IBM® Informix® 12.10

Questions about administering and using Informix

These topics provide short answers to some frequently asked questions about administering and using Informix®.

Starting the database server, user information, and privileges:

Configuration:

Backup and restore:

Assertion failures:

Other administrative tasks and questions:

Where can I find information about what to do if the server does not start?

If an oninit command encounters an error, the database server returns an error message and a return code. The message text describes why the database server did not start. For a list of oninit return codes, the text of the messages that the server returns, and the action you can take to solve each problem, see the Return codes for the oninit utility topic in the IBM Informix Administrator's Reference.

For example, the message for oninit return code 175 tells you that the server failed to initialize the root dbspace. If you go to the Return codes for the oninit utility topic and scroll to return code 175, you learn that you should check the root dbspace related parameters in the server configuration (onconfig) file to make sure that the path for the root dbspace is valid.

What can I do if I get a message that the password is incorrect or user informix is not known on the database server?

If you try to connect to Informix using DB-Access on a Windows operating system, and you receive error 951, you can troubleshoot the problem by following the guidelines in this IBM® Technote, which is available from the IBM Support portal: http://www.ibm.com/support/docview.wss?uid=swg21200068

How can I tell which users have DBA privileges on a database?

From DB-Access or your application, run this query:
select username,usertype from sysusers;

The output shows user names (for example, public and informix) followed by one of the following codes:

For information about the database privileges, see Database-level privileges.

Where can I find information about the oncfg file?

This oncfg file contains information about dbspaces, chunks, and logical-logs that are used during a whole system restore. For information about this file, see the Facts about the ONCFG file IBM Technote that is available from the IBM Support portal: http://www.ibm.com/support/docview.wss?uid=swg21106660

If you are looking for information about the onconfig.std file, which contains configuration parameters, see onconfig Portal: configuration parameters by functional category in the IBM Informix Administrator's Reference.

How can I create a customized onconfig file?

You can customize the default onconfig.std file by using the genoncfg utility, which is in the $INFORMIXDIR/bin directory. For more information, see The genoncfg Utility in the IBM Informix Administrator's Reference.

How much temporary space is needed for oncheck sorting?

The oncheck utility requires sort space when examining an index. The amount of sort space required is the same as the amount that is needed to build the index. For information about calculating the amount of temporary space needed, see Estimating temporary space for index builds in the IBM Informix Performance Guide. If you receive the error "no free disk space for sort," you must estimate the amount of temporary space needed and make that space available.

Where can I find information about ON-Bar return codes?

ON-Bar return codes are accompanied by messages in the ON-Bar activity log. For a complete list of ON-Bar return codes, see the ON-Bar return codes topic in the IBM Informix Backup and Restore Guide.

Where can I find information about solving errors that might occur during backup and restore?

The Knowledge Collection: Informix Backup and Restore Utilities IBM Technote (http://www.ibm.com/support/docview.wss?uid=swg21404291) contains links to technotes on backup and restore utilities, such as the ON-Bar and ontape utilities.

These technotes contain answers to many common questions, information about solving some errors that might occur, and sample commands. These Technotes supplement the information in the IBM Informix Backup and Restore Guide.

Can ON-Bar start continuous log backup even if all logical logs are 100 percent full?

The ON-Bar utility can start a continuous backup of the logical logs even when the logical logs are full. However, a better practice is to make a regular backup of the logical logs and set the alarm program ($INFORMIXDIR/etc/alarmprogram.sh) to back up each log as soon as the log becomes full. For more information, see Starting a continuous logical-log file backup.

How do I correct an assertion failure?

You can correct some types of assertion failures yourself. For others, you must contact IBM Software Support. For more information, see Correcting assertion failures.

Can I prevent assertion failures?

Yes, you can prevent many assertion failures by performing routine administrative tasks and thoroughly testing your applications. However, you should be prepared for assertion failures and other problems. For more information, see Prevent and prepare for assertion failures.

How can I plan responses to severity 4 and 5 event alarms?

Event alarms have a severity level, which specifies the seriousness of the event on a scale from 1 to 5. A level 5 alarm, which is the most serious event alarm, indicates that the database server failed. For more information, including details on how to respond to server failures, see the Event Alarms IDs and Severity 5 event alarms topics in the IBM Informix Administrator's Reference.

What can I compress?

You can compress and uncompress the following items:

How do I estimate compression ratios?

You can estimate the percentage of space that you can save if you compress data in tables or table fragments.

To estimate the compression benefit, run the admin() or task() function with the estimate_compression argument. For more information, including examples, see the Estimating compression ratios topic in the IBM Informix Administrator's Guide.

If you use the IBM OpenAdmin Tool (OAT) for Informix, see the OAT help topic on estimating compression.

What can I do to prevent automatic update statistics (AUS) from running?

To prevent AUS from running, disable both the Auto Update Statistics Evaluation task and the Auto Update Statistics Refresh task, as follows:
  1. Update the value of the tk_enable column of the ph_task table to F where the value of the tk_name column is Auto Update Statistics Evaluation.
  2. Update the value of the tk_enable column of the ph_task table to F where the value of the tk_name column is Auto Update Statistics Refresh.

For more information and examples of using this method, see Disabling AUS in the IBM Informix Performance Guide.

You can also use the IBM OpenAdmin Tool (OAT) for Informix to disable the AUS process. See the OAT help topic on configuring and enabling automatic update statistics.

Must I run update statistics manually to set the level of statistics before using automatic update statistics (AUS)?

No. The AUS maintenance system automatically identifies and runs the necessary update statistics. However, if you manually update statistics, AUS does not run the statistics at a lower level, resolution, confidence level, or sampling size. For more information, see Automatic statistics updating in the IBM Informix Performance Guide.

What is the purpose of star-join directives and where can I find information about them?

You can use star-join directives to specify how the optimizer joins tables that have a star schema.

Many data warehouse databases use a star schema, which consists of a fact table and a number of dimensional tables. In a typical star join, the fact table joins with all dimensional tables on a foreign key. For more information, see Star-join directives in the IBM Informix Guide to SQL: Syntax.

How do I get the onpladm utility to start jobs in the background on the Windows operating system?

If you want onpladm utility jobs to start in the background, set the INTERACTIVE_DESKTOP_OFF environment variable to 1. Otherwise, the utility starts each onpload job in a new command window because, by default, the INTERACTIVE_DESKTOP_OFF environment variable is set to 0. For more information, see the topics about using the onpladm utility on Windows" in the IBM Informix High-Performance Loader User's Guide.

What are the benefits of and what are some examples of defragmenting partitions?

Because a frequently updated table can become fragmented over time, leading to performance degradation, the ability to merge non-contiguous extents can improve performance. Defragmenting a table or partition brings data rows closer together and avoids partition header page overflow problems.

For more information about and examples of defragmenting partitions, see defragment argument: Dynamically defragment partition extents and the developerWorks® article about the defragmenter (http://www.ibm.com/developerworks/data/library/techarticle/dm-1011informixdefragmenter/index.html).

How do you use fragment-level statistics?

Fragment-level statistics affect the way UPDATE STATISTICS MEDIUM and HIGH operations gather data and generate column distributions on fragmented tables. For information about the circumstances under which you might want to use fragment-level statistics and details about setting up your system to use fragment-level statistics, see the developerWorks article about taking advantage of fragment-level statistics (http://www.ibm.com/developerworks/data/library/techarticle/dm-1104fragmentstats/index.html).

Also see relevant reference topics such as Fragment-level statistics and Statistics options of the CREATE TABLE statement.


Examples exchange | Troubleshooting