Oracle sql question

Shouldn't be the join itself.

Can you post the SQL?

The following example works for me:

drop table tab1 purge;
create table tab1(col1 number);

drop table tab2 purge;
create table tab2(col1 number);

insert into tab1 values(1);
insert into tab1 values(1);
insert into tab1 values(2);

insert into tab2 values(1);
insert into tab2 values(2);
commit;

select t1.col1, t2.col1 from tab1 t1 join tab2 t2 on t1.col1=t2.col1 group by t1.col1, t2.col1 having count(t1.col1) > 1;

Select allOpen in new window

We really need to see your basic query. It could be that the use of having count(column) might be the problem. Typically, you would have something along the lines of -
select a.id, a.colx, b.coly, count(*)  ==> instead count of a certain column
from a inner join b on a.id = b.id
group by a.id, a.colx, b.col
having count(*) > 1

I should've been more precise, I'm not using join per se, but rather selecting from two tables and in the where clause setting  table1.column1 = table2.column1 kind of approach, does that approach factor into the discussion?

That is still a join.  It is the older syntax.

Updating my query above, still works for me.

select t1.col1, t2.col1 from tab1 t1, tab2 t2 where t1.col1=t2.col1 group by t1.col1, t2.col1 having count(t1.col1) > 1;

Select allOpen in new window

Maybe there's something I'm missing;

select 
dv.donor_id, 
dv.visit_datetime,
dv.unit_number,
dn.fullname,
dn.address_line1||' '||dn.address_line2||' , '||dn.city||' '||dn.state_cd||' '||dn.zip_cd5 Address, 
'(' ||dn.evening_phone_area ||')' ||' '|| dn.evening_phone_prefix ||' '||dn.evening_phone_number EvePhone,
dn.date_of_birth,
Substr (Months_Between (sysdate, dn.date_of_birth)/12,0,2) Age,
Lower (dn.email_address) Email,
dv.donation_type_cd,
dv.historical_blood_type
from
rptdba.donor_visit dv,
rptdba.donor dn
where 
dn.donor_id = dv.donor_id and
dv.donation_type_cd Like 'T1' and
dv.historical_blood_type like 'O-' and
dv.batch_date >= to_char(to_date('20120101','YYYYMMDD')) and dv.batch_date <= Sysdate and
dv.visit_status_cd Like 'DN' 
Group By
dv.donor_id,
dv.visit_datetime,
dv.unit_number,
dn.fullname,
dn.address_line1,
dn.address_line2,
dn.city,
dn.state_cd,
dn.zip_cd5, 
dn.evening_phone_area,
dn.evening_phone_prefix,
dn.evening_phone_number,
dn.date_of_birth,
dn.date_of_birth,
dn.email_address,
dv.donation_type_cd,
dv.historical_blood_type
Having count(dv.donor_id) > 1

Select allOpen in new window

slightvw, right on pro, many thanks, I'll adjust my code accordingly and go from there.

Note that my count over just counts the total number of visits for a single donor id.  that likely isn't really what you are after but without sample data and expected results, it is the best I can guess from the SQL you posted.

You would want to partition by all the columns used for the 'duplicates'.

It will also return all the rows.  If you only want the true duplicate rows, you likely want ROW_NUMBER instead of COUNT.

Sorry, I didn't realize that this question was still open. many thanks slightwv.