If you have a database where querying against v$ Performance views is painful, consider gathering Fixed Object Stats.

Fixed objects are the 700+ X$ tables that feed a lot of these system and performance views.

Fixed Object Stats are _not_ gathered by the default stats gathering procedures. They are not gathered at all unless you run the proc to gather them. In this case Oracle relies heavily on Dynamic Sampling… but that isn’t really a substitute for really gathering stats.

Yes, the mention of Cartesian Joins usually makes a DBA's heart skip a beat. The cartesian joins caused by missing join conditions are definitely a pain to deal with - these are the types of joins that can "blow up" temp space and cause all types of alarms to go off.

I didn't find anything in Oracle's official 11g documentation on this particular join method but I did find plenty of articles about issues with it in their support DB. I have chased down a few of these in the past couple of weeks and here is what I found.

I came across this PL/SQL page today which has pointers to a lot of interesting stuff you can do with PL/SQL.

From the page:
"This library is a collection of various utility packages for PL/SQL, as well as links to useful libraries hosted and maintained elsewhere."

http://code.google.com/p/plsql-utils/

It includes a lot of things I never really figured I would need to do in PL/SQL, but it is nice to see someone is thinking of it. If you have any libraries to share, please contact the maintainers of the page.

...well not yet...

An Infoworld article was released today describing an Oracle vulnerability. The SCN space can be exhausted via bugs or attackers exploiting bugs. Metalink doc: 1376995.1 [Bulletin : Information on the System Change Number (SCN) and how it is used in the Oracle Database] covers a technical description for what SCNs are used for.

You can use this query to find out how close you are to the SCN limit.

Here are two runs of the same SQL from different programs. The top is an automated process, the bottom is Toad. (SQL is the same, but SQL_ID is different because of white space issues):

So why is one finished in 7.2 minutes and one still at 4.7 hours and still counting up?

I normally get a lot of mileage out of Application Express interactive reports. Unfortunately there are some restrictions on the SELECT statements that can be used with them. Here is a demonstration of a technique I use when I want to add some interactivity to a plain old Apex report.

I will create a Select List dropdown that submits and refreshes the report data automagically. The value in the select list will be applied as a filter to the report when it refreshes.

1) Create the Select List

• Items -> New Item -> Select List
• Choose "Select List with Submit"
• Next

I just noticed there is a 11gR2 version of the Developer Days VM available now. You can have a running copy of 11gR2,
Oracle TimesTen In-Memory Database, Oracle XML DB, Oracle SQL Developer, Oracle SQL Developer Data Modeler, Oracle Application Express, Oracle JDeveloper, and Hands-On-Labs (accessed via the Toolbar Menu in Firefox) on a copy of Oracle Enterprise Linux 5.

This is great as a sandbox for development or an easy way to stand up a DB for messing around with. The image supplied by Oracle runs inside Oracle VirtualBox VM engine which is also free.

We moved to a new web host on Black Friday. My old host doesn't appreciate repeat business so it was time to make the change and try something new. I just started up with DreamHost and so far I am impressed. Migration of the site went quicker than expected and so far I only made one mistake that caused any issues but I will blame that on waiting for DNS changes to propagate :)

Had an odd one come across my desk today. I have seen Bind variables used in the WHERE clause thousands of times. I have seen Binds in the SELECT lists hundreds of times. I have seen it used in the HAVING clause zero times... until today when it turns out it was causing a performance issue.

With the Bind variable in the HAVING clause we were seeing an execution plan with lots of Full Table Scans and 12+ second runtimes. When editing the query and using a literal in the HAVING clause we saw 100ms or less performance and very efficient Execution Plans.