i remember of the need to set some environment variables when running tsm
for example, my cron daily backup script looks like :

#!/bin/bash
export DSM_DIR=/opt/tivoli/tsm/client/ba/bin
export DSM_CONFIG=${DSM_DIR}/dsm.opt
cd ${DSM_DIR}
${DSM_DIR}/dsmc incr ...

it's possible that an activation at the /etc/inittab level will have a very
reduced environment, thus causing the problem ; and more, i'd say it's not
the standard way to start a service. so i'd urge you to write a start/stop
script that will reside in /etc/rc.d/init.d, that will be called by the S##
or K## links, and that will set the appropriate environment.

then when it works, post it ;-)



A 16:27 08/11/01 -0500, vous avez écrit :
>We recently installed 4.2.1.0 client software on a Linux system with a
>2.4.10 kernal. I have no trouble starting the client scheduler process
>with the 'at now' command, but the client scheduler process is
>consistently crashing with a segment fault when it is started from
>inittab during system initialization. We find messages like:
>
>11/08/01   05:04:42 main thread, fatal error, signal 11
>
>in /opt/tivoli/tsm/client/ba/bin/dsmerror.log. The location of the log
>file is odd, since dsm.sys specifies that dsmerror.log should be written
>in /var/adm/adsm. When the client scheduler process is started using the
>'at now' command it does in fact write its logs in that directory. A
>'kill -l' command on the Linux system lists signal 11 as SIGSEGV.
>
>
                        - * - * - * - * - * - * -
Si nous avons chacun un objet et que nous les echangeons, 
   nous avons encore chacun un objet.
Si nous avons chacun une idee et que nous les echangeons,
   nous avons alors chacun deux idees.

Thierry ITTY
eMail : Thierry.Itty AT Besancon DOT org               FRANCE