News & Resources

Informix Stored Procedure Language (SPL) has a smaller command set than other database products, but is more than adequate when making full use of built-in SQL functions. An SPL routine is called a PROCEDURE if no data is returned, or FUNCTION otherwise, as explained here. User Defined Routines (UDRs) can also be written in C or Java, whereas this article is only concerned with SPL UDRs. There may be multiple levels of SPL routines called by another, and it can be difficult to know which has failed if an error results. This article provides a method of recording and retrieving the details of any failure, with only a tiny code addition at the top of each routine.

[Read More…]

The Informix statements LOAD and UNLOAD allow data to be transferred between a flat file and a database, but are in fact only implemented in certain clients such as DB-Access, 4GL variants, and AGS Server Studio. You may therefore need functions in Informix Dynamic Server (IDS) that do this, such as when coding in Java with JDBC. Furthermore, external tables are much lighter and quicker for this purpose, so you might in any case prefer the functions described in this article, which use them in Informix Stored Procedure Language (SPL).

[Read More…]

Informix supports views like other database products, which behave much like tables but are in fact SELECT statements stored for convenience and reuse. If a table or view is dropped and recreated during a schema change, any views in the same database that depended on it will have been lost. This article covers this problem, and includes SQL to list dependencies and definitions of views.

[Read More…]

Informix database server products, including both Standard Engine and Dynamic Server, have for decades been delivered with a character interface tool for database exploration and SQL execution. This is referred to as “DB-Access” in the documentation, and is invoked with “dbaccess” from the operating system command line or within scripts, as described in the latest DB-Access User’s Guide. This article covers some lesser known and very useful environment variables that improve its usability, and a way to enable command history and editing when used directly as an SQL CLI.

[Read More…]

If an Informix instance warns that one of your temp dbspaces is full, it is useful to have SQL quickly to list what it contains, or to be run automatically if dbspace usage goes over a maximum acceptable percentage. Such a “sysmaster” query is provided in this article.

[Read More…]

This article provides a script to analyse what database objects were locked when IBM Informix Dynamic Server (IDS) “asserts” (saves an evidence file) including what sessions were waiting on another, which is particularly useful to diagnose deadlocks.

[Read More…]

This article provides a stored procedure to run InformixHQ (HQ) Java processes automatically whenever IBM Informix Dynamic Server (IDS) has been restarted, as originally posted on the community forum here. It follows on from the previous article InformixHQ Setup and uses the shell script “informix-control.sh” it contains which starts or stops HQ.

[Read More…]

When testing a database migration to a new platform that is not compatible, for example with a different OS or locale, you cannot restore from a level 0 archive on the new server. Data must be copied either via flat files or directly using SQL tools. If this is from a live database that is being updated and cannot be exclusively locked (which rules out dbexport), you cannot guarantee consistency of the data being reloaded, so NOVALIDATE must be set to ensure foreign keys can be reapplied. This article provides stored procedure language (SPL) code to identify and delete any resulting invalid data.

[Read More…]

This new feature, documented here, was introduced with IBM Informix Dynamic Server (IDS) version 14.10. It is an enhancement to Enterprise Replication (ER) allowing for data changes to be handled by a stored procedure rather than being applied directly to a replica table. It is very likely to be used in a Loopback Replication arrangement, so that the stored procedure fired is contained in the same database as the source table. Such a replicate is also known as an “asynchronous post-commit trigger”. This is because, unlike a conventional trigger, it runs in background with respect to the calling application and occurs after the transaction has been committed. ER infrastructure should ensure reliable buffering and delivery should a backlog build up or if the server is restarted. This article gives an example of how this might be useful to perform whatever type of action you wish when data changes have been committed.

[Read More…]

This relatively new feature, documented here, was introduced with IBM Informix Dynamic Server (IDS) version 12.10.xC11 in March 2018, enabling replication of tables within the same database or instance. It is an enhancement to Enterprise Replication (ER) which was previously only supported between different instances. This article gives an example of how it can reduce downtime when slow schema changes are required on large tables in continuous use.

[Read More…]

This new command, documented here, was introduced with IBM Informix Dynamic Server (IDS) version 14.10.xC2 in September 2019. In theory, it reduces the effort involved in creating a new active-active replica instance to a single command. It uses Enterprise Replication (ER) which supports participating servers being on different Informix versions or operating systems, providing a unique route for platform migration with almost zero downtime. The complex work to achieve this with ER was previously a major disincentive. This article looks at the current reality of “cdr migrate server” (CMS), showing that it can achieve its goal if some pitfalls are avoided. CMS should not be confused with the earlier ifxclone command to automate the creation of another node in a high-availability cluster, which must be on the same platform and version as the source server.

[Read More…]