On 12/22/2011 10:17 AM, Francois Labreque wrote:
> Le 2011-12-21 18:35, Jim Henderson a écrit :
>> On Wed, 21 Dec 2011 09:15:01 -0500, Tom Austin wrote:
>>
>>> On 12/20/2011 11:14 AM, Jim Henderson wrote:
>>>> On Tue, 20 Dec 2011 07:35:35 -0500, Tom Austin wrote:
>>>>
>>>>> What should I be looking at to help determine where the problem is
>>>>> originating from?
>>>>
>>>> With network problems, I tend to start with a sniffer to see if there's
>>>> something weird going on on the network. If you can determine the
>>>> source of the delays (like a name lookup failure/retry cycle, lots of
>>>> CRC errors or other issues that are causing problems), that can often
>>>> point to a hardware issue or some other cause that you hadn't
>>>> considered.
>>>>
>>>> Jim
>>>
>>>
>>> Wireshark has been suggested - I may look into trying it if I get the
>>> time.
>>
>> That's a good one, and the one I use. I find it's better to do this
>> rather than try to guess - it can save time spent looking at
>> blinkenlights
>> on switches and whatnot, because it tells you exactly what's happening on
>> the wire.
>>
>> Jim
>
> Keep in mind that in a switched environment, the switch will only send
> the packets that your workstation should see (i.e.: packets intended for
> you and broadcasts/multicasts). Without configuring the switch for
> port-replication, you will not be able to see everything that goes on in
> the network.
>
> In other words, take automatic diagnostics of "high level of broadcasts
> from XYZ" with a grain of salt since - by design - it's the only frames
> coming from that station that you will see.
>

Yep, I am with you on this one - we have a switch - and likely in the 
mix I would have forgotten about it's 'benefits'

Post a reply to this message