Informix 11.70.xC6

This article is written in English and Portuguese
Este artigo está escrito em Inglês e Português

English Version:

As usual, I’m just reviewing the latest IBM Informix fixpack. The software was just released (yesterday, October 23). It’s important to note that this fixpack is clearly on the “mature” phase of 11.70 and also that IBM is already working with selected customers on the next major release. As such, we should not expect too many news… Nevertheless, and I’ve been mentioning this to several customers, IBM uses the Agile methodology in Informix development. And one of the goals of this methodology is to provide new functionalities to customers as soon as possible. So, what am I saying with this mixed message? That even for a “quiet” fixpack we can find surprising new features. As usual, let’s take a look at the release notes and comment on the new features section:

  • Administration
    • Enhancements to the OpenAdmin Tool (OAT) for Informix
      Just a version refresh for several components (Apache, PHP and PDO_INFORMIX)
    • Support for the same XID for transactions on different databases
      A new parameter IFX_XA_UNIQUEXID_IN_DATABASE allows that a global (XA) transaction ID can be repeated in different databases in the same instance. Or in other words, the scope of the transaction ID can be changed from the instance to the database level. This is essentially a compatibility feature.
    • Coordinating transactions within a high-availability cluster
      This is one of the surprises… A new parameter CLUSTER_TXN_SCOPE allows a DBA to define if the applications receive the COMMIT confirmation immediately or after the transaction information was propagated through the cluster (or just the current secondary server). This can be a bit confusing and some experimentation is required, but it basically avoids that an application that takes advantage of several cluster nodes hits some apparent data consistency issue caused by the asynchronous nature of the logical log propagation. It can also be useful if an application that changes data, triggers some other process that will read that same data from another node in the cluster. I would say this can be an invaluable option for future expansion of the clustering capabilities in Informix
    • Easier failover configuration for Connection Managers in a high-availability cluster
      We can now specify the failover policy at the engine level by setting a new parameter HA_FOC_ORDER that will take priority over the policy specified in the connection manager(s) connected to the instance
  • Application development
    • Enhanced support for OUT and INOUT parameters in SPL routines
      Another surprise. For most cases and most people this will seem irrelevant. But I recall a couple of situations where I’ve found the lack of support for OUT/INOUT parameters in SPL a big…

Auteur : (Fernando Nunes)

No comments yet.

Leave a Reply

%d bloggers like this: