowledge.extremenetworks.com/articles/Q_A/What-is-ELRP/