Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.misc -> Re: Verify_user function not validating as designed

Re: Verify_user function not validating as designed

From: Jim McMahon <jp_mcmahon_at_hotmail.com>
Date: Mon, 15 Nov 2004 22:48:14 GMT
Message-ID: <41993181.589445556@news.charter.net>


For what it's worth, here's what I got when I ran your function from TOAD using the arguments 'uname', 'abcdefg!1', and 'abcdefg!2'. I had stripped it down a bit, eliminating all the earlier checks for this test.

ORA-20006: Password should differ by at least 3 characters.
ORA-06512: at "MY_SCHEMA.VERIFY_USER", line 52
ORA-06512: at line 12


As you can see, the differ logic seems to work. The only other thing I can think to suggest is that you try a test case where the old/new passwords are *not* enclosed in double quotes on the alter user statement.

Hope this helps.

Jim

wilcoxd.forum_at_gmail.com (Dan) wrote:

>Jim,
>
>Thanks for the feedback.
>
>I should have mentioned that I tested all the other test cases in the
>function and they all errored out correctly. So my user is correctly
>associated with the profile, etc. It is only this one test that seems
>to fail, which is why this is so strange. I figured if someone else
>was using similar logic and/or had tried this very same function they
>could confirm that it fails for them too, etc. and maybe get to the
>bottom of it that way, and then either apply a fix they had already
>engineered to this script or get an alternate method to do this that
>Oracle would not have issue with.
>
>If you have any other suggestions or recommended resources that I
>could follow-up with the assistance would be appreciated.
>
>Thanks
>
>jp_mcmahon_at_hotmail.com (Jim McMahon) wrote in message news:<4192e033.175353495_at_news.charter.net>...
>> I glanced at the code and don't see any obvious problems, but how do
>> you know the function is even being invoked ? You don't show anything
>> that would imply it is, such as associating the user joe with a
>> profile that uses the function.
>>
>> If you haven't already done so, you might consider directly invoking
>> the function, putting in a couple of DBMS_OUPUT.PUTLINE statements, or
>> changing your test to exercise one of the earlier (and simpler)
>> checks. Does it act as expected if you:
>>
>> SQL> alter user joe identified by "a"; -- should raise -20002
>>
>>
>> P.S. You don't need to commit after altering the user.
>>
>> Hope this helps.
>>
>> Jim
>>
>> Being ordinary and nothing special is a full-time job.
>> jp_mcmahon_at_hotmail.com (Jim McMahon in real life)

Being ordinary and nothing special is a full-time job. jp_mcmahon_at_hotmail.com (Jim McMahon in real life) Received on Mon Nov 15 2004 - 16:48:14 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US