Network Oddities -- Failure to Communicate

[Nicholas Edwards]

Let me first say that I have the EXACT setup, two netgear (the real DEC
21410-AE ones), cards and a crossover cable. I had a little problem
getting them to work in Full Duplex 100mbit mode, but got them working
after a while.

When using a crossover cable, it seems that if one machine is off when the
other initializes (loads the driver) its card, that because there is no
link beat, that it just uses whatever default it has. Then when the second
machine gets turned on, it tries to operate at a different mode than the
card in the first. Of course, if one were using a hub or switch,
everything would autonegotiate fine, but with x-over, both machines need
to be on.

The solution seems to be that one should compile the tulip driver as a
module and then do an:

insmod tulip options=14

to force MII 100baseTx Full Duplex

this seems to work real well for my setup, all three lights on the card go
on, and I get more bandwidth than I know what to do with. Of course, I
*really miss* the collision light :)

If this works, let me know.

IF it doesnt, please tell me which lights *are* functioning on your card
when you go through the setup you described in your previous post.

Nicholas

On Sun, 20 Sep 1998, Matt Shirel wrote:

> Hi,
> 
> I'm at a loss in trying to get 2 machines to speak to each other via  a
> crossover cable.  Each machine has a Netgear FA310tx card (with the DEC
> 21140 chip).  Each machine initializes the card correctly (as far as I
> can tell).  And routing is set up accurately.
> 
> When I attempt to communicate with the other card, I get no response. 
> If (for example) I attempt to ping the other machine, ping indicates no
> response.  However, I can watch the lights on the cards flash as if
> something is being sent or recieved.
> 
> I just had something interesting happen (I'm trying hard to brainstorm
> this even as I'm composing this message.)  Everything has been the same
> working in both directions... (ie defining either machine as machine 1
> or machine 1)
> 
> Scenario 1)  Machine 1 running ping and tcpdump.
> 	ping hangs getting no response.
> 	tcpdump shows 	machine 1 > machine 2 echo request
> 
> Scenario 2) Machine 1 running ping, Machine 2 running tcpdump
> 	ping hangs getting no response
> 	tcpdump shows	machine 1 > machine 2 echo request
> 			machine 2 > machine 1 echo reply
> 
> Scenario 3) Machine 1 running ping and tcpdump, Machine 2 running
> tcpdump
> 	ping responds with normal timing loop.
> 
> I'm at a complete loss.  I don't know what the normal network traffic is
> associated with ping and don't have a set up to determine that right
> now...  But this behavior seems very strange to me..
> 
> 
> I'm running 2.0.35 (my own compile from RH sources), tulip.c v 0.89H.
> 
> I'll include the output from tulip-diag at the end of the message.
> 
> If anyone has any suggestions, I'd greatly appreciate hearing them...
> 
> Thanks in advance...
> 
> MATT
> 
> ------------------------
> tulip-diag -a
> 
> Unknown PCI device ID bffffb84.
> tulip-diag.c:v1.05 8/28/98 Donald Becker (becker@cesdis.gsfc.nasa.gov)
> Index #1: Found a Digital DS21140 Tulip adapter at 0xec00.
> Digital DS21140 Tulip Tulip chip registers at 0xec00:
>   ffa04800 ffffffff ffffffff 00fff028 00fff228 fc000102 320e0200
> fffe0000
>   e0000000 ffffcbf8 ffffffff fffe0000 ffffff80 ffffffff 1c09fdc0
> fffffec8
>  The Rx process state is 'Stopped'.
>  The Tx process state is 'Stopped'.
> Transmit stopped, Receive stopped, full-duplex.
>  The transmit threshold is 128.
>  Port selection is MII, full-duplex.
> 
> -------------------------
> tulip-diag -e
> 
> Unknown PCI device ID bffffb84.
> tulip-diag.c:v1.05 8/28/98 Donald Becker (becker@cesdis.gsfc.nasa.gov)
> Index #1: Found a Digital DS21140 Tulip adapter at 0xec00.
> EEPROM transceiver/media description for the Digital DS21140 Tulip chip.
> 
> Leaf node at offset 30, default media type 0800 (Autosense).
>  CSR12 direction setting bits 00.
>  1 transceiver description blocks:
>   Media MII,  block type 1.
>    MII interface PHY 0 (media type 11).
>  MII PHY found at address 0, status 0x782d.
>  MII PHY #0 transceiver registers:
>    1000 782d 7810 0000 01e1 41e1 0001 0000
>    0000 0000 0000 0000 0000 0000 0000 0000
>    0000 0000 4000 0000 38c8 0010 0000 0002
>    0001 0000 0000 0000 0000 0000 0000 0000.
> 
> ----------------------------
> tulip-diag -m 
> 
> Unknown PCI device ID bffffb84.
> tulip-diag.c:v1.05 8/28/98 Donald Becker (becker@cesdis.gsfc.nasa.gov)
> Index #1: Found a Digital DS21140 Tulip adapter at 0xec00.
>  MII PHY found at address 0, status 0x782d.
>  MII PHY #0 transceiver registers:
>    1000 782d 7810 0000 01e1 41e1 0001 0000
>    0000 0000 0000 0000 0000 0000 0000 0000
>    0000 0000 4000 0000 38c8 0010 0000 0002
>    0001 0000 0000 0000 0000 0000 0000 0000.
> ------------
> thanks again...
>