zeng yubo ([EMAIL PROTECTED]) reports a bug with a severity of 2
The lower the number the more severe it is.
Short Description
Oncurrent transactions about select for update
Long Description
Two concurrent transactions control by "select for update", they can shared the
data with insert ?
Sample Code
The table Name is num_center .
select count(*) from num_center ;
we support the result execute is 100.
--(1) first , transaction A exceute:
begin work;
select * from num_center where type='lock_row' for update ;
insert into num_center( type , currval ) values ('test1',100);
--(2) then Transaction B exceute:
begin work;
select * from num_center where type='lock_row' for update ;
select count(*) from num_center ;
--(3) Transaction A exceute:
commit;
--(4) Transaction B exceute:
commit;
--the Transaction B Result is 100 .
why not be : 100 + 1 = 101 ?
No file was uploaded with this report
---------------------------(end of broadcast)---------------------------
TIP 6: Have you searched our list archives?
http://archives.postgresql.org
