The ORACLE-BASE Blog

I’ve said it before and I’ll say it again, Oracle products don’t run properly on Tru64 anymore!

The latest issue came from an observation by a developer and involves problems with using NVL inside an EXISTS subquery. The following code is the test case I sent in my TAR to Oracle support.

CREATE TABLE tab1 (
code         VARCHAR2(10),
description  VARCHAR2(50)
);

ALTER TABLE tab1 ADD (
CONSTRAINT tab1_pk PRIMARY KEY (code)
);

INSERT INTO tab1 (code, description) VALUES ('1', 'ONE');
INSERT INTO tab1 (code, description) VALUES ('2', 'TWO');
INSERT INTO tab1 (code, description) VALUES ('3', 'THREE');

CREATE TABLE tab2 (
code         VARCHAR2(10),
tab1_code    VARCHAR2(10),
description  VARCHAR2(50)
);

ALTER TABLE tab2 ADD (
CONSTRAINT tab2_pk PRIMARY KEY (code)
);

ALTER TABLE tab2 ADD (
CONSTRAINT tab2_tab1_fk FOREIGN KEY (tab1_code)
REFERENCES tab1(code)
);

CREATE INDEX tab2_tab1_fk_i ON tab2(tab1_code);

INSERT INTO tab2 (code, tab1_code, description) VALUES ('1', '1', 'ONE');
INSERT INTO tab2 (code, tab1_code, description) VALUES ('2', '1', 'ONE');
INSERT INTO tab2 (code, tab1_code, description) VALUES ('3', '1', 'ONE');
INSERT INTO tab2 (code, tab1_code, description) VALUES ('4', '2', 'TWO');
INSERT INTO tab2 (code, tab1_code, description) VALUES ('5', '2', 'TWO');
INSERT INTO tab2 (code, tab1_code, description) VALUES ('6', '2', 'TWO');
INSERT INTO tab2 (code, tab1_code, description) VALUES ('7', '3', 'THREE');
INSERT INTO tab2 (code, tab1_code, description) VALUES ('8', '3', 'THREE');
INSERT INTO tab2 (code, tab1_code, description) VALUES ('9', '3', 'THREE');

COMMIT;

VARIABLE p1 VARCHAR2(1)
BEGIN
:p1 := '%';
END;
/

SELECT t1.code
FROM   tab1 t1
WHERE  EXISTS(SELECT rowid
FROM   tab2 t2
WHERE  t2.tab1_code = t1.code
AND    t2.code like :p1)
ORDER BY t1.code;

CODE
----------
1
2
3

3 rows selected.

SELECT t1.code
FROM   tab1 t1
WHERE  EXISTS(SELECT rowid
FROM   tab2 t2
WHERE  t2.tab1_code = t1.code
AND    t2.code like NVL(:p1,'%'))
ORDER BY t1.code;

CODE
----------
1
1
1
2
2
2
3
3
3

9 rows selected.

SELECT t1.code
FROM   tab1 t1
WHERE  EXISTS(SELECT rowid
FROM   tab2 t2
WHERE  t2.tab1_code = t1.code
AND    t2.code like COALESCE(:p1,'%'))
ORDER BY t1.code;

CODE
----------
1
2
3

3 rows selected.

From this you can see that then the bind variable is set to ‘%’ the query without the NVL works as expected, but the query using the NVL displays a row in the master table for each row found in the subquery. In comparison, the query using the COALESCE function works fine.

Now I realise that this is not a great query anyway, but you would still expect it to work!

Happy days in Tru64 land again 🙂

Once again I’m drawn to the inevitable conclusion that running Oracle products on anything but a tier-one platform is a bad idea. What’s more, judging by the recent arrival of Oracle 10g Release 2, there is now only a single tier-one platform. I don’t see a version for Solaris, HP-UX or Windows yet.

As for me, I’m still waiting for 10.1.0.4.0 to be released on Tru64. Release 2 is a long way off 🙂

Cheers

Tim…

I was recently asked to review a product called SQLdetective, so here it is.

Q. What’s it for?
A. It’s for developers and DBAs and anyone else who does anything with an Oracle database.

Q. What features does it have?
A. It has everything you’ve come to expect from a general purpose DBA/Developer IDE. I’m not even going to attempt to list them because it would take forever. Suffice to say, if you’ve seen it in another product, it’s probably here.

Q. What’s it look like?
A. You can look at a bunch of screen shots here. Like most current IDEs it’s a bit busy. Not as plain as PL/SQL Developer, but not as busy as TOAD. The icons are “different”, but I’m not sure if that’s a good or bad thing.

Q. What’s it feel like?
A. I seems quite quick and responsive, but it does some odd things, like when you highlight a table in the navigator the default operation is to show you a page of data. I doubt this would be the first thing I would want to see.

Q. What’s it cost?
A. The basic price is US$99, but if you want all the add-on modules it’s going to cost a total of US$198. So the cost depends on what you need. It’s difficult to compare prices because you rarely have to pay list prices and each IDE has different optional modules. It’s more expensive than the free version of TOAD, and now JDeveloper is free you can do all your PL/SQL development and debugging using that if you’ve got the hardware to run it 🙂

Q. Tim, would you buy it?
A. If I was forced to use an IDE I would use TOAD or JDeveloper for free. I have both loaded on my PC and I never use them. If I was forced to buy a license for an IDE I would consider this product. My company recently bought licenses for PL/SQL Developer, but I still use a text editor and SQL*Plus 🙂

Conclusion:

If you really want to pay money for an IDE give this one a try. It’s got a lot of stuff in it. Alternatively, write good intrumented code and save yourself the cash.

I guess I could just copy and paste this review about almost any of the current IDEs, but at least I’ve kept my promise…

Cheers

Tim…

Looks like Firefox 1.0.6 is out, but the update checker doesn’t know it… Again.

Got through a night without a server crash on the production RAC. Feels like the old days again 🙂

Cheers

Tim…

PS. Posted two more chapters to the publisher today. That leaves me with one new one to write and one to tighten up then I’m finished. Must crack on and get it sorted before I have a nervous break down cause by a complete lack of sleep 🙂

It’s been an “interesting” couple of days.

Since we moved to the new configuration we’ve have had some intermittent server crashes, including one last night where both nodes crashed simultaneously. We’ve applied an OS patch that is meant to solve the problem, but we’re still in the “fingers crossed” stage at the moment. It’s meant some busy nights and annoyed users, but we’re moving forward.

You’ve gotta love it…. Not!

Cheers

Tim…

This weekend our production system was switched from a 3-node to a 2-node RAC.

We were originally using a 2-node RAC (2 CPUs per node) and we added a third node because the system was struggling to cope with the workload. The third node helped us out in some ways, but it caused a lot of trouble in others. Ever since it’s inclusion it became impossible to take one node out of the RAC without bringing the lot crashing down, so much for high availability. In addition, a substantial proportion (about 30%) of the wait states on the system were due to inter-node communication. Now I expected with more nodes there would be more inter-node communication, but it seems a bit excessive. Heaven only knows what would happen in a 4-node cluster…

After a lot of banter with Oracle and HP we’ve finally decided to try a 2-node RAC again, but this time with 3 CPUs per node. OK, it’s actually 4 CPUs per node, but one CPU in each node is permanently offlined, so as not to affect our current Oracle licensing.

All the hardware modifications are complete and all tests indicate that the system is up and running normally. Of course the true test will happen tomorrow morning when the users log in and start to break things 🙂

The best news of all is that the move back to a 2-node cluster means that we can once again shut down one node at a time if we need to do maintenence. This is a big plus.

If everything goes quiet over the next few days it means that I’m fire-fighting and the switchover didn’t go well.

I’d be curious to see how many people out there are using RAC on more than 2 nodes. I’ve only done this on Tru64 with 1og Release 1, but I can say without a shadow of a doubt that it doesn’t work properly. I’m curious if this is Tru64 specific problem or if there is a fundamental flaw in RAC for clusters with more than 2 nodes.

Cheers

Tim…