Log Buffer #140: A Carnival of the Vanities for DBAs

This is the 140^th edition of Log Buffer, the weekly review of database blogs. Welcome.

Let us begin with Oracle this week. Dan Norris illustrates how to start database services automatically after instance startup. He says, “Services are an essential component for managing workload in a RAC environment. If you’re not defining any non-default services in your RAC database, you’re making a mistake.”

Riyaj Shamsudeen also was sleuthing—into dynamic_plan_table, x$kqlfxpl and extreme library cache latch contention. Dion Cho, the Oracle Performance Storyteller, responded with an item about a similar experience of his: heavy reads on the library cache related v$ views.

From detectives to cloak-and-dagger. Tanel Poder reveals the real history of Oracle database! X$KZSRT! X$KCPXPL! CCCP! KGB! CIA! Wow!

Pythian’s Grégory Guillou published How To Access MySQL from Oracle With ODBC and SQL.

There’s some big MySQL news floating about in the offing, and Domas Mituzas provides some relevant links and a place to talk about it: the possibility of IBM acquiring Sun Microsystems. It’s just speculation and rumour now, but the very promise of it is too tantalizing to ignore. I imagine that the employees of MySQL—so newly moved to their new home at Sun—trembling at the prospect of another transition. Here are Craig Mullins’s thoughts: Will IBM Acquire Sun?

On the technical side now, Yves Trudeau, a Big DBA Head, argues in favour of high-performance replacement of the MySQL Memory storage engine with NDB.

On MySQL Insights, Ivan Novick looks into the matter of auto increment stability: “The message here is not that the implementators of auto-increment have not done a good job, but that creating this functionality is actually really difficult. . . . I would recommend against using (it) in large-scale complex applications, unless you really know what you are doing and are accepting the limitations that exist.”

MySQL-dump has a post on DELETE, innodb_max_purge_lag and a case for PARTITIONS

Venu Anuganti was also on the trail of partitions. He examined MySQL’s 5.1 Mix Of Handlers In Partition.

Arjen Lentz wants to hear your SAN and VM horror-stories, please. He also shares his thoughts on Oracle (and MySQL), enterprise, suitability and sense, a response to Baron Schwartz’s “50 things to know before migrating Oracle to MySQL,” which I mentioned here last week.

Ido Kanner has tried quite a few different DBMSs, offers Why I find MySQL to be so bad, the first in a series.

In the SQL Server world, Paul S. Randal returns with the results of his survey on the importance of validating backups.

The SQLBatman swoops over NOCOUNT and User Options: “How do I disable (NOCOUNT) for only one database on an instance? The answer is you don’t, at least not for MS SQL.”

On the Journey to SQL Authority, Pinal Dave gives a through demonstration of the difference between union vs. union all.

Paul Nielsen and his readers discuss multiple assignment variables and order by in SQL Server 2008. “Though nearly undocumented, (multiple assignment variables have) been around since at least SQL Server 2000. (They’ve) been bashed as a kludge and hack by some and revered as the embodiment of elegance by others.”

Linchi Shea and commenters have an interesting discussion of the value of a DBA utility database in every SQL Server instance. Linchi writes, “I’m curious how you manage your maintenance objects out there in the real world. Do you put them all in a utility database or distribute them in the user databases?”

Louis Davidson writes, “File this under the ‘I can’t believe there is still stuff that I keep learning about SQL Server 2005!’ though thankfully most things I find I learn are things I wouldn’t be all that likely to use.” The issue is: granting rights to all objects in a database.

From the PostgreSQL world, Dan Langille announces the release of PGCon 2009 Schedule.

Joshua Drake offers a funny remembrance of excellence from past PostgreSQL conferences. With LOLcats, even.

On his select * from depesz;, Hubert Lubacziewski examines set operations in the shell.

Josh Berkus and his readers variously ask and answer the question, how do I examine the Linux page cache? “I thought, if PostgreSQL has easy tools to monitor the contents and use of its cache, Linux must have some tool to examine the contents of the page cache, right? Wrong.”

On the DB2 Express-C Team Blog, Leons Petrazickis asks for your participation in a survey on DB2 Express-C download experiences.

Here’s Craig Mullins again, this time on his DB2PORTAL Blog with a short introduction to lock avoidance.

And that is that. Please add to the comments your favourite DB blogs from this week, and/or send me an email if you’d like to publish an edition of Log Buffer on your own blog. Till next time!