[c-nsp] autonegotiation fail

[Vincent De Keyzer]

Welcome to the club! I guess you also have noticed the CRC errors on the
interface counters.

I would say that forcing to 100/FD is usually a good idea on Cisco trunks
(switch/swith but sometimes also switch/some routers, like the 7206 VXR).

On access ports, you'd better leave it to auto-negotiation. Workstations
usually negotiate better with a Cisco than another Cisco (conflicting
interests I assume :-).

Vincent


> -----Original Message-----
> From: cisco-nsp-bounces at puck.nether.net [mailto:cisco-nsp-
> bounces at puck.nether.net] On Behalf Of Horváth Szabolcs
> Sent: jeudi 22 septembre 2005 17:23
> To: cisco-nsp at puck.nether.net
> Cc: Timár Csaba
> Subject: [c-nsp] autonegotiation fail
> 
> Hello!
> 
> We have many switches in our network: 2912MfXL, 2924CXLv, 2924MXL,
> 2924XLv,
> 295024, 295024C, 3548XL, wsc1900LiteFx. The software versions are very
> heterogenous.
> 
> We observed that in some cases the autonegotiation between Cisco Catalyst
> switches failed. For example:
> 
> We did the measures with iperf (dast.nlanr.net/Projects/Iperf/).
> The results:
> 	individual test: ~85 Mbit/s up, ~85 Mbit/s down
> 	bidirectional test: ~80 Mbit/s up, ~7 Mbit/s down(!)
> 
> So we set the link to 100 Mbit/full duplex by manually.
> The results:
> 	bidirectional test: ~85 Mbit/s up, ~85 Mbit/s down(!)
> 
> We want to force every switch-switch connection to 100 Mbit/full duplex.
> 
> What do you suggest?
> What are the common pitfalls in this situation?
> 
> Thanks for your reply,
> Szabolcs Horvath
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/