i have coda client and server working nicely on the same machine. but
when i try to setup venus on a remote client to access this new server,
it dies. 

the same remote venus client machine can be pointed at the CMU
testserver with much better results. this is with reboot, remove/install
of venus RPM, and re-run of the appropriate venus-setup command between
attempts.

only fishy thing i can find is the response to the cfs checkservers
command, when run on the remote client box it gives:

  # cfs checkservers
  Contacting servers .....
    VIOCCKSERV: No such file or directory

is there something i have to do on the remote client side to account for
the ``two secret tokens of _exactly_ 8 characters'' that i used when
installing the server?

anyone seen this before and have a clue for me?

TIA
--sean 
                          sean dreilinger, mlis
                          mailto:sean_at_durak.org
                          http://durak.org/sean

# venus

Date: Thu 02/11/99

13:38:34 /usr/coda/LOG initialized at size 0x86f9c
13:38:34 /usr/coda/DATA initialized at size 0x21be70
13:38:34 brain-wiping recoverable store
13:38:35 loading recoverable store
13:38:35 starting VSGDB scan
13:38:35        0 vsg entries in table
13:38:35        0 vsg entries on free-list
13:38:35 starting VDB scan
13:38:35        1 vol entries in table (0 MLEs)
13:38:35        0 vol entries on free-list (0 MLEs)
13:38:35 starting FSDB scan (833, 20000) (25, 75, 4)
13:38:35        0 cache files in table (0 blocks)
13:38:35        833 cache files on free-list
13:38:36 starting HDB scan
13:38:36        0 hdb entries in table
13:38:36        0 hdb entries on free-list
13:38:36 Initial LRDB allocation
13:38:37 Venus starting...
Killed