Hello all !

We are running ADSM clients on Solaris servers.

We are starting ADSM automatically in accordance with the ADSM installation
manual using the following entry in /etc/inittab:

adsm::once:/opt/IBMDSMba5/dsmc sched > /dev/null 2>&1 # ADSM scheduler

We have had no problems with this until the other day with a new machine.
When doing 'init 0', ADSM did not terminate, the machine hung and had to be
shut down forcibly, which resulted in corrupt file systems.

When including run levels in the inittab entry:

adsm:23:once:/opt/IBMDSMba5/dsmc sched > /dev/null 2>&1 # ADSM scheduler

'init 0' was successful.

The machine in question is a Sun Ultra: sun4u sparc SUNW,Ultra-Enterprise.

Does anyone know if there is any special reason for not including run levels
in the inittab entry or if it is a simple typo in the installation manual ?

//  Karl-Ake Eriksson
L M Ericsson Data AB