(Illustration by Gaich Muramatsu)
Jan and Rune - Thank you for the prompt replies. You are right, the SrvLog did show that 127.0.0.2 returned as it's IP address. After reconfiguring the server started properly and I was able to create my root volume. Rune - I do agree with your overall assessment of the vice-setup script, however in this case it could easily be resolved by have the script request for an address. Adding something as simple as this to the vice-setup, would probably work: hnip=`hostname -i` echon "Is your host ip address ($hnip) ? (y/n) " read yesno if [ x$yesno = x -o x$yesno = xyes -o x$yesno = xy ] ; then ${exec_prefix}/sbin/codaconfedit server.conf ipaddress $hnip else echon "Enter your host ip address:" read hnip ${exec_prefix}/sbin/codaconfedit server.conf ipaddress $hnip fi Thanks again. >>> Jan Harkes <jaharkes_at_cs.cmu.edu> 10/21/2006 10:57 AM >>> > I just ran through vice-setup and I didn't receive a magic command. > Is the documentation outdated or did I miss something? The > User/System Admin Manual mentions using a createvol_rep command, but I > don't I have all the information needed to complete my installation. ... > Congratulations: your configuration is ready... > > Shall I try to get things started? (y/n) y > - Coda authentication server (auth2 &) > - Coda update server (updatesrv) > - Coda update client (updateclnt -h coda0.omni-ts.com) > Creating /vice/spool > - Coda file server (startserver) > coda0:/ # The basic configuration is ready, however the initial root volume did not get created. So I assume the server startup failed somehow. Check /vice/srv/SrvLog and possibly /vice/srv/SrvErr. My guess is that the server was unable to resolve it's own name to a usable ip-address, i.e. it either couldn't resolve it's own hostname, or got the address of the loopback interface (127.0.0.1). This is a fairly common problem, we used to just continue from here and the volume would seem to get created correctly. However if the server thinks it's own address was 127.0.0.1, the created volume could not be accessed by clients. A client asks for 'volume location information', receives the 127.0.0.1 address and is unable to connect to the volume at that address (since it loops back to the client). If it is the hostname resolution problem the fix is fairly simple. Edit /etc/hosts, remove the servers name from the loopback interface and add a new entry which has the server's real ip-address and hostname. Then restart the server with the 'startserver' command and if it started successfully (File Server started message in /vice/srv/SrvLog), create the root volume by hand with 'createvol_rep / coda0.omni-ts.com'. I think that is really all that is needed. JanReceived on 2006-10-23 12:59:17