Re: ORA_ROWSCN and ROWDEPENDENCIES

DA Morgan wrote:
> Yesterday, I can't find it now, someone wrote that ROWDEPENDENCIES
> has a granularity of 3 seconds.
>
> While that may have been true in the past ... it seems it another bit
> of advice that has outlived its usefulness.
>
> In 10gR2 the following produces a unique SCN for each row.
>
> CREATE TABLE t (
> testcol NUMBER(10))
> ROWDEPENDENCIES;
>
> BEGIN
> FOR i IN 1 .. 1000 LOOP
> INSERT INTO t VALUES (i);
> COMMIT;
> user_lock.sleep(0.1);
> END LOOP;
> END;
> /

You seem to be missing out on some pretty basic oracle concepts here.

Do different rows in a table that are inserted/changed and committed on different commits ever share the same SCN? Received on Wed Jan 17 2007 - 15:24:21 CST