Re: Strange NFS performance

[Richard Nyberg <rnyberg@xxxxxxxx>]

At Tue, 20 Apr 2004 13:36:44 -0400,
Dave Cuthbert wrote:
> 
> Have you been able to rule out hardware problems?
> 
> I'm thinking specifically of the cabling -- I've encountered far too 
> many questionable CAT-5 cables which have either been improperly 
> constructed (e.g., 3&4/5&6 paired instead of 3&6/4&5) or had one too 
> many chairs roll over them.  Far less commonly, our switches in the 
> office also have a tendency to lose a port or forget their duplex 
> settings at times.
> 
Hello Dave.

I'm too little of a network person to really be sure that there isn't
any hw problems. However, when I run fbsd 5.2 instead of dragonfly on
Comp3 the network performance is really good in both directions, even
in both directions simultaneously. So I think there is some problem in
dragonfly.

Trying to send as much data as possible in both directions using netcat
and tcp with dragonfly on Comp3 shows performance like this
[bsd//rnyberg]$ netstat 1
            input        (Total)           output
   packets  errs      bytes    packets  errs      bytes colls
        18     0      17116         21     0      17770     0

When running fbsd 5.2 on Comp3 the values are >12000000 instead of >17000
in both input and output columns.

If it was a hw problem, shouldn't that affect fbsd as well?

Anyway, since these are extreme cases, it is not a huge problem for me.
I was just curious about the strange performance.

Regards,
	-Richard