Re: how to monitor the progress of inserts

As Niall demonstrated so beautifully, there are no relational rules when it comes to v$ views (actually the x$ "tables" upon which the views are built).
As to V$SESSION_LONGOPS being inaccurate, cryptic and useless, I can't follow you there. I don't find them any more cryptic than any other table or view and within the confines of being an estimate, I find them remarkably accurate and extremely useful in gaging eta of a process - or detecting escalating deterioration which is a very useful observation in its own right.

Gogala, Mladen wrote:

> To do otherwise would seriously break the relational
> rules. Sadly, V$SESSION_LONGOPS is inaccurate, cryptic and, generally
> speaking, useless.

-- 
Regards

Wolfgang Breitling
Centrex Consulting Corporation
www.centrexcc.com
--
http://www.freelists.org/webpage/oracle-l