Links of the Week – #3

Once again, here’s what I enjoyed reading this week. Not all of it is database related:

Jeremy Schneider diagnosed busy waits in a RAC environment. This article and its follow up are probably the most educational articles I’ve read this week.

On oracle-l mailing list, Tanel Poder explains how to set the window title for SQL Plus client running in windows.

Don Seiler had lots of fun with CBO, and lived to tell the tale.

Jeff Hunter was very unhappy with Oracle support. Getting Oracle to backport fixes is indeed difficult.

Dominic Delmolino introduced me to V$OSSTAT and explained how he finds out about new features.

Marc Andersen found a good quote in a book about the panic of 1907.

Gretchen Rubin wants everyone to start a happiness project. Maybe I’ll give her ideas a try.

It was fashion week in NY this week, get updated on what to wear this season. I can’t wait to see how many readers will click on this link.

Enjoy the weekend.


Links of the week – #2

Here’s what I enjoyed reading this week. Warning – not everything here is related to databases.

Kevin Closson wrote an article that really left me with my jaw dropped. He wrote about good IO throughput on NFS systems. The article is excellent, as usual, but the part that really knocked me over was the fact that he wrote his own dd, because he was not satisfied with the throughput he measured with oracle’s dd. Isn’t this guy unbelievable?
That’s why I was happy to read that he sometimes botches RAC installations too.

Tanel Poder wrote about running remote scripts from sqlplus through http or ftp. Definitely something that can be helpful in production, especially someone else’s production.

H.Tonguç Yılmaz wrote best practices for upgrades. He has very good advice there, so it is definitely recommended reading.

On oracle-l list, Anurag Verma asked a good question about block recovery and got some good answers.

My favorite SQL Server blogger wrote a good parable about data types. I’m still troubled by the fact that he ran into so many DBAs that misuse data types so badly.

Paul Graham wrote more advice for startups, and Marc Andersen has a tip too. The Corporate Cynic describes one of the most annoying things you can hear from a manager or a co-worker.

In BMJ, a medical ethics journal, I’ve read an article about Hyperactivity in children: the Gillberg affair. The article describes an academic disagreement between Swedish psychiatric researchers that quickly escalated into law suits. If you think that some Oracle consultants are fighting dirty, you should read what those Swedish scientists do.

I think I’ve got addicted to writing blog summaries, and I have to admit that its more enjoyable writing them when I don’t have to look for Postgres articles in the name of database objectivity.

Botched installations and funny new features
Tips for upgrading and NFS measures
Blocks that recover and startups that jinx
These are a few of my favorite links


Log Buffer #59: a Carnival of the Vanities for DBAs

As I’ve hinted at the beginning of the week, this week I’m writing up edition #59 of Log Buffer. Dave Edwards of Pythian Group contacted me couple of weeks ago and offered this opportunity. I hesitated quite a bit, this is a big responsibility, and I was not sure my rather new blog is ready for such exposure. I’m glad I agreed. Dave Edwards sent me a bunch of helpful links to other interesting blogs, so I got to discover many valuable blogs I’ve managed to miss so far, and to read about databases that I usually don’t see much. Despite the fact that the links Dave sent included many databases besides Oracle, this Log Buffer will probably be a bit biased toward the worlds most popular DB.

Oracle 11g is still the hottest topic in Oracle blogsphere. Tanel Poder started an 11g internals series with an article about automatic memory management, getting into the implementation in Linux. Kevin Closson continued looking into this topic with a typically well researched article.

Blogging About Oracle also has a series about Oracle 11g, written by different authors. I particularly liked Part 7 – Function Result Cache, written by Roel. He gives a good example that demonstrates the value of this feature.
Virag Sharma, in Database, RAC, Dataguard and Apps writes about SQL Performance Analyzer, one of the hottest features in 11g, and Oracle Brains explain how to tune using 11g’s invisible indexes.

If you are still not tired of 11g, Eddie Awad posted 40+ links to 11g articles and blog posts. This should keep everyone busy during the weekend.

Oracle Open World is three month away, but Dan Norris and Shay Shmeltzer are already getting ready for the event. I also need to start looking for interesting sessions. there are always so many to choose from.

And in other news: Tanel Poder published a seriously cool script for tracking session level performance stats that work even when you are not allowed to do anything on the database.
Alex Niujten, on the highly useful Amis Technology blog explained how to create table as select when you have nested tables.
In Eye on Oracle, Mike Brunelli is collecting information about the quality of Oracle support. I hope his project will generate many responses and maybe it will even cause Oracle to rethink their support organization.

Daniel Fink, the Optimal DBA has fun with SQL and SQL*Plus while dynamically assigning column names. Don Seiler at Die Seilerwerks writes how he used trace 10053 to determine how CBO does its job, and Jakub Pawlowski points to training material about PL/SQL that Steven Feuerstein published in his site.

SQL Server fans continue to blog about SQLServer 2008, Bob Beauchemin writes about SQL Server’s support for extended events. Jeff Smith writes about composite primary keys. I truly hope that everyone already knows about them, but if you don’t – its a must read, and Mladen Prajdić , at I want some Moore, explains how to modify data in a result set that uses joins. I wish he had written this article few month back when I needed it.

Meanwhile, on the OpenSource web:
Ronald, at Technical Notes blog posts a bunch of links for Oracle DBAs learning MySQL and also advice regarding backup and recovery. Last week I received a task to take ownership of MySQL server and write a recovery procedure for it, I can testify that both articles are very useful.
Charlie Cahoon ends in his Summer of Code blog with a release of his MySQL proxy. Corra is already using MySQL Proxy on Ubuntu.
Morgan Tocker and Brian “Krow” Aker try to decide how big transactions should be, and last but not least </depesz> explains how to secure PostgreSQL.

Whew, that has been one long post. I didn’t realize how much interesting things are being written by DBAs every day. There were so many great articles that choosing and picking them for this post was much more difficult than I expected. I highly recommend every database blogger to try writing Log Buffer once or twice, its a unique experience.


Links of the Week

I’ll be doing Log Buffer – Carnival of Vanities for DBAs next week. As a practice run, here are the posts I’ve read and loved last week:

Kevin Closson wrote about configuring lots of DBWR processes, I’ve read the first article, but now he already has two new posts on the subject. He really gives good specific advice on when you need multiple DBWR processes and how to configure your system for maximum efficiency in this case.

Dizwell posted an article going over all the new parameters in Oracle 11g.

Gregory Guillou at Pythian Blogs wrote an excellent article about the SQL Performance Analyzer in 11g, with the most detailed example imaginable.

I’ve discovered utPLSQL, a unit test framework for PL/SQL developers. Guaranteed to improve your life and your code.

Citrix bought XenSource. I think its pretty important news which very few people noticed.

Jeff’s business and technology blog posted What Databases should do for me. A post with significant number of factual mistakes but with an interesting idea about automatic indexes.

You can also find a good article on how to install 11g on NFS over at Oracle-Base.

Enjoy the week and don’t forget to check back on Friday for the real Log Buffer!