In this post, I will explain how to install Oracle Fusion Middleware Forms and Reports 11g R2 for a high availability architecture on a Windows platform. For the Linux platform, it’s approximately the same, even though the node manager is a daemon.

Even if we (dbi services) usually advise to setup Oracle Enterprise Linux for Oracle deployments on Linux, it might appear that some customers still prefer other certified distributions such as SuSE Linux Enterprise Server (SLES). This post describes the configuration of the tmpfs filesystem size on SuSE Linux Enterprise Server 11.0 (SLES 11, without any service pack) in order to use the Oracle 11.2 Automatic Memory Management feature (AMM) through the Oracle instance parameter memory_target.

In this post, I will describe how to migrate from Oracle APEX 3.0 to Oracle APEX 4.1. I will also focus on some encountered problems as well as the new features of Oracle Application Express (APEX) 4.1.

A few days ago, the dbi services consultants had the pleasure to welcome Oracle guru Martin Bach for a three day Oracle Grid infrastructure training. As many of you probably know, Martin Bach is an OakTable member and the author of 'Oracle Database 11g RAC on Linux'. He also is Oracle Certified Master and Oracle ACE.

Le marché du Database Activity Monitoring (DAM) devenant de plus en plus important avec des produits tels que Imperva, Guardium ou Sentrigo, je me suis intéressé au produit Oracle Database Firewall, qu'Oracle présente comme un substitut aux différents acteurs du marché des DAMs.

Oracle has introduced three new caching features in its Oracle 11g release:

• Query result cache
• Client OCI result cache
• PL/SQL function result cache

In this posting, I will present the query result cache feature.

Ayant rencontré récemment des problèmes de blocs corrompus sur des bases Oracle en version Oracle 10g et Oracle 11g, des différences notables sont apparues notamment au niveau de la vue v$database_block_corruption. Je vous livre dans ce post le résultat de mes tests.

A few days ago I downloaded the Cloud Control 12c binaries for Linux x86_64 from Oracle Technet. My friend Hervé Schweitzer, configured an OEL 5.7 VM for test purposes and I was quite eager to install this new software. I checked the OS settings, the required parameters, etc. - everything was fine. My OMS repository database (11.2.0.3) was ready. Unfortunately, as soon as I started the installer, it directly crashed.

My colleague Hervé Schweitzer reported a very interesting issue to me. It is about the usage of Grid Control 11g while monitoring the available space on an Oracle database – in our case a 10.2.0.3 target database. This post continues my last post about the free space management in a tablespace in relation to the recycle bin management.

Last week at a customer site, we experienced an excessive redo log generation rapidly filling up the FRA (Flash/Fast Recovery Area) with archivelog files. This eventually caused the database to crash.

Last week at a customer, I used DBUA to upgrade an Oracle database from 10g to 11g and the migration ran successfully. However, the day after I got a monitoring message that the diag directory is growing very rapidly.

Aprés le post sur « Utilisons les securefile », revenons sur le sujet LOB avec une approche performance, stockage. Nous rencontrons parfois des modèles de données étranges, nous aurions envie en tant que DBA de tout reprendre à zéro pour refaire un modele relationnel en 3eme forme normale (3NF), plus efficace et mieux conçu dans notre « RDBMS » favori, mais la réalité étant ce qu'elle est nous devons faire avec, et au mieux. De plus Oracle n'est pas qu'une base de données relationnelle :-) on l'oublie trop souvent.

This post describes some basic Oracle database management issues database administrators might be confronted with. It is about space management and differences observed while monitoring the available space from several point of views (tablespace/datafile) in Grid Control 11g. It also relates and explains an issue (ORA-01652) observed on pre 11.2 Oracle releases (targets) concerning the tablespace/datafile space management and the recycle bin.

At dbi services, we have a few people passionate about diving. Let me take this opportunity to let you dive into the two databases lakes Oracle and SQL Server to see their similarities and differences. For this first level of diving (named ODD), we will not be going too deep for reasons of narcosis or databases toxicity. As always, we will start with a short briefing to present the environment and the different themes. Have a nice dive !

During the last weeks, I experienced an uncommon bug during upgrade of security sensible databases: The upgrade process was sticking for hours on the script c1101000.sql which tried to fill the column DBID with the database DBID of the tables AUD$ and/or FGA_LOG$ (auditing and fine grained auditing tables).

Lors de mon activité de consultant, et suite à un bug qui n'affiche pas correctement le nombre de CPU (OMS 10.2.0.5, agent 10.2.0.5 et linux x86_64), je me suis intéressé à la manière dont Oracle Enterprise Manager Grid Control (OEM GC) récupérait les informations de mémoire et de CPU sur les différents hôtes surveillés par ce dernier.

Last week, I was not able to complete a backup because of the ORA-19566 error: "exceeded limit of 0 corrupt blocks". Here is what you can do to fix it.

It's already quite a while that I posted my last article on the Oracle basics, so I decided to take some times in the train to continue this small serie. This time, I'm going to speak about one of the root of Oracle administration, starting and shutting down a database server.

Sometimes, you need to modify the statistics of a table. There are various reasons: either for the performance analysis or because the execution plan changed. Oracle offers two methods: statistics restore or statistics export. Both are in the dbms_stats package.

While discussing with some tuning gurus, I was made aware about this feature. It is not so far from the adaptive cursor sharing perfectly described in Chris Antognini's blog: http://antognini.ch/papers/BindVariablePeeking_20090718.pdf

This feature compares the real execution (number of returned/actual rows - "A-rows") with some expectations (number of expected rows - "E-rows") and chooses another plan for the next executions if the difference between the expectation and the actual result is too big. "Cardinality Feedback" is not well documented in the Oracle documentation.