Parsing giving different SQL ID from actually executing SQL?

Question

I think this is worth a thread on its own.
Env: 11.2.0.3 on HP-UX Itanium
Scenario:
1. parse a sql statement using the following
  c := dbms_sql.open_cursor;
  dbms_sql.parse(c, v_sql, dbms_sql.NATIVE);
  dbms_sql.close_cursor(c);2. Row appears in v$SQL with expected SQL_ID and plan_hash_value of 0.
3. In Toad, actually execute the <sql statement>
3. Second row appears in V$SQL with different SQL_ID and plan_hash_value non 0.
What's going on? Why do we get two different SQL_IDs for the same sql text executed by the same user?
Edited by: Dropbear67 on Feb 13, 2013 6:23 PM

Answer 1

>
I think SB is right, it's in the translation.
As for why it's important.. Well I am being given SQL_ID's by the dev team and being asked to go create and fix baseline plans for them. I can get the SQL from V$SQL from the SQL_ID provided, but when I go to run the query myself, I end up with a different SQL ID (and different PLAN_HASH_VALUE because I'm running it in a different optimizer env, but thats irrelevant for here). This causes more work for me to then go search the v$SQL views to try and find the SQL I just ran.
Why do I need to find my new SQL_ID? Because I then use it to lock in the baseline plan which I just created by running the query. I need to know it's SQL_ID and the PLAN_HASH_VALUE.
One would have thought you could just run
select * from table (dbms_xplan.display_cursor()); but that doesn't seem to always play ball - why? maybe a matter for another thread one day
>
Well now, after a day and a half, that we know what it is you are really trying to do you might want to look through this 4 part article: SQL Plan Management - Creating SQL Plan baselines.
It is written by Maria Colgan, a developer on Oracle's Optimizer development team. - can't get a better source of info than that.
Sounds like you already have significant experience with baselines so you may already be familiar with most of this. But it will certainlly be useful to others.
https://blogs.oracle.com/optimizer/entry/sql_plan_management_part_1_of_4_creating_sql_plan_baselines
>
This first post in our series, describes the concepts of SQL Plan Management and how to create SQL plan baselines. The second part will describe how and when these SQL plan baselines are used. The third part will discuss evolution, the process of adding new and improved plans to SQL plan baselines. Finally, the fourth part will describe user interfaces and interactions with other Oracle objects (like stored outlines).
You can create a SQL plan baseline in several ways: using a SQL Tuning Set (STS); from the cursor cache; exporting from one database and importing into another; and automatically for every statement. Let's look at each in turn. The examples in this blog entry use the Oracle Database Sample Schemas so you can try them yourself
>
And Tanel Poder's article 'SQL_ID is just a fancy representation of hash value might be of interest.
http://blog.tanelpoder.com/2009/02/22/sql_id-is-just-a-fancy-representation-of-hash-value/