[PLUG] Network issues

[Denis Heidtmann]

On Thu, Jan 21, 2010 at 6:33 PM, Mike Connors <mconnors1 at gmail.com> wrote:
> Denis Heidtmann wrote:
>> Don't jump to conclusions based upon a novice like me reading a
>> script!  There could be stuff in there that I missed.
>>
>> It sounds like the consensus here is that to duplicate what happens to
>> the network during boot I would need to [remove the driver (rmmod),
>> restart the network( ifup?) , ping ], repeat 'till bored.  Is this
>> correct?
>>
>> -Denis
> Yes, that seems the correct test to run.
>
> I'm still trying to understand what causes the "1000baseT/Full" to show
> up or not show up for "Advertised link modes"/ 10baseT/Half 10baseT/Full.
>
> My understanding is that during auto-neg each side advertises it's
> highest capability and the 2 link partners negotiate down to a common
> speed/duplex capability.
>
> Also, auto-neg on the link is exchanged when:
> 1. Link is initially connected
> 2. Device at either end of the link is powered up
> 3. Device is reset or initialized
> 4. Renegotiation request is made
>
> Which means the testing you've has forced auto-negotiation to reoccur.
> Which leads me to believe that it's not strictly a auto-neg problem.
>
> I think we really need to understand the missing Gigabit link mode
> advertisement. Why would the NIC not advertise it sometimes...?

I am glad you have focused on this.  Since it is the only thing I have
found which is consistently different between failed and working
modes, it deserves some scrutiny.  I think it is interesting that the
unused capability (1000baseT) is what is missing when in the failed
mode.  Does it mean that the negotiation failed at that point?  Is
there any way to snoop on the negotiation process?  I have wireshark,
but am not familiar with running it.

Thanks for you interest.

-Denis