For those of you who have kids and have had a hard time explaining to them how it works, you might enjoy this youtube video. Its only 8 minutes long. When my kids were younger I always encouraged them to watch shows like this. I have to admit I am still a fan of myth busters. 

Take it for what it is.....

 http://www.youtube.com/watch?v=--FTr0vD1xQ

Posted by Michael Corey

www.ntirety.com

I saw this blog entry on neworacledba. To reach the original blog entry follow the link. I saw quite a few useful entries in this blog. I would encourage you to check out the blog http://neworacledba.blogspot.com/2008/06/asm-instance-creation.html . 

Here is a copy of the Blog entry

ASM instance is used to manage database files.If a node hosts multiple databases,single database,files can be managed using ASM.

Two possible ASM deployments are:
1) Seperate oracle home is there are multiple databases on a single node
2) No seperate oracle home if the node hosts a single database.

ASM instance needs to be created and started before the Oracle database instance is started.

css - cluster synchronization service daemon is used to establish connection between an ASM instance and oracle database instance.

To startup ASM instance alone and restrict the connection to Oracle database instance startup ASM instance as follows:

SQL>STARTUP RESTRICT;

Steps to startup an ASM instance are as follows:

1) Check is css is running using
ps -efgrep css

We can also use crsctl command to check if css appears healthy
crsctl check css

If css isn't up and running startup ASM as follows
1) Login as root user
2) export PATH=PATH:$ORACLE_HOME
3) localconfig add

The localconfig command creates /etc/oracle file,startsup css.The $ORACLE_HOME/css/admin has control script for css startup and shutdown.

It adds init.cssd to /etc/inittab file.

If we use DBCA to create an instance css is started automatically.

2) Create a initialization parameter file initasm+.ora with the following parameters set.

Instance_type=+ASM
ASM_POWER_Limit =Value from 1 to 11
ASM_Diskstring = '/dev/sda*',
ASM_DISKGROUPS=dg1,dg2

3) Startup the instance.ASM doesn't have a data dictionary.It is mandatory to login as a user with SYSDBA/SYSOPER privilege.
For local authentication use Operating system(OS) authentication.
For remote authentication use a password file.

4) Set ORACLE_SID=+ASM (ASM is the default value).The value set for instance_type is mapped internally to db_unique_name.

$sqlplus / nolog
SQL>connect / as sysdba
SQL> startup pfile='/asd/initasm+.ora'

This step creates an ASM instance.ASM cache is the SGA component specific to ASM.It takes as little as 25MB- 30MB space.Total spce needed for an ASM instance is 100MB as it doesnt have any datafiles.

5) Check is ASM instance is up and running using :

ps -efgrep asm

The step above lists background processes like:

asm_pmon_+asm
asm_smon_+asm
asm_lgwr_+asm
asm_ckpt_+asm
asm_dbw0_+asm
asm_mman_+asm

Few special is ASM like

asm_rbal_+asm
asm_psp0_+asm
asm_gmaon_+asm

Posted by Michael Corey

www.ntirety.com

I just saw this online. Its from  Michael Arrington at Techcrunch. I have included a section of this article. 

We’ve got multiple sources at both Yahoo and Microsoft telling us that official talks are back on between the two companies. But we’re hearing something different than CNET - the talks are about a full buyout again, not a sweetened search-only deal.

The information we have is thin, but what one source is saying that Microsoft is talking a price lower than the $33 they were offering when the talks disintegrated in May. Given Yahoo’s recent share price (it’s below $21 today), and the fact that just about everyone other than their board and top execs are publicly screaming for a deal, I’m not surprised.

Microsoft official comment is “no comment,” which actually contains more information than it appears to. For well over a month, Microsoft has officially been saying they’re no longer interested in Yahoo. They didn’t say that today.

To read the full article click here. 

http://www.techcrunch.com/2008/06/24/sources-microsoft-and-yahoo-talks-back-on/ 

Anything that can challange google is important to the future of Information Technology I am a big fan of Yahoo. I use the site quite a bit, especially yahoo messenger. I have a number of friends around the world I keep in touch with. 

Posted by Michael Corey

www.ntirety.com

I read an interesting blog entry titled “The Sin of Band-Aids” by the Oracle Alchemist. It started like this…..

Let’s take a trip to the past…we weren’t actually there, but we can pretend. Let’s look into a conversation between a DBA, developers, and management at some time ‘way back when.’

    Management: Our system has been performing poorly for months now.
    Developers: We didn’t change anything.
    DBA: I didn’t change anything.
    Management: We need a fix.
    Developers: We don’t have time to go through all our code and tune it.
    DBA: *frantically types: alter system set optimizer_index_caching = 35; alter system set optimizer_index_cost_adj = 25;* Fixed!

God does that sound familiar. How many times as a DBA, have we made an adjustment to the database as a Band-Aid fix to an application problem. We tell management it’s a band-aid, that eventually wont work, but thed don’t here that part.

The blog entry goes on to say…

Fast forward…

Queries are not performing as expected. Sometimes a query will run fine, sometimes it runs horribly. Sometimes a combination of queries and users results in massive buffer busy waits or ‘latch: cache buffers chains’ waits. Nothing is working and so finally, a consultant is called.

You’ve probably guessed that I was the consultant. Move to the head of the class! In particular, there were three or four main offenders; queries that would sometimes perform well, then all of a sudden ‘turn horrible.’ The core of the problem was truly unstable execution plans.

Using ASH I was able to go through different times of day and compare reports from times where the system performed well with times the system performed horribly. Doing so allowed me to compare SQL_IDs between different times of day to see which queries seemed to be going through dramatic changes. Once a SQL_ID was found that experienced such a dramatic change, I used DBMS_XPLAN.DISPLAY_AWR to find all the different iterations of the query.

In case you’ve never used it, DBMS_XPLAN.DISPLAY_AWR is very useful. It allows you to input only a SQL_ID, and Oracle will show you the explain plans for that SQL_ID as recorded in the AWR. For instance:
select * from TABLE(dbms_xplan.display_awr(’95t7cmj84u4jz’));

The next step was to look through the results to find any inconsistencies large enough to result in huge differences in performance. In this situation, I found that ‘good’ explain plans were using decent indexes, NESTED LOOPs, and low overall bytes. The ‘bad’ explain plans were using index joins, where an index fast full scan would join with an index range scan into a view (e.g. index$_join$_003). These plans resulted in many more block touches and much more CPU used.

Using Grid Control, I looked at the timeline for each of the hash values of the SQL_ID and saw that the poorly performing plans lined up with their bad performance periods.

Of course, now the burning question: what can be done about this?

Well it’s the Oracle Alchmist blog entry not mine, only right that you go to the source to see the answer to the question and also his conclusions.

http://www.oraclealchemist.com/oracle/the-sin-of-band-aids/

Posted by Michael Corey

www.ntirety.com

I just saw a useful white paper I thought I would share. Its from Microsoft and compares SQL Server 2008 and Oracle 11G. I am a big fan of SQL Server andwhat a tremendous vaue it represents. At Ntirety we support a large number of SQL Server databases, its a great product and only gettting better and better.

This is from the Microsoft Site:

I saw an interesting article in the BBC. It came out June 16th, 2008. To quote the article

“More bloggers than ever face arrest for exposing human rights abuses or criticising governments, says a report.

Since 2003, 64 people have been arrested for publishing their views on a blog, says the University of Washington annual report.

In 2007 three times as many people were arrested for blogging about political issues than in 2006, it revealed.

More than half of all the arrests since 2003 have been made in China, Egypt and Iran, said the report.”

It goes on the say “The report said the rising number of arrests was testament to the "growing" political importance of blogging.”.

It says the average jail time was 15 months, with the longest sentence being 8 years. To read the original article

http://news.bbc.co.uk/2/hi/technology/7456357.stm

Being an active Blogger it is changing the world and how we exchange information. I am amazed at who reads my blog and what part of the world they are from. A particular favorite blog entry of mine was the
How to “GO Green” As A Database Administrator (DBA).
http://michaelcorey.ntirety.com/blog/tabid/51101/bid/5335/How-to-Go-Green-As-A-Database-Administrator-DBA.aspx

The abilty to share my thoughts with people all over the world is amazing. I published another blog entry World’s 50 Most Powerful Blogs (European Style)
 

http://michaelcorey.ntirety.com/blog/tabid/51101/bid/4190/World-s-50-Most-Powerful-Blogs-European-Style.aspx

it was from the guardian.co.uk, another UK based article. I found the list quite captivating, and a few of those blogs are now on my must read list.

It is clear to me Blogging is part of our society around the globe in a big way. The fact that people are still getting arrested all over the globe for expressing their opinion, makes me realize just how lucky I am to live in the United States of America. I don’t always agree with what everyone says, but I am glad that here at least you are free to express your opinion without fear of being locked up for an average of 15 months.
 

Posted by Michael Corey

www.ntirety.com

I was on my way home from a seminar series we recently did jointly with Microsoft. At this seminar I presented "Choosing a DBA Managed Service instead of a consultant” “which is based off the research I did for my white paper Choosing a Remote DBA Vs Consultant Link on the topic trying to better understand what companies benefit most from a remote DBA service.  The other presenter was Tony Piltzecker of Microsoft who presented “What’s Coming in SQL Server 2008”.  Which looks great I might add. It looks like Microsoft is taking another quantum leap for its SQL Server database

I am on my way home, after many stops. It’s late and I go to purchase gas. The Price off the Highway was almost $4.00 dollars a gallon. I think to myself this is the second time this week; I have had to fill this tank up. On the radio shortly after that is a discussion on the price of gas. Yes I listen to talk radio. Tonight was WBZ AM Radio 1030.  They were equating the current gas problem to the TULIP Craze, The Internet Boom and so on. How this price of GAS is a bubble that will eventually burst.

I think even if that’s true, it’s a bubble that is still going to hurt a lot before it does burst. It got me thinking, which I am sure it has a lot of other people thinking. What can I do to “GO Green”? Ntirety business is completely Remote Database Administration. So to answer this question, I have to answer to questions.

1.    What does it mean to “GO GREEN” with your Oracle database administration
2.    What does it mean to “GO GREEN” with your SQL Server database administration

Given that fact we are 100% focused on database Administration. What does it mean to “Go Green” as a database administrator. So I am looking for thoughts, ideas, and suggestions on how to “GO GREEN” as a database administrator.

As a company we handle databases all over the world for lots of different companies. We have built an infrastructure that allows us to work completely remote. So in once sense we are “Going Green” because our database administrators do not need to drive into the office every day and pay $4.00 a gallon for gas to do their jobs.

Database Administrator (DBA) Go Green Suggestion 1: “Set up your infrastructure so you can work remotely in a safe and secure manner. Eliminating the need to go into the office to fix things that could be done remotely.

How many times when I was working as a consultant for a companies database. The organization would have a power failure. The Server and databases would not come up cleanly after the power failure. The first thing I did was create a startup procedure that would ensure on a reboot of the server everything would come up clean. I would also set up a variable to be set. In case I did not want the database to start up on system reboot. So my seconds go green suggestion.

Database Administrator (DBA) Go Green Suggestion 2: Set up procedure so the database automatically restarts itself upon a power failure or server reboot. This will prevent you getting having to deal with it. Which means a trip into the office, or other activities that will waste time and energy.

How many times have you seen reports printed that the users never looked at. The worst case of this was when I worked for Honeywell Corporation. We had a report printed for a user every week, which took 7 boxes of green-bar paper. When I question who needed a report every week that was 7 boxes thick. I guess even back then I was a “GO Green” database administrator. I learned that they only looked at the last 10 pages. My next “GO Green” suggestion.

Database Administrator (DBA) Go Green suggestion 3: Eliminate unnecessary paperwork. If it can be done electronically, do it that way.

How many DBA have multiple screens on your desk. So they can watch the progress of what is happening on the various systems. Do you really need that? We have to change the way we do things, so we are more energy savvy. So my next DBA Go Green suggestion

DBA GO Green Suggestion 4: If you are not using a piece of equipment shut it off.

I am hoping that reading this blog entry will get every one thinking. I am looking for ways that a Database Administrator can GO Green in their jobs. Please take the time to share any ideas you have no matter how big or small.

Posted by Michael Corey

www.ntirety.com

I just saw an article by Ryan Paul on the Internet Explorer team at Microsoft sending a cake to the Firefox 3 team at http://arstechnica.com/journals/linux.ars/2008/06/17/the-cake-is-a-lie-ie-team-bakes-a-treat-for-mozilla

“In honor of today's Firefox 3 release, Microsoft's Internet Explorer team gave a cake to Mozilla. The tasty treat, which prominently displays IE's blue "e" icon, just arrived here at Mozilla headquarters in Mountain View.”

“The cake congratulates Mozilla on shipping Firefox 3 and expresses love from the IE team. Microsoft's IE developers also sent a cake to Mozilla in 2006 following the release of Firefox 2, so this is clearly becoming a tradition.”

Over the years I have seen many such things happen like this. They always make me smile. It reminds us all what a small world we live in. The article goes on to say “Closer cooperation between Microsoft and Mozilla on standards and web innovation would be a big win for everyone.”.

If you enjoyed that article check out http://www.arcanology.com/2008/06/17/ie-sends-mozilla-a-new-cake-for-firefox-3/   There are a few more pictures you might enjoy, plus see a few more faces behind Firefox 3.

Posted by Michael Corey

www.ntirety.com

 The June meeting of the Boston MySQL User group featured Keith Murphy, Editor of the MySQL Magazine (www.mysqlzine.net). Talking about Mysql backups.

Here are some of the links mentioned in the Video

To play the Keith Murphy Video on MySQL backups click here

Click here to download the Keith Murphy Backup Video located at http://technocation.org (351 MB)

Mylvmbackup by Lenz Grimmer

http://lenz.homelinux.org/mylvmbackup/ http://lenz.homelinux.org/mylvmbackup/.


mylvmbackup is a tool for quickly creating backups of a MySQL server's data files. To perform a backup, mylvmbackup obtains a read lock on all tables and flushes all server caches to disk, creates a snapshot of the volume containing the MySQL data directory, and unlocks the tables again. The snapshot process takes only a small amount of time. When it is done, the server can continue normal operations, while the actual file backup proceeds.

The LVM snapshot is mounted to a temporary directory and all data is backed up using the tar program. By default, the archive file is created using a name of the form backup-YYYYMMDD_hhmmss_mysql.tar.gz, where YYYY, MM, DD, hh, mm, and ss represent the year, month, day, hour, minute, and second of the time at which the backup occurred. The prefix backup and the date format can be modified. The use of timestamped archive names allows you to run mylvmbackup many times without danger of overwriting old archives.

Alternatively, instead of tar, you may use rsync. This process is nearly identical, with the exception that the file suffix is not used.

INNOBASE – INNODB Hot Backup
 http://www.innodb.com/hot-backup/order/


InnoDB Hot Backup is the ideal solution for online backups of InnoDB tables in MySQL—and for setting up replication. It allows you to back up a running InnoDB database under MySQL without setting any locks or disturbing normal database processing. You get a consistent copy of your database, as if the copy were taken at a precise point in time. InnoDB Hot Backup is also the ideal method of setting up new slaves if you use the MySQL replication on InnoDB tables.

Prices are at:

Pricing
Price (excluding VAT)
EUR         USD
1-Year License         € 390     $ 605
Perpetual License     € 990     $ 1540
1-Year Email Support     € 590     $ 915

Zmanda Open Source Backup
 http://network.zmanda.com/shop/home.php?cat=1,3

Zmanda Recovery Manager (ZRM) for MySQL simplifies the life of a Database Administrator who needs an easy-to-use yet flexible and robust backup and recovery solution for MySQL server. With ZRM for MySQL you can:

    * Schedule full and incremental backups of your MySQL database.
    * Start immediate backup or postpone scheduled backups based on thresholds defined by you.
    * Choose to do more flexible logical or faster raw backups of your database.
    * Perform backup that is the best match for your storage engine and your MySQL configuration.
    * Backup your remote MySQL database through a firewall.
    * Configure on-the-fly compression and/or encryption of your MySQL backups to meet your storage and security needs.
    * Get e-mail notification about the status of your backups and receive MySQL backup reports via RSS feed.
    * Monitor and browse your backups.
    * Define retention policies and delete backups that have expired.
    * Recover a database easily to any point in time or to any particular transaction, e.g. just before a user made an error.
    * Parse binary logs to search and filter MySQL logs for operational and security reasons.

Posted by Michael Corey

www.ntirety.com

One of the things I like best about writing a blog is that I am able to take it anywhere I want. Sometimes we talk about hard-core issue concerning Oracle databases, other times, we talk about important social issues like the GO Green DBA.

Go Green DBA

A blog entry I hope would get us all thinking about how we can do our part to make a difference.

This latest entry is about a web site I was made aware of that caught my attention. It’s a database of Police speed traps all over the United States and other parts of the world. I checked it out for my local town and it was quite accurate. Its amazing the types of information that is stored in a database and the uses of it. Be it an Oracle database, SQl Server database or some other vendor product.

So here is website link to help you find that speed trap and avoid the insurance surcharge we all hate.

Speed Trap Database
 

Posted by Michael Corey

www.ntirety.com