Hello,

Now that seems to work. I think that the previous error was because I had an error when creating
data partition (in a file) of 130M. I run now with a 90M data partition and it's ok.

I have 6 anothers questions:

1) Sometime, when I try: kclog manu, I get:
	<<Local login only, could not contact venus>>
	Why ? (Venus is runnin on the client and the scm).


2) If I do a "cfs mkmount /coda/pc/shiva pc.shiva" on the pc.shiva server, on a client I can't see
anything in /coda (with ls -l) but I can open a file in /coda/pc/shiva/. Is it normal ?


3) How kclog works ?
	When I run kclog manu (with or without a kinit manu before), kclog ask me for the password and it
doens't seems to contact kerberos host (kauth2 is launched, no auth2).


4) Files on server are located in /vicepa, right ? But where are stored the names,... of the files ?
	Is it possible to use coda to see an entire machine (i.e. see the / of the server) as I can do with
NFS.
	Related question: is it possible to replace /vicepa by an existing directory ?
		i.e.: clients can see /home/user1 files under /coda/user1 for exemple.


5) Can we really use Code (v 4.6.5) with coda-fs-module-k2.0.34_c4.6.0 module in a production
environnement ?
	i.e. Is it (very) stable or is it know to crash and lose datas ?


6) What are the differences between coda-fs for kernel 2.0.x and kernel 2.1.x ?


Thank you very much for your help. Coda seems to be a very good solution for sharing files.

Manuel


--
____________________________________________________________________
Manuel GUESDON  -  SOFTWARE BUILDERS        <mguesdon_at_sbuilders.com>
http://www.sbuilders.com                        PGP Key Id: 12C3E391
PGP Signed/Encrypted mails prefered