I have a 2.5 Gb coda volume on a server.
I have a pair of clients, each with a 5Gb cache.
The clients are hoarding the entire volume.

Client A is my notebook computer. I walk into my home, sit down, open it up
and get on the net.

Fifteen minutes later, I cd down into the coda volume on client A and create a
700kb file (by copying one of my students' dissertation to a new file).
Then I say "cfs strong."
Ten minutes later, it's still not visible on client B.
During this ten minutes, venus on client A spins, maxing out my load average.
Simple things like /bin/pwd and cfs cs hang.
Finally, I see a reintegration on client A's codacon, the loadav returns
to normal, and codacon on client B shows the upload; the file is now
visible on client B. The /bin/pwd and cfs cs complete.

Then I try it again, with a new copy, and it works instantly.

I have no idea why it is unusably slow in one case and fast in the other.

Client A is running Fedora Core with the coda-client-6.9.1-1 rpm.
However, it is running an older kernel: 2.6.17.14 with the coda
kernel that came *with* that kernel. (Do I need to be compiling from the latest
coda kernel modules?)

Client B is running Ubuntu feisty with the 6.9 coda debs from 
    deb http://www.coda.cs.cmu.edu/debian unstable/

The server is running Debian with the coda debs from the same place.
    -Olin