I know someone has discussed running a faircom server (c-tree) on netware, I am trying to run it on a linux cluster. The service will start and stop fine when run manually, but will never finish loading when run from the cluster.

Here's the scoop
init script named /etc/init.d/SMserver8 starts and stops the Faircom database server.
I can run "/etc/init.d/SMserver8 start" and everythings starts up fine from the prompt
I can run the cluster load script manually /etc/opt/novell/ncs/SMserver8.load and everything loads fine
I start the resource from the cluster "cluster online SMserver8" and the resource stays in the loading stage for 2 minutes (the length of the timeout on the resource) then goes comotose.
I can verify that during that 2 minute window the faircom service is running and I can connect to it with a client, but cluster services must never get the message that it finished loading because it eventually gives up waiting and unloads the resource.

When loading manuall I notice that after faircom loads it doesn't return a prompt until you press enter, that is why I have a sleep 15 and a couple of echo commands in the script. Like I said, this script runs fine when I execute it manually, and the echo's return me to a prompt.


.. /opt/novell/ncs/lib/ncsfuncs

exit_on_error mount -t ext3 /dev/evms/SMserver8 /SMserver8
# add the ip addres
exit_on_error add_secondary_ipaddress

# start the faircom server
/etc/init.d/SMserver8 start &
sleep 15


exit 0

Anybody have any ideas. I have tried loading this dozens of ways and haven't gotten anywhere with it. I would appreciate any suggestions.