UIM Historic Table Are Empty

Document ID : KB000006842
Last Modified Date : 14/02/2018
Show Technical Document Details
Issue:

The HN_QOS_DATA_xxxx historic data tables are empty or show a very small, uniform size.

Cause:

This can happen if there are problems with the configuration parameters table which is used by data_engine. When data_engine runs maintenance, it is calling spn_de_datamaint. This stored procedure is pulling its configuration parameters from tbn_de_DataMaintConfig. If data_engine is unable to correctly update this table, then it's possible that the maintenance routine is following invalid retention policies.
The following document will help confirm whether data_engine is performing data manitenance:
https://www.ca.com/us/services-support/ca-support/ca-support-online/knowledge-base-articles.TEC000003967.html

Resolution:

Confirm that data_engine is configured to maintain historic data and that enough time has passed that there should be data stored in the HN_QOS_DATA_xxxx tables.

Confirm the RN_QOS_DATA_xxxx tables are maintaining data for the period specified in the data_engine configuration.  So, if data_engine is configured to compress data before delete and the "Delete raw data older than" setting is for 30 days, confirm that there is at least 30 days worth of data in the appropriate RN_QOS_DATA_xxxx table.

Ensure the SQL user being used by UIM has all the appropriate privileges.  You can find more information on privileges in the install guide:  https://docops.ca.com/ca-unified-infrastructure-management/8-5-1/en/installing-ca-uim/pre-installation-planning/install-and-configure-your-database-software

During normal operation, the maximum timeout for data_engine maintenance is about 18.2 hours - so ensure you have waited about 18 hours to ensure there really is a problem.  If a question remains, you can use the following query, (if it is still failing after the 18 hours), to get the latest tbnlogging results:
SELECT * FROM TBNLOGGING WHERE EVENT NOT LIKE '%SNAPSHOT%' ORDER BY ID ASC;

To resolve some of these problems, please complete the following:
1. deactivate data_engine
2. issue the query: TRUNCATE TABLE TBNLOGGING;
3. activate data_engine
4. highlight data_engine in IM and press CTRL+P to bring up the probe utility
5. execute the callback "run_admin_now"
6. Wait about 15 minutes and then check some HN_QOS_DATA tables to see if they are increasing in size.  You should see data beginning to populate there.

 

If these steps do not resolve the problem, please collect the following data and open a Support case:

- level 3 data_engine logs

- export the following to csv file for review:

  •   select * from tbn_de_DataMaintConfig
  •   SELECT * FROM TBNLOGGING WHERE EVENT NOT LIKE '%SNAPSHOT%' ORDER BY ID ASC;