Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 7 Mar 2004 11:23:47 -0600
From:      David Sze <dsze@alumni.uwaterloo.ca>
To:        Mike Tancsa <mike@sentex.net>
Cc:        freebsd-stable@freebsd.org
Subject:   Re: Odd network issue ... *very* slow scp between two servers
Message-ID:  <20040307172347.GA40032@mail.distrust.net>
In-Reply-To: <6.0.3.0.0.20040306210359.08fc35a8@209.112.4.2>
References:  <20040306130937.N71806@ganymede.hub.org> <6.0.3.0.0.20040306180314.08adede0@209.112.4.2> <20040306210515.M13247@ganymede.hub.org> <20040306211328.H13247@ganymede.hub.org> <6.0.3.0.0.20040306210359.08fc35a8@209.112.4.2>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Mar 06, 2004 at 09:11:09PM -0500, Mike Tancsa wrote:
> 
> I would in the following order
> a) check the settings on the switch. AutoNeg *only* works when both sides 
> are set to auto-neg. If both sides are not auto, make sure both sides are 
> the same for speed and duplex settings.

To add to this point, even with both sides set to AutoNeg, it seems there
can still be problems with em devices.

The company I work for has a customer with IBM xSeries 345 machines, which
have dual em devices onboard.  They had the machines hooked up to a Cisco
2924 switch, both sides were set to AutoNeg.  Both sides also said they
negotiated to 100/Full, but the number of interface errors was through the
roof.

The problems went away when both sides were forced to 100/Full insted of
AutoNeg.  In this case, Marc may be SOL if he must use the unmanaged switch.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040307172347.GA40032>