Re: Keeping my triggers enabled

"Connor McDonald" <connor_mcdonald_at_yahoo.com> wrote: <3A070DCF.5DD4_at_yahoo.com> <3A070DCF.5DD4_at_yahoo.com>

> Why not regularly monitor the ability to issue ALTER TRIGGER instead ?

The procedure is mainly created to prevent myself from doing anything stupid. In need the ability to disable triggers at times however.

> If you really do you want to monitor triggers - you'd be better off
> looking at the source tables (not the view) which would be more
> efficient...
>
> You can sleep DBMS_LOCK.SLEEP or host commands to sleep in SQL PLus, but
> I would still recommend preventing the problem rather than reacting to
> it.

Thanks for that. Seems a bit strange to put this procedure in the DBMS_LOCK package, but there you go. I am now looping in my SQL*Script, saving the server from branching a new process every 4 seconds. Also, querying the base tables is a lot cheaper indeed.

I agree that preventing the problem would be better, but you can _never_ be sure that no trigger is disabled. The output of the triggers is used for some interfaces. If a trigger is disabled, we won't notice (and that is the essential part: we won't notice) until the end of the day. By then, it would be impossible to re-create the output of the triggers. I am sticking to my little guard dog for my own peace of mind. When you give it some thought, I am sure you will agree that shutting down the database is the most sensible option.

-- 
Jan

jantah_at_hot.mail.com
... and you know what to do with hot.mail, right?