Hello everyone,

I have a senior Oracle Applications DBA opening here at Pythian’s headquarters in Ottawa. Working at Pythian is different than working in-house or as a consultant, because you’ll be making your contributions available to each of the customers assigned to your team, allowing you to see more use cases, more technologies, and work with more and varied environments, all the while building interesting and long-lasting working relationships with your peers.

I will gladly sponsor a work visa to come to Canada for the right candidate anywhere in the world.

We support some of the most interesting and mission-critical Oracle Applications environments in the world, including one that is FDA-regulated in support of a global biosciences company. Meaning lives are at stake. It doesn’t get any more mission-critical, or more personally rewarding, than that.

Top criteria:

• Outstanding Oracle Apps DBA on UNIX skills, bonus points for Apps RAC and major implementation and upgrades experience
• Exceptional troubleshooting, problem-solving and learning skills
• Superior productivity per hour and overall getting-the-job-done-right abilities
• Fluent communication skills in English, both written and oral, are mandatory. Second or third languages are also a huge benefit (we have customers all over the world and are always eager to add a language to our repertoire)
• Core DBA skills are a prerequisite for Apps DBA at this level, but you know that already
• Publications, blogging and presentations experience and interest a plus
• Experience with Oracle Applications Server and Portal a plus
• SAP, Peoplesoft and other ERP experience a plus

Job highlights:

• Work in an elite team of Apps DBAs for an elite group and growing of customers; you’ll learn more here in a year than in any in-house DBA job no matter how long you stay; I personally guarantee it.
• Work and gain valuable experience on every mainstream platform, including AIX, HP/UX, Solaris, Linux, Tru64, Windows, etc.
• Support every mainstream database technology and feature, including Oracle RAC, advanced queuing, advanced replication, every flavour of dataguard, RMAN, streams, etc. etc.
• Work across multiple industries including health care, manufacturing, media, dot-com, education, retail, services, and many more.
• Work in a company that values hard work, not long work.
• Work in a company that will allow you to research and write articles, presentations and blog posts on company time, and pay for you to present your research at just about any user conference worldwide where it gets accepted.

Learn more about Pythian and see our customer list at http://www.pythian.com.

To apply:

Send me an email with a one-paragraph introduction of who you are and why you are exceptional to me at vallee@pythian.com. Feel free to attach your resume in any format (Word, PDF, RTF, ODT, whatever makes you happy.)

More information about Ottawa:
Ottawa is a city of around 1,000,000 and is a great place to live.
Encyclopedia: http://en.wikipedia.org/wiki/Ottawa
Cost of living: http://www.finfacts.com/costofliving.htm
Quality of life:
http://www.finfacts.com/irelandbusinessnews/publish/article_1000846.shtml

http://www.canadaka.net/modules.php?name=Forums&file=viewtopic&t=3052 (Globe & Mail article reprint)

Hello everyone,

I think this will be a great log buffer.

Dave has been sick these past two days and as a result, we do not have a comprehensive log buffer ready the way we or a volunteer usually do. This was bound to happen to log buffer at some point and today it has happened.

So I had two choices - cancel this week’s log buffer, or try to make it great despite this adversity. Never one to accept defeat easily, I’ll go for the second option. At least if this is the lamest log buffer ever it won’t be because I didn’t try something new that had a shot at a good result.

So this week’s log buffer is as follows: If you came here to read interesting content from the database community, you probably spend some time each day learning and reading about things we would all find interesting. We are counting on each and every one of you, our faithful readers, to propose the one article you read in the last week (and preferably was written in the last week but I can’t exactly be choosy at this point can I?), and include a short paragraph as to why this article was interesting to you and why it should interest us. Do this in the comments to this post; the akismet spamfilter will tend to eat comments with URLs so please include the magic word “contribution” somewhere in your comment so I can go dig them out for approval.

Feel free to link to your own blog posts if you are proud of them. It is a carnival of the vanities, after all.

A typical log buffer gets over a thousand reads over the week it is active. I am going to call this experiment a success if there are at least 25 interesting links written by the community posted below! And if there are not, well it was worth a try and much better than saying it was canceled.

And so this will be the first log buffer truly written by the readership - join us to make it a good one!

Thanks

Paul
P.S.

Get well soon, Dave

A short note to let everyone know that I’ll be heading to Dubai later today to participate in Pythian’s exhibit in the Business Solutions Hall.

For those of you who haven’t heard of it, GITEX is like COMDEX for the Middle-East - it’s literally the third largest tradeshow in the world where COMDEX is #1 - and it’s big, really really big, like 120,000 attendees big.

I would like to invite everyone who plans to attend GITEX this year to stop by and visit me. Our new COO Roger Hatch will also be there, as will our VP Sales Martin Wisniewski, so there will be no shortage of company executives on hand to discuss your database infrastructure issues and how Pythian might be a help. Or we can just shoot the breeze!

If you’re already in Dubai and want to attend, please send me a note at vallee@pythian.com (I’ll have my blackberry) and I’ll hook you up with one of our free guest passes if I have any left.

If you’d like to find us, we’re in the Business Solutions Pavilion, Hall 6 Stand C6-40, right near the cafeteria (so we hope for good and varied passers-by!), and between Oracle and SAP. Oracle is a huge presenter at this show, as is Microsoft, but MySQL does not yet attend. If you are a MySQL customer and at the show, I have a very good idea of the services on offer over there now (thanks to a budding partnership between us - more to come on that later) and you can drop by at the Pythian booth to find out how Pythian and MySQL together might be able to contribute to your database administration strategy.

Hope to see some of you there.

Paul

There’s an interesting dynamic going on right now in the DBA world. MySQL’s growth and installed base, as a function of its size three or five years ago, is perhaps five if not ten times larger than it was. In 2002 when Pythian’s MySQL services launched, we took on the platform at the explicit request of an existing customer that was primarily an Oracle shop, but that was adopting MySQL for some bolt-on systems. (Pythian trivia moment: That customer is one of ten customers that Pythian has retained from 2002 or earlier right up until today).

Today, MySQL is our fastest-growing practice in terms of new customer acquisition.

The point I want you to take away from that is simply this: there are about five to ten times more high-value environments running MySQL in the world today than there were three years ago.

Most of you working in database administration know this all too painfully already: it’s hard to find an entry-level DBA role in any company. Most responsible technology managers entrusted with high-value environments demand DBAs with a minimum of three to five years production experience. Search craigslist for DBA jobs for a confirmation of this fact.

Together, this means that the population of database administrators that IT managers want to manage their high-value MySQL databases is limited to those DBAs that were already running MySQL three or five years ago. Which is too few to satisfy market demand by a factor of five to ten times.

See my point? (more…)

I just read a fascinating article on clustering architectures for databases from Kevin Closson of Polyserve (now HP). Kevin, for those of you who don’t know him, is a Golden God, at least according to StorageMojo Robin Harris, but all I can say is that he has one of the most informed and incisive views and insights on clustering, SMP, high-availability and high-performance environments in the industry. So, done with the sucking up, you can read Kevin’s bio here, and then turn your attention to his take on the shared-nothing vs. shared-everything debate. Very instructive.

I thought I would share this with the broader community because I think a lot of MySQL, SQL Server and EnterpriseDB folks who need to read this and think about this subject might otherwise miss it, simply because they may not be regular readers of Kevin’s blog.

I am sad to report that my RSS feed for oracle-wtf definitely made me say WTF! but not in a good way. Actually visiting the site confirmed my suspicion, it’s been hi-jacked by a splogger. Take a look and be sure not to reward this guy by clicking anything.

If my memory serves, this has happened to some other high-profile bloggers in our community and although I forget the victims it was a difficult process for them to actually recover the articles; if I remember correctly google’s terrible customer service in this regard actually triggered quite an exodus of blogs away from the blogger platform. Doug - didn’t this affect you?

It will be interesting to see how google handles this now, perhaps they’ve learned something in the last year or two. In the meantime, Tony, Scott, Will, James, Thai and Adrian - best wishes and good luck from all of us here at Pythian. Let us know how you make it out of this mess.

And it strikes me that this might be a good time for all of us bloggers to take a second to verify that our passwords are secure… :-}

As you might imagine, the traffic to the open letter from the oracle.com domain has been spiking in the last few days. Two days ago, in fact, Christo received an email from Oracle putting into question the fact that AWR data collection could not be disabled without a license to the diagnostic pack, and promptly forwarded that note to me.

This email then referred to a MetaLink Doc Id explaining how to do it. And a genuine and polite request was made to please correct the open letter so that at least it was factually accurate in this regard. The thing is, my searches on MetaLink did not turn up any published notes with that Doc Id. Seeing as I’m reasonably well-connected in the Oracle world (ha!) I asked someone inside to take a look and got the gist of the proposed Doc and package, and understood that it’s status was as yet unpublished.

So as it turns out, Oracle has been working on a package to disable the AWR data collection without requiring a license for at least two months. But as of yesterday, it had not yet been published.

And so, yesterday evening, and let’s give Mark Brinsmead most of the credit for the timing of this, Oracle published a method for disabling AWR collection without requiring a license to the diagnostic pack. They nicely gave me the opportunity to announce its existence, and thus behold, note 436386.1, helpfully titled “Package for disabling AWR without a Diagnostic Pack license in Oracle”! (You will need Oracle MetaLink credentials to access that note.)

A couple points now are in order:

• Thank you very much, Oracle, for doing the right thing and making it so that your licensing term for this data is supported by the functionality of your database engine. A week ago, the fact that the restriction was not supported was a substantial stressor and risk from a licensing compliance point of view.
• I think I should highlight that clearly Oracle is reading and responding to our blogs, and this is the first step to integration into the Oracle blogosphere. Whether this is as a result of your efforts, Justin, or not, I think an “‘attaboy and good job” is order here too. Please keep it up, I am looking forward to the point where Oracle users and bloggers are as tightly integrated with Oracle employees and bloggers as they are for some Oracle competitors. (Colour me jealous green!)
• Speaking for myself, I still think Oracle is missing an opportunity by making this instrumentation available to the DBAs, at least at the data layer. There are very good arguments in support of this contention, made by Mark in his letter, and by several of the commentators to the letter. So I still feel this solution is a step backwards onto solid ground, instead of a step forwards onto solid ground.
• That being said, until the licensing terms change, my review of the licensing terms and of this note leads me to conclude that running this on every system where the diagnostic pack is unlicensed is a best practice. That means, if you’re running Oracle 10g and you do not have the diagnostic pack licensed, put this on your agenda. Note that Oracle disagrees with this assessment, in saying
  

  Oracle, therefore, recommends that all customers, with or without Diagnostic Pack license, leave AWR enabled so that they can benefit from features that do not require pack license but implicitly use AWR.

  … but, without meaning to pick a fight in any way, I in turn respectfully disagree with them. The risk, ramifications and costs of a post-hoc licensing liability finding are just too high in today’s Sarbanes-Oxley world.

Of course, the open letter will need a slight correction now. Mark and I will discuss how to handle that later on. In the meantime, if you think the instrumentation should be included in 11g, by all means sign the open letter!

Thanks!

Paul

A short post to draw your attention to this article by Kevin Burton titled “MySQL and the Death of Raid”. Although it’s written from the MySQL point of view, he does bring up some interesting points on the advantages of what he calls a “RAISe” or Redundant Array of Independent Servers” architecture (actually I coined the RAISe acronym just now :-) ) over the traditional RAID approach of hardening the availability and performance of your disk.

Take a look and let me know what you think.

In the meantime, I’d also like to welcome Gregory Guillou to our team here in Ottawa, as he has arrived today. He’s an Oracle specialist, one of the world’s 100 or so OCMs in fact, and has particular expertise in Oracle’s clustering technology, RAC.

This is more of an essay than a blog post, but this subject comes up time and again, and since I tripped across this interesting blog post by Pedro Timóteo about why he has decided not to be a sysadmin any more, I thought now’s as good a time as any to comment on what I think is a significant industry trend in production engineering work.

Pedro here has discovered the essential truth of most sysadmin or DBA jobs: if you’re any good, you will soon be bored and under-appreciated. (That’s a different blog post of his, also worth reading, linked there.)

That’s because the best way to be excited and appreciated in a production engineering role is to let things fail, and then come in riding your white horse to fix things up when everyone notices. There’s only one problem — this is also distasteful to those among us who believe in doing a good job by ensuring things never fail.

Another good way to stay busy is to do everything manually all the time — clone databases, create users, etc. This is also distasteful to those of us who have seen and put to work the huge savings available by automating routine production engineering tasks such as daily verifications.

So what do we do instead? We tune, automate, streamline — in Pedro’s words “some software upgrades here, some tuning there, some cron entries here, some scripting there, some changes to the network, and so on.” Next thing you know, “most of your job is done.”, you’re not so busy any more and there’s plenty of time, which your employers are happy to fill with

dumb, repetitive, non-sysadmin (and therefore non-scriptable) tasks -— which, since you have free time, you probably can’t refuse, or at least feel you can’t. Any raise or promotion will certainly not go to you, but to your “hard-working” co-workers, who are always so “busy” and have so much “work” that they stay at work every day after 6, that they can never do a task “right now”, but only in a week’s time, and that, even their own results are much inferior to yours, it’s you who’re not “dependable”, “dedicated”, or “competent”.

This is the quintessential problem of the best production engineers. It is the impetus bringing the finest of them to come work at Pythian, and driving a large part of our growth.

It’s also a huge piece of the dynamic for most of our competitors’ profit models as MSPs, and the primary differentiator of Pythian vis-a-vis the lot. And finally, it is the underlying factor in why we believe here at Pythian that we are the vanguard of a broad IT shift away from using solely in-house resources for the production engineering functions of database and systems administration. Let me explain.

(more…)

It seems Google couldn’t wait to have their code evaluated for merging into the main source tree, and decided to release it to the general public as patches to 4.0.26 for the community to evaluate.

What have we added and enhanced?

The high availability features include support for semi-synchronous replication, mirroring the binlog from a master to a slave, quickly promoting a slave to a master during failover, and keeping InnoDB and replication state on a slave consistent during crash recovery.

The manageability features include new SQL statements for monitoring resource usage by table and account. This includes the ability to count the number of rows fetched or changed per account or per table. It also includes the number of seconds of database time an account uses to execute SQL commands.

Here’s their blog post announcing the release, a page with more details about the release, and just to make this page a bit useful, here are the direct links to the major releases by feature:

• SemiSyncReplication - block commit on a master until at least one slave acknowledges receipt of all replication events.
• MirroredBinlogs - maintain a copy of the master's binlog on a slave
• TransactionalReplication - make InnoDB and slave replication state consistent during crash recovery
• UserTableMonitoring - monitor and report database activity per account and table
• InnodbAsyncIo - support multiple background IO threads for InnoDB
• FastMasterPromotion - promote a slave to a master without restart