I am a DBA, not a Systems Administrator. Therefore, the metrics/alerts that matter to me are primarily for the instance, secondarily for the VM, and pretty much never for the host (especially when the instance and VM metrics are all Normal). The "All Metrics in a Category" alerts are a great time saver because of how inefficient the interface is for setting up blackouts. (See , for example.)
However, "All" does mean all, and so it includes the host metrics that I don't care about. I'd like to see a set of "All non-host Metrics in a Category" alerts, so I can avoid alerts like this one without having to set up 5 individual alerts (one for each non-host metric) and potentially 7 blackout periods (one for each day of the week) for each one:
Alert: All Metrics in a Category - CPU
Database Instance: <redacted>
Execution Time: Sunday - October 02, 2016 02:21:02
View Alert Status: http://Redacted:8123/iwc/alertMain.iwc
View Resource Charts: http://Redacted:8123/iwc/healthMetricsDetail.iwc?db_id=6&tab=CPU&pm=P
Resource: Host CPU Usage - HIGH
Alarm Level: Critical
75% of data values were in critical state
Resource: Instance CPU Utilization - NORMAL
Alarm Level: Normal
No data values were in warning or critical state
Resource: O/S CPU Utilization (%) - NORMAL
Alarm Level: Normal
No data values were in warning or critical state
Resource: Signal Waits Percent - NORMAL
Alarm Level: Normal
No data values were in warning or critical state
Resource: VM CPU Ready Time - NORMAL
Alarm Level: Normal
No data values were in warning or critical state
Resource: VM CPU Usage - NORMAL
Alarm Level: Normal
No data values were in warning or critical state
This message was system-generated. Do not reply to this message
