Skip navigation.


Impact of US Daylight saving changes on Oracle

Well, just for a briefing, since 1966, most of the United States has observed Daylight Saving Time from at 2:00 a.m. on the first Sunday of April to 2:00 a.m. on the last Sunday of October. But in 2007, most of the U.S. will begin Daylight Saving Time at 2:00 a.m. on the second Sunday in March and revert to standard time on the first Sunday in November.

So, Oracle has released patches to adapt these Daylight saving time changes. The databases that are using the following will be impacted…

1. Databases using TIMESTAMP WITH TIME ZONE and TIMESTAMP WITH LOCAL TIME ZONE data types and TZ_OFFSET function as they take their time zone information from Oracle's time zone files.

Oracle 10gR2 RAC on Windows Server x64 and Comparison with RHEL

Anu Chawla's picture

Recently Performance Tuning compared Oracle 10gR2 RAC on Windows Server 2003 x64 vs. RHEL. You can download the paper from

The following behavior was observed during testing of the Oracle RAC databases on Red
Hat Enterprise Linux x86_64 and Microsoft Windows Server 2003 Enter:
Oracle RAC Stress Test
o Transactions Per Minute were roughly equivalent for 2 - 150 user sessions.
o Transactions Per Minute were up to 16% higher for MS Windows for 150
- 250 users.
o CPU usage was above 90% for all of these tests for both Linux and MS
Windows Server.
o The response times for the “New Company Registration” test component

Yet another alert log puzzle

A handy alert log is invaluable for troubleshooting database problems. A RAC database has multiple alert logs.
I prefer to monitor them through a single table.

"A master failing to make an entry in the vessel's official
logbook as required by this part is liable to the Government for a
civil penalty of $200.

United States Code. Title 46 - Shipping. Subtitle II -Vessels and seamen, Part G - Merchant seamen protection and relief. Chapter 1113 -Official logbooks.

How meticulously do you keep your book log, Captain?
Luckily for us, our databases are as much ships as they are first mates. They are intelligent enough to keep their own records. And those logs are as important as vessel logs of the past, because the information carried by an early 21st century database could be easily worth more than 1450 tons of tea carried by Cutty Sark in 1870 en route from Shanghai to London.

As Oracle does all the mundane work, our role becomes more creative - to inspect and properly use the gathered information.

Experimenting with the continous mining

cbruhn2's picture

Today I have been experimenting with the new feature of continous mining with logmnr.

My setup

  • Oracle Database 10g Enterprise Edition Release - Production With the Partitioning, OLAP and Data Mining options
  • ALTER DATABASE ADD SUPPLEMENTAL LOG DATA or we don't see the funny things in the redologfiles
I now starts the logminer session in one window with :
dbms_Logmnr.Start_Logmnr(StartTime => SYSDATE - 1 / 24, Options => dbms_Logmnr.dict_From_OnLine_Catalog + dbms_Logmnr.Continuous_Mine);

FROM v$Logmnr_Logs;

Strange behavior from resetlogs.

cbruhn2's picture

During a test of some backup setup with rman I ran into problem running the restore / recover command.

Selects, Cursors, RefCursors, Cursor Expressions, some simple examples of rowset manipulation in SQL and PL/SQL

Kevin Meade's picture

So a buddy of mine, Ray, asked me for some examples of how data could be returned from a PL/SQL procedure. After a short discussion and some fiddling in SQL*Plus, we produced a neat document with some easy examples of what he could do. I still don't know how he is hooking this up to his coding tools, but it is a good demonstration of alternatives for manipulating sets of rows so I figured I'd post it. Hope someone besides Ray finds it useful. Ray's original question was "Kev, is there a way to return data from PL/SQL code so I can use it like a set of rows?". If you are an advanced PL/SQL developer you probabely already know this stuff but then again, it might be worth a five minute look for you.

Oracle Menu - a very useful shell script

Hi there,

I have never joined an Oracle forum before so hello.

I have just spent the last 9 years of my life working for Oracle Consulting UK. I am now contracting in the Cayman Islands.

I spent a lot of time writing a shell script and wondered if you'd like to see it. it really can save you a lot of time and is a nice cockpit to navigate around in.

works great for no database, core, or 11i (eBus)

tell me what you think:

it only runs on unix (ksh) or bash with a very simple documented mod. I even coded it around pdksh's bugs (rightmost pipe not in current process).

Securing Your Database: A simple Brute Force Blocker

Natalka Roshak's picture

If you've had to change a Unix password, you know that most systems won't let you use just anything as a password - your password must be a minimum length, and must not be too easy to guess or brute-force. Oracle doesn't come with this capability installed out of the box, but Oracle 8 and above let DBAs define a custom function to complexity-check user passwords. Oracle provides a default password verification function to do some basic checking, although DBAs need to know about it and install it manually.

Easy XML - Let the Database do the Work

Kevin Meade's picture

I don’t want to learn XML. I don’t want to learn XPATH. I don’t want to learn XQUERY. I don’t want to learn XSLT. I don’t want to learn XSQL. I don’t want to learn XSU. XDB has some way cool stuff. But to-date, I have been pretty successful in not learning things I don’t want to learn. So you might think that when the time came for my databases to start sharing data via a XML transport, I would be in trouble. Fortunately, God gave me at least a few smarts when he loaded me up with laziness. In this article we will discuss one way to move XML formatted data in and out of Oracle using what we already know: object views, instead of triggers, collections, and PL/SQL packages.

What a strange way to write NULL

gojko's picture

A few days ago, while hunting for a bug in PL/SQL code, I stumbled upon the strangest way to write NULL. If the e-mail address parameter was empty, the genius who wrote this PL/SQL procedure set it to , then compared it with ten lines below, in order to log a problem. I really don't know what is it about NULL that scares people so much, but over the years I got used to occasional -1 and 0, or even 'EMPTY'. However, this is the first time I ran across Donald.