Resolving space use exceptions for index databases

Resolve the space use exceptions that you received for your index databases.

Before you begin

Ensure that you have completed the analysis steps for index database exceptions and that you have identified the possible causes of the database exceptions. If you have not yet analyzed the space use exceptions, see Analyzing space use exceptions for all full-function database types.

Procedure

If you receive space use exceptions for index databases, the method you apply to resolve the exceptions depends on the results of the analysis steps as follows:
Tip: Depending on the status of the data set, applying several of the following methods can be an effective approach to resolving the exceptions. Before making any changes to your environment, evaluate all the methods in the following list and determine the best approach to resolve the condition.
  • If the out-of-space condition is caused by a primary-space-used-up condition (that is, the data set has only primary space, and the size of used space is reaching the size of the primary allocated space), and if the data sets allocation size is smaller than the maximum size (4 GB or 8 GB), increase the size of the allocated space. For instructions for increasing the size of allocated space, see the topic "Changing the amount of space allocated" in IMS Database Administration.
  • If the out-of-space condition is caused by a data-set-full condition (that is, the data set approaching its maximum size of 4 GB based on the value of the DBX_RBA_HIGH_USED data element), new index segments cannot be created in the index database data set.

    Consider the following methods to resolve a data-set-full condition:

    • If your HIDAM primary index database is approaching its maximum size, you must reduce the number of root segments in the HIDAM database. To reduce the number of root segments, split the HIDAM database to several smaller databases, or migrate to a PHIDAM database.
    • If your secondary index primary data set, overflow data set, or both are approaching their maximum size, you must reduce the number of index pointer segments. To reduce the number of index pointer segments, split the database to several smaller databases, or migrate to a HALDB.

      If you do not need certain pointer segments to be built, consider suppressing index entries (sparse indexing). For more information about sparse indexing, see the topic "How sparse indexing suppresses index entries" in IMS Database Administration.

    • If your partitioned secondary index database data set is approaching its maximum size, split the partition of partitioned secondary index into several smaller partitions.
  • If the out-of-space condition is caused by an extent-resources-used-up condition (that is, when the data set cannot be expanded), typically, changing the data set allocation definitions and reorganizing the database will resolve the condition.

    Consider the following methods to resolve an extent-resources-used-up condition:

    • If the data set has reached the maximum number of data set extent counts, and if the DASD volumes have enough space, increase the primary and the secondary extent sizes for data set allocation.
    • If the DASD volume does not have enough space, or if the data set has reached the maximum number of extents that are defined for that volume, increase the number of DASD volumes.
    Generally, as the number of extents or the number of DASD volumes increases, I/O operations also increase. Therefore, consider using a DASD volume that has sufficient size, and specify an extent size that is large enough for data set allocation.