This is Part III in a multi-part series, demonstrating how to take EM12c from out of the box to enterprise level.
Monitoring templates, when created via an example database and then used as the default for all databases, allows the DBA control to create incidents and notify on what metrics are important to the business.
Creating a Template
As templates are used at the global level, we create a template by clicking on Enterprise à Monitoring à Monitoring Template.
There are no templates by default, but as they are based off existing databases, it is easy to create one. Click on “Create” to begin the Template wizard.
For our first Template, we are going to create one based off an existing “Host” target:
We leave the default “target” for our copy choice and then click on the magnifying glass to choose an existing host from our list of targets.
Name the template a clear name and add a description. Choose to make this the default template, (although you can also edit the template and add this later, but if you don’t, it will not be applied to future discovered targets of this type.)
Click on the “Metric Thresholds” tab.
I recommend the following settings be applied to the host monitoring template:
|Disk Device Busy (%)||90|
|File system Space Available (%)||20||5|
|CPU Utilization (%)||90|
|Memory Utilization (%)||90|
There are rarely any critical metric thresholds. The DBAs main indicator for issues will be at the database level, not the host level, so critical alerting on these settings is removed and warning thresholds are increased to eliminate unnecessary incident creation in the EM12c console.
Save the template.
Single and Cluster Database Template
Click on “Create” to start the next template creation, first for a single database instance and then create a second template with the same metrics that you see in the table below, along with the additions shown in the cluster database metric table and use a cluster database for the source.
Again, choose the default target type and then click on the magnifying glass and choose a database instance, (single instance) and select it as the source for metrics to be sourced for the template.
Name the template a descriptive name and add a description and click in the check box to make it the default for all discovered targets of this type. Click on “Metric Thresholds” to now edit the metric values for warnings and critical.
Update or add the following values for metric thresholds:
|Archiver Hung Alert Log Error||ORA-|
|Data Block Corruption Alert Log Error||ORA-|
|Generic Alert Log Error||ORA-0*(600…|
|Media Failure Alert Log Error||ORA-|
|Session Terminated Alert Log Error||ORA-|
|Archiver Hung Alert Log Error Status||0|
|Data Block Corruption Alert Log Error Status||0|
|Media Failure Alert Log Error Status||0|
|Session Terminated Alert Log Error Status||0|
|Archive Area Used (%)||80||97|
|Broken Job Count||0|
|Failed Job Count||0|
|Database Vault *(3 total)||0|
|Data Failure Detected||.*|
|Deferred Transaction Count||100|
|Deferred Transaction Error Count||0|
|Dump Area Used (%)||80||97|
|Exadata- Error Code||0|
|Incident- Access Violation||.*|
|File Access Error||.*|
|Generic Internal Error||.*|
|Inconsistent DB State||.*|
|Internal SQL Error||.*|
|Oracle Data Block Corruption||.*|
|Out of Memory||.*|
|Redo Log Corruption||.*|
|Need to Instrument with OCM||1|
|Op Error- Archiver Hung||.*|
|Data Block Corruption||.*|
|Generic Operational Errors||.*|
|Streams Process Errors||0|
|Streams Process Status||Disabled||Aborted|
|Tablespace Space Used||85||95|
|Tablespace Space Used(Dict)||85||95|
|Segments Approaching Max||0|
|Segments Not Able to Extend||0|
|Temporary File Status-Status||OFFLINE|
|User Audit- Audited User||SYS|
|Database Service Status||Up with Warning||Down|
Save the template after completing the updates/changes to the metric thresholds.
You should now have three monitoring templates- one for host and then one for single database instance and another for cluster. You can add others as required by the business, but these are the three required to alert of CRITICAL only issues and create incidents for managing during business hours of warning thresholds on metrics. As these are “default” templates, any new targets added for these target types will automatically “adopt” these metric settings and thresholds.