In answering Will I miss any changes if I replace an oracle trigger while my application is running?, I went looking to see if the trigger was locked by an INSERT statement. It wasn't and I can't find anything on the internet to suggest that a trigger can be locked.
If I run the following in one session:
create table test_trigger (id number);
create table test_trigger_h (id number);
create or replace trigger test_trigger_t
after insert on test_trigger for each row
begin
insert into test_trigger_h (id) values (:new.id);
end;
/
insert into test_trigger
select level
from dual
connect by level <= 1000000;
and then in a second session try to find out what locks are occurring I get the following:
select object_name, object_type
, case l.block
when 0 then 'Not Blocking'
when 1 then 'Blocking'
when 2 then 'Global'
end as status
, case v.locked_mode
when 0 then 'None'
when 1 then 'Null'
when 2 then 'Row-S (SS)'
when 3 then 'Row-X (SX)'
when 4 then 'Share'
when 5 then 'S/Row-X (SSX)'
when 6 then 'Exclusive'
else to_char(lmode)
end as mode_held
from v$locked_object v
join dba_objects d
on v.object_id = d.object_id
join v$lock l
on v.object_id = l.id1
join v$session s
on v.session_id = s.sid
;
OBJECT_NAME OBJECT_TYPE STATUS MODE_HELD
-------------------- -------------------- --------------- ---------------
TEST_TRIGGER TABLE Not Blocking Row-X (SX)
TEST_TRIGGER_H TABLE Not Blocking Row-X (SX)
According to Oracle, the trigger is not being locked.
However, if I try to replace the trigger whilst the INSERT statement is running it will not be replaced until after the statement has completed (not including a commit), which implies that the trigger is locked.
In this situation, is the trigger locked and if so how would one determine that it is?
To determine if a trigger(as well as any other stored procedure) is locked or not, the V$ACCESS dynamic performance view can be queried.
Session #1
insert into test_trigger
select level
from dual
connect by level <= 1000000;
Session #2
SQL> select *
2 from v$access
3 where object = upper('test_trigger_t')
4 ;
Sid Owner Object Type Con_Id
--------------------------------------
441 HR TEST_TRIGGER_T TRIGGER 3
Those kinds of locks are library cache pins(library cache locks are resource(TM
type of lock) locks), needed to ensure that an object is protected from being modified while session is executing it.
--session sid # 441
insert into test_trigger
select level
from dual
connect by level <= 1000000;
-- session sid #24
create or replace trigger test_trigger_t
after insert on test_trigger for each row
begin
insert into test_trigger_h (id) values (:new.id);
end;
-- Session # 3
select vs.sid
, vs.username
, vw.event
from v$session vs
join v$session_wait vw
on (vw.sid = vs.sid)
join v$access va
on (va.owner = vs.username)
where vs.username = 'HR'
Result:
Sid Username Event
--------------------------
24 HR library cache pin
....
441 HR log file switch (checkpoint incomplete)
Here we can see that the session #441 waits for a log file switching and session #24 waits for library cache pin.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With