Category Archives: Uncategorized

#DOAG2016, definitely a crazy week.


#DOAG2016, the largest Oracle Community gathering in Europe. Taking place in Nuremberg, at the Nuremberg Convention Center NCC, one of the more impressive places to hold such a conference, towering 4 stories high, with a big central atrium!!
It is a huge effort to get all of this together!

In this blog-post I want to highlight some of the crazy things I experienced this week… And… I did try to follow my own schedule, but I wasn’t overly successful.

Young talent

One of the things that was somehow quite clear this week, is that we have a lot of young talent out there, eager to learn and share experiences. It is not just the #NextGen “movement” of DOAG, of which Carolin Hagemann made me aware, but just young people on the conference itself.

Discussing “Young PL/SQL” at the unconference session made us all aware that our part of the IT trade is no very sexy and popular with the youngsters. This all despite what was mentioned above. In universities we train SQL, but we don’t train to create real-life business applications, leveraging the power of the one language that keeps SQL close to the data it feasts on, PL/SQL. But, more on that below (Thick Database Paradigm).
To promote PL/SQL, basically two ground requirements were defined:

  1. Create a free ‘PDB as a Service’ for schools;
  2. Inspire teachers to talk about data centric computing

By finding somebody to be regionally or globally owner of this quest, it should be possible to get young professionals as familiar using PL/SQL for creating performant and business-ready applications as they were familiar using Microsoft Excel to do their accounting “back in the days”

ACE program

“There is a disturbance in the force!”

For everybody not acquainted with the Oracle ACE Program by the Oracle Technology Network… You should be!! Please read up, as it is an incredible cool initiative.

The disturbance, you ask?
Well, to retain your “status”, Oracle expects you to do “stuff” and this “stuff” is then evaluated on a yearly basis. Basically the initiative, the disturbance, is to get some transparency in “the stuff”. And, as always, everybody wants change, but few actually are good at “change”. There are rimples and things that change, but in the end; everything will be fine, unless, obviously, when it will not be fine.

Talks

I was honored to (co)host to talks at #DOAG2016:

Bad Boys of Replication – Changing everything…
With Oracle ACED and good friend Björn Rost, about an intense migration project we did some time ago. We were even offered to host our talk in Tokio, the biggest hall at DOAG!

Saving lives at sea at an industrial scale using Oracle Cloud Technology
An insightful (at least I like to think so) talk with my colleague Oliver Limberg. The talk is about the rapid development of a global portal for the maritime logistics branch.

I had a blast, and I hope you did too!

Community spirit

Oracle User Group conferences are about sharing and are about fun. Mr. Martin Widlake wrote a good post about that.

Apart from all the “more formal” things that happened, there were quite a few extracurricular activities, mostly involving an Irish Pub or a restaurant.

This all may sound quite funny and exciting, and, yes, it is alto talk with your co-workers: “Oh, hey, you are going to have fun and party all week!” Of course it is not a drag and a bore, but it has very profound function!
Whenever you run into trouble, these are the exact same people that are not only able, but probably also inclined to help you out, as you would help them out, as friends do among each other. In the end, they, you, your boss and your clients benefit. This is not to be underestimated too much.

The extra, special bit, that DOAG offers are the so called “unconference sessions”.
Not scheduled, no slides, nothing official, just getting together and discussing subjects of interest. Our “Young PL/SQL” was one of these “unconference session”, which turned out to be a great (and valuable) success!

Meeting people

Just to name a few, heroes of long and of yet to come for #DOAG2016:

Dietmar Neugebauer
Frank Dernoncourt
Joel Kallman
Johannes Ahrends
Kamil Stawiarski
Laurent Leturgez
Maja Veselica
Marcel Hofstetter
Piet de Visser
Sabine Heimsath
Stefan Kinnen
Stew Ashton
Uwe Hesse
Zoran Pavlovic
And alle the ones I forget to mention here!!

Thick Database Paradigm

Noting new in IT…

Well, no.

The Thick Database Paradigm (opposed to the “No PL/SQL Paradigm”) is nothing new. We have actually all been doing this since the eighties. Program your business rules, your constraints, everything that makes sure that your data is all that you want it to be, close to that data.
There are so many reasons that speak in favor of this approach that it is nearly overwhelming and deserves at least a book in itself. But, let me make a small attempt to highlighting a few here:

  • spare yourself network bandwidth, by not sending data all over your network to be processed
  • safeguard your data inside the (Oracle) database, so it can be protected by all that has been invented to do so
  • Transact data where it lives and combine and aggregate it there, you will be amazed by the efficiency
  • Remind yourself why you used to think “business logic in middle teer” was a good idea

If you leave possibile religious believes aside, there is no other conclusion possible then that the reinvention of “Thick Database” is the (re)discovery of 2016, right from the time when IT still made sense.

Yes, there are cases where an “Enterprise Service Bus” makes sense, but, as with every technology withing IT, it has a very specific area where it actually adds value or even makes sense. At best, a lot less than all the places where it is used currently!
Not to get carried away in this joyful blog-post, I will leave this topic at this.

The end

I hope to see you at the next Oracle User Group conference, somewhere… Please watch for the asterisk at his page for the conferences that I will attend.


#doag2016 my picks and suggestions

As many conferences evolve over the years, the number of sessions on offer can easily be overwhelming. I have overheard many conference attendees wrestling with their choices for what to see and which sessions to attend.

For DOAG 2016 I have a short overview with my picks and with one or two tip-sessions. I hope this helps, though it is just my personal preference of course…

Please note that this post is based on the printed version of the conference planner and this may obviously be subject to change. Find the on-line version of the conference planner here!

Tuesday, November 15th
08:30
Goto-session
Connecting Oracle & Hadoop by Tanel Poder
Tip
Structuring an APEX Application by Alex Nuyten
Meet your match: Advanced Row Pattern Matching by Stew Ashton

11:00
Goto-session
How to identify the Right Workload for Database In-Memory by Andy Rivenes
Tip
Die Schlechten ins Kröpfchen – SQL analyse für DBAs by Martin Klier

12:00
Goto-session
Was die IT von der Luftfahrt lernen kann by Uwe Küchler
Tip
Using image copies for Oracle Database Backup by Ilmar Kerm
Using SQL Transaction Framework to rewrite Bad SQL on the fly by Kerry Osborne

13:00
Goto-session
Plötzlich Multitennant – was ändert sich für den DBA by Uwe Hesse
Tip
Oracle VM auf Exadata – Erfahrungen aus der Praxis by Christian Pfundtner
Einsatz von Maps in APEX by Denis Kubicek

14:00
Goto-session (TOP-tip)
–> Session got cancelled, but will be at UKOUG!
Patch you application with No Downtime (& No extra Costs!) by Oren Nakdimon
Tip
Hacking Oracle’s memory – About Internals & Troubleshooting by Stefan Koehler

15:00
Goto-session
XML in der Oracle DB by Wolfgang Nast
Tip
PL/SQL Performance – Best practices für Laufzeitoptimierung by Jan Gorkow

16:00
Goto-session
The Oracle Optimizer – Upgrading Without Pain by Nigel Bayliss
Tip
Erfahrung nach einem Jahr Fusion Middleware 12c by Jan-Peter Timmerman

17:00
Goto-session
Active Session History: Advanced Analytics by David Kurtz
Tip
MySQL for Oracle DBAs by Philipp Michaly
Deploying PL/SQL Applications, Building Rome in a Day by Alan Arentsen

Wednesday, November 16th.
08:00
Goto-session
Logical Replication in 12cR2 – What are the options now? by Vit Špinka
Tip
Function madness: Use and Abuse of PL/SQL Functions by Piet de Visser

09:00
Goto-session
Ensuring your Physical Standby is Usable by Michael Abbey
Tip
RMAN – From Beginner to Advanced by Marcin Przepiorowski

10:00
Goto-session
Oracle Secure Backup – eine Livedemo by Sven-Olaf Hilmer
Tip
Oracle Hacking Session by Kamil Stawiarski
Advanced Interactive Grids by Patrick Wolf

11:00
Goto-session
The Battle: Linux vs. Windows by Dierk Lenz, Johannes Ahrends and Martin Klier
Tip
Adaptive Features or: How I Learned to Stop Worrying… by Ludovico Caldara
Controlling Execution Plans – Workshop by Kerry Osborne

12:00
Goto-session
Application Express für den DBA? Geht das? by Joel Kallman
Tip
Und Sie bewegt sich doch by Lothar Flatz
APEX Desktop Apps – Interaktion mit dem Client System by Daniel Hochleitner

13:00
Goto-session
Hash Joins and Bloom Filters by Toon Koppelaars
Tip
Ansible für Oracle DBAs by Alexander Hofstetter

14:00
Goto-session
Delivering Continuous Availability for Database Services by Michael Timpanaro-Perrotta
Tip
Dbvisit – Oder doch lieber Data Guard by Andreas Kother
Chase the Optimizer Every Step of the Way by Mauro Pagano

15:00
Goto-session
Top 7 Plan Stability Pitfalls & How to Avoid Them by Neil Chandler
Tip
Advanced RAC Programming Features by Martin Bach
Weblogic 101 for DBA by Osama Mustafa

16:00
Goto-session
Bad Boys of Replication – Changing Everything by Björn Rost and yours truly

17:00
Goto-session
Oracle System Statistics by Paul Matuszyk
Tip
Compression – Technik und sinnvolle Umsetzung by Klaus Reimers
Node.js der Alleskönner by Kai Donato

Thursday, November 17th.
08:00
Goto-session
FAQ about Masking Sensitive Data in Oracle Database by Maja Veselica
Tip
Data Guard in Oracle 12.2 – Crash Course by Zoran Pavlovic

09:00
Goto-session
Mining the AWR v2: Trend Analysis by Maris Elsins
Tip
Regular Expressions: Say What? by Alex Nuyten

10:00
Goto-session
Databases Clone Using ACFS by David Hueber
Tip
R.I.P. Oracle Database by Markus Lohn

12:00
Goto-session
Writing Efficient SQL Statements by Joze Senegacnik
Tip
Validate User Input in APEX by Richard Martens

13:00
Goto-session
Backup und Recovery PoC auf der Recovery Appliance by Frank Schneede
Tip
Ready, Steady, GIT: Einführung eines Versionskontrollsystems by Carolin Hagemann

14:00
Goto-session
Warum sollte man die Multitennant Database Option Verwenden by Johannes Ahrends
Tip
Collections in PL/SQL by Frank Haney

15:00
Goto-session
Saving Lives at Sea – At an Industrial Scale Using Oracle Cloud Technology by Oliver Limberg and yours truly
Tip
Part 1: The NoPL/SQL and Thick Database Paradigms by Toon Koppelaars and Bryn Llewellyn

16:00
Goto-session
Part 2: The NoPL/SQL and Thick Database Paradigms by Bryn Llewellyn and Toon Koppelaars

And!!
Do not forget…
The first ever APEX Hack’a’thlon is going down on Friday the 18th of November at the DOAG Education day. If you are interested or just want more information, don’t hesitate to drop a line.

@HrOUG_2015 in Rovinj, Croatia

In a hectic year it is good to attend and contribute to Oracle user group sessions. This adds an element of a ‘Working Holiday’ to someones schedule. I can promise you, the vacation isle of Rovinj is a perfect venue for this and especially since it is the last week of the opening of the Hotel for this season.
Of course you can find all information about contributing to these events right here!!

@HrOUG_2015, as the official twitter-account of the conference goes, brings just this!! Content combined with pleasure. Ranging from quality sessions by Rock star speakers to relaxation in the pool and late night party in “The Castle”.

Currently the biggest worry is rain… At least for the attendees. As always the (very) hard working organizers are doing their best to create a super experience for everyone attending the conference, and my personal biggest worry is that the participants will actually bring their laptop to the hands-on experience. Actually doing logical replication yourself is so much cooler than seeing it demonstrated. It will be an interesting experience anyhow.

This conference also led to another series of Oracle Hero’s I got to meet in person!

And as always there is really serious stuff going on as well. One of the main challenges or worries today is the developments surrounding Oracle Database Standard Edition Two, and the impact it brings for the development of the European market.
Eliminating this database version forces emerging projects to use the Oracle Cloud, as the super-sharp priced project startup version is no longer available. We had this with Standard Edition One. It also counters Oracles own statement, quite recently presented by Andrew Sutherland, of hybrid cloud functionality, since there is no “on-premise” equivalent for a small scale project anymore!
We are hoping for a good discussion on Friday during the Standard Edition Round Table version at HrOUG, co-hosted by Philippe Fierens, as this development is very heartfelt in Croatia as it is in many European countries.

If you want to read more about this years event in Croatia, please checkout the many tweets and facebook entries by @helifromfinland, @alexnuijten, @roelhartman (ps. Vote for Roel as member of the ODTUG board) and many more!!

Oh, and as far a basic life’s needs go… The Internet on the island is the best ever!!

Updating SQL Developer to use newer Java version

I was being teased by SQL Developer.

Everytime I started it came nagging about that it was being forced to live in an old Java version called jdk1.7.0_45 and that is was not feeling happy about it.
So, I should remedy this, I thought to myself.

First visit was, inspired by some search-work on the WWW, a file called product.conf. Which offered two possiblities:

java

SetJavaHome to some logical location
or
SetJavaHome to nothing, and then SQL Developer would kindly ask me to point it to somwhere to live.

Well… no. My SQL Developer refused it all and just started with this jdk 1.7.

Same hack done in another file on another location, a file called sqldeveloper.conf.
Same result.

Freshly downloaded SQL Developer, put in place… No help!

Erm…

Rename
drwxr-xr-x  3 root  wheel  102 Jan  6  2014 jdk1.7.0_45.jdk
in /Library/Java/JavaVirtualMachines
to
drwxr-xr-x  3 root  wheel  102 Jan  6  2014 xxx-jdk1.7.0_45.jdk

Nope! Still the same nagging…

What now?

In the end, I wound up with one of Jeff Smits’s helpers.
This guy aksed me to “start SQL Developer from the commandline”. Right, but how?

So I finally found:
/Applications/SQLDeveloper.app/Contents/MacOS/sqldeveloper.sh

And that did start SQL Developer from the command-line…

But… wait… an .sh-file!! Interesting!!

And, behold… in this .sh-file lies the answer:sqldev_startup1

So the file reads:
export JAVA_HOME=`/usr/libexec/java_home -v 1.7`
Which I hacked to:
export JAVA_HOME=`/usr/libexec/java_home -v 1.8`

And, presto, error-message gone and SQL Developer now happily lives in Java 8.

Hope this helps somebody out!!

Communication limit with mod_plsql through PlsqlMaxParameters

There is always a lot to do when you run your page. And, somehow, these days there are some default limits to the amount of work you are supposed to do.
Okay, but what does this mean…

In an application, we have a screen to create some appointment between a bunch of people. This bunch can become quite substantial indeed.
In those days we were running Oracle HTTP Server 9i to lift mod_plsql up in the IP, and we never had any issue creating these somewhat larger appointments in our application.

Recently we got a call on our service desk, which said there were some issues in creating an appointment with a greater number of participants. The message read, somewhere in the error-text:

HTTP-400 Too many arguments passed in. Got 2009 parameters. Upper limit is 2000

Okay, somehow this application function is not going to work!

On research we identified the parameter PlsqlMaxParameters to be the problem. This parameter was introduced with the first version of Oracle HTTP Server right after 9i. If unset, PlsqlMaxParameters defaults to a value of 2000.
I am guessing here, but my bet would be this to be a restriction to prevent a buffer-overrun of some kind.
To resolve the issue the PlsqlMaxParameter needed to be added to plsql.conf. (We tried to do this in dads.conf to keep most of Oracle HTTP Server unaffected, but got errored out).
We chose the value of 4000, meaning a 100% increase while maintaining a fair and safe limit.

For Oracle HTTP Server, the parameter is set in plsql.conf which can be found in either:

  • $ORACLE_BASE/Oracle_WT1/instances/instance1/config/OHS/ohs1/moduleconf
  • $ORACLE_BASE/ohs/Apache/modplsql/conf

For Oracle EPG, the parameter can be set as follows:

  • dbms_epg.set_global_attribute(‘max-parameters’, ‘100’);

And this is where it gets tricky!

Currently we are leveraging Oracle 11g Standalone HTTP Server (11.1.1.7), so without this bulky and difficult WebLogic overhead and we thought we’d aught to test this installation with these big appointments.
As expected, we received an error-message, so the behavior is consistent for this part. The scary bit though was that the error-message is no longer an error-message but a hint to go in the woods…

The request could not be understood by the server due to malformed syntax.

Which could virtually mean anything, and if we hadn’t run into the previous error, we’d have a helluvatime trying to resolve this.
Luckily the theater was set and we knew what we were searching for and testing, so we quickly resolved this also by adding PlsqlMaxParameters.

I hope this post will help you troubleshooting this catch-all message too!

Update on July 25th 2014
As we’re running into problems a new, I decided to call upon Oracle Support too… Raise and SR and see what comes floating to the top.

Well, a confirmation of what we already concluded, there is no conclusion…

I'm not able to determine what the maximum value is that you can set this to as it does not appear to have been documented in the bug, and the parameter has not been documented at all in the product documentation. I don't have access to the source code, so I'm not able to see if a hard limit has been set or not.

Raising the number of MaxPlsqlParameters is not considered to be a definite solution. Which kinda sounds logical since where you need a page-call with more than four thousand !! parameters (or even 2,000 when you think of it), you would think something could be optimized.

And the suggestions of MOS, which I will support:

  • I would suggest you raise the value to what you need it to be for your environment. If there is a hard limit in the code, and you configure this above that limit, then you will get an error.
  • I would then suggest you look into your application and work to reduce the number of parameters as per the recommendation of the developer of the parameter.