DX Unified Infrastructure Management

Hi All,

We are observing that, the UIM database is growing rapidly and the disk space on the server is almost full. We have increased the disk space several times in last 6 months and unable to predict the future data growth.

Below are the DB maintenance settings on data_engine probe.

When I run the below query, I see the Daily Average Age value is 744. Is this the one causing the DB growth?

How can I change the setting for Daily Average Age (DAILYAVGAGE) to a value lower than 744? Please let me know.

Also, below are the tables do not have data in them. Can we suspect any issue with UIM database? Please clarify me.

Regards,

Rajashekar

Also, please let me know how to verify what is causing the DB growth.

We are using ORACLE as UIM database.

in raw config you can change the following value

   daily_avg_age = 744

this is most likely not your problem

Please see the below KB artciles

hope this helps

Hi Gene,

If possible, can please provide me the oracle version of the queries listed in the mentioned KB articles?

I have analyzed raw data tables mentioned above in my post and found that the tables which are being used by QOS_ORACLE_SYSTEM_WAITS, QOS_ORACLE_SYSTEM_STATISTICS AND QOS_ORACLE_SYSTEM_STATISTICS_DELTA metrics are big in size (space all these 3 tables is about 75% when compared to remaining tables size). I hope, this is expected because of the nature of monitoring of these three metrics.

How can we tweak these three metrics to store less data?

Regards,

Rajashekar

the queries above should be ok for oracle may have to adjust for case sensitivity.

If you are having a problem with a specific one please let me know and I will see what I can do.

Generally speaking the only way to reduce these is to reduce what is being monitored. Either the number of metrics or the frequency which they are captured.

you may want to work with your DBA and go through the oracle probe template and see if all of the metrics being collected are really needed or not. Any that are not remove.

You can also reduce your retention period in the data_engine to reduce the time the QOS is stored.

Hope this helps

The other thing you can do is go into the data_engine on the Quantity of service tab.

Find these three QOS messages and then set a different retention time period just for these QOS.