All,
I was fiddling with a test script (trying to figure out a quirk in exception handling after cancellation of a procedure) when I got another access violation, reading from address -1 (FF … FF)
Please find the errorlog below. It may be incorrect: it has a timestamp dating it to 2015-03-31 10:01, just like the prior issue (I copied that one earlier today)
I checked for other install directories, but the one I found is even older. It looks like Toad either failed to write the error log, or did so without updating the timestamp…
For the sake of completeness, I’ll paste my test script below.
I do not recall the exact sequence of events, but basically, this is what I did:
- ran the procedure one statement at a time (not the trigger definition), using F9
- made some changes, added the trigger, and restarted by executing the truncate table using F9.
create table test_exceptions(txt varchar2(256 char) );
truncate table test_exceptions;
create trigger test_exceptions_rbi
after insert on test_exceptions for each row
begin
:new.txt := :new.txt || ‘modified by trigger’;
end;
begin
for i in 1 … 9
loop
sys.dbms_lock.sleep(seconds => 3);
end loop;
sys.dbms_output.put_line(‘Test completed normally’);
exception
when others
then
sys.dbms_output.put_line('Test completed with error ’ || sqlerrm);
insert into test_exceptions
values (‘Error occurred’);
commit;
end;
select *
from test_exceptions;
Kind regards,
Abe Kornelis
[View:~/cfs-file.ashx/__key/communityserver-discussions-components-files/86/Toad_5F00_76_5F00_access_5F00_violation_5F00_2.txt:550:0]