Re: Data Dictionary: relative_fno question

"Howard J. Rogers" wrote:
>
> What this meant, of course, was the file number was no longer absolute
> ("When I tell you file 6, there can be no argument about which file I'm
> talking about) but relative ("When I tell you file 6, I also have to tell
> you which object (and hence tablespace) we're on about, because there could
> be another file 6 lurking somewhere else.
>
> Hence the use of the term 'relative file number' (though you can still use
> the absolute file number up until the point where you add the 1023rd
> datafile to your database and recycle the file numbers).

Hi Howard-

You've confused me a bit here - when you say "file number," are you using that figuratively or are you talking about the file# column of v$datafile (or equivalently the file_id column of dba_data_files)? v$datafile.file# (or dba_data_files.file_id) is most definitely absolute and unique - a primary key if you will. v$datafile.rfile# (or dba_data_files.relative_fno) is the column that can contain duplicates. Seems like you were saying it's the other way around? Or maybe I just misunderstood.

Just so you know I'm not crazy:

Connected to:
Oracle9i Enterprise Edition Release 9.2.0.1.0 - 64bit Production With the Partitioning, OLAP and Oracle Data Mining options JServer Release 9.2.0.1.0 - Production

SQL> select count(1)
  2 from v$datafile;

  COUNT(1)

      1284

SQL> select count(distinct file#)
  2 from v$datafile;

COUNT(DISTINCTFILE#)

                1284

SQL> select count(distinct rfile#)
  2 from v$datafile;

COUNT(DISTINCTRFILE#)

                  956

Regards,
Sean Received on Wed Jan 15 2003 - 19:34:06 CST