We have 2 Solaris 7 servers, server X has one Oracle 8.1.6 database
and tuxedo 7.1 environment, server Y has 3 Oracle 8.1.6 databases and
three tuxedo 7.1 environments. We can start and stop tuxedo for all 4
environments manually using tmboot and tmshutdown respectively. We can
also start and stop the single environment on server X as part of a
backup job via cron. The server Y environments receive the following
error when running the same job:
+ tmshutdown -y
tmshutdown: NLS:6: Message not found, catalog CMDTUX_CAT, set 1, num
703
: CMDTUX_CAT:4849: NLS:6: Message not found, catalog CMDTUX_CAT, set
2, num 4849
The script is invoked from root cron which has the entry:
su tuxedo -c "/usr/local/bin/xtuxedoonoff DATABASE STOP"
The server environments are identical as are the scripts that are run.
Has anyone any ideas?
Thanks
[Moderator's note: Have the script print its environment with an env command.
Perhaps one of the environments is messed up.]