Cloud Database Monitoring using SelectStar – Part VIII

We are going through the setup and use of SelectStar, a new heterogeneous monitoring solution for cloud databases from Blue Medora. In the previous blog post here, we added an Oracle database from our on-premise server's collector. The Oracle database then appeared in the SelectStar database overview page, and we then drilled down on the recently added “SaiProd” Oracle database.

We had a look at the Alerts and Recommendations displayed for this database, and were warned there were no recent backups for the database. It was also recommended to multiplex the database online redo log files and archive logs, and set up database flashback. In the Advanced tab, we saw various graphs and pertinent information for this particular database.

We moved back to the Overview page, and clicked on the Queries graph to drill down to further detail. This displayed the top Queries running in the database, meaning those with the highest performance impact.

At this point, move back to the Overview page, and click on the Sessions tab.

This displays the active, inactive, killed or cached sessions as shown below. You can select the date range as before. The count of each type of session is shown in a table in the lower half of the screen.

Move back to the queries tab. Click on one of the queries as displayed.

This drills down to the individual statement. You can click on the Show All button to see all the statements again.

Move back to the Overview page. To look at the Setup,  go ahead and click on the Setup gear button.

This brings up the configuration of the Oracle connection as seen in the following image.

You can change the configuration details here, test the connection and save it on this page. Next, click on the Alerts link on this page.

On this page, you can modify the existing Alert Rules, or add new rules. You can turn an Alert rule on or off.

As you can see, some existing alert rules are based on sessions usage or processes usages going beyond a critical threshold percentage, the days since the last RMAN backup (we saw this alert in the initial investigation of this database), and the datafile read time.

The session usage alert has been repeated twice, first for the usage exceeding 100%, and another alert for the usage exceeding 70%. This should be a warning alert and a critical alert, but in this case the metric alert has been set up twice for two alerts. Let us click on the setup gear button for one of the alerts.

On this page, we can modify the Alert rule for the Sessions usage. The critical threshold can be set up with a new operator, value and unit, and the alert text can also be changed.

The volatility is set to "Stable" which means it will only trigger the event after three consecutive collections. The other options for the volatility are "Volatile" (trigger or resolve after ten consecutive collections), or "Immediate" (trigger or resolve immediately). You can also add a warning threshold on this page.