EM 12c Email Notification Setup

Question

Hello,
I've got EM 12c up and running.  I've got some notifications already setup.  They are for the database, critical/fatal/warning, that I set up through the incident manager.  Not sure if this was the right way to go about that but anyways it seems to be working.
I'm trying to get it setup so I get notified if say the filesystem is running out of space, say if /u02 is 85% used to send an email.  If I go into Enterprise, Monitoring, Monitoring Templates. I've created a new template for the hosts, with all the metrics that I want monitored.  Now how do I get that to email me those alerts?  Is there a document I'm not seeing that will help me set this up?
Thanks!
ETA:  I can go to the host homepage, and see that there metrics that I'm monitoring are alerting.  I just can't seem to figure out how to email based on those alerts.

Answer 1

Hello,
To answer your question, here are the high level steps:
1) Setup your SMTP gateway.   However, if you are already getting email, then perhaps you've done this step already?
For reference, here is the link to the doc for details:
http://docs.oracle.com/cd/E24628_01/doc.121/e24473/notification.htm#sthref154
2) Set up your email addresses. 
Under your username drop-down menu,  you should specify the email addresses to which  notifications should be sent.
Here is the link to the doc for details:
http://docs.oracle.com/cd/E24628_01/doc.121/e24473/notification.htm#autoId2
3) Set up rules (incident rule sets) that specify the metric alerts for which you want emails to be sent
The rule sets were designed to provide a lot of flexibility in covering the different types of events that EM12c supports as well as the many types of notification requirements that many data center needs.
Think of a 'metric alert'  as a type of event.  
For your specific requirements (sending email for host metric alerts)  you can do this:
3.1) Create a rule set and specify the targets for the rule.   In your case you can specify all targets of type 'host'  (if you want the email to be sent for all host targets' alerts) or choose the specific hosts you're interested in.   
3.2) In the Rules tab,   create an event rule (this is the first option in the choice of rules to create).   When asked for a 'Type', choose 'metric alert' as the event type.   Then choose 'Specific events of type Metric Alert',   then choose the filesystem metric and other metrics for which you want notifications to be sent.
3.3) In the Actions part of the rule, specify the EM user(s) in the 'Email to' field of the Notifications section.
3.4) Save the rule
This should get you up and running with email notifications. 
However, I would also recommend looking into these other areas to optimize your use of EM12c monitoring:
1)   If this notification requirement applies not to all hosts but a subset of host targets,  consider creating a group of those targets and specifying the group as the target of the rule set  (step #3.1 above).  Once you have a group, you can perform operations on the group instead of individual targets, makes it easier to manage later on.  As you add more targets to EM, if they need to be monitored in the same way as the other targets in the group,  then simply add the new target to the group,  the rule will automatically apply to the newly added target in the group without requiring changes to the rule.
2) Use incident management features
In EM12c, we introduce incident management as a way to focus and better manage the more important issues that impact your datacenter.   Hence when an important event comes in, consider creating an incident for it.  Once you  have an incident, you can leverage features such as specifying an incident owner,  setting its priority,  resolution status, etc.    Using Incident Manager will provide you in-context diagnostic links and access to MOS articles to  help further diagnose/resolve the event.
Here is the link to the doc for more details on incident management:
Using Incident Management
3) Leverage other rule options
The rule has an option to apply to all metric alerts of specified severities.   So  if you want to receive email for all critical and warning alerts for the host targets,  then instead of choosing the specific metrics, you can change the rule (step #3.2 above) and choose the option 'severity in critical, warning'  instead of choosing individual metrics.     If you do so,  then say later on you want to add thresholds to new metrics (which will generate metric alerts).  You won't have to change the rule to add in the new metrics since your rule already covers all metric alerts with critical and warning severities.  However, since it will cover all metric alerts for your chosen targets, you need to make sure that you have the appropriate thresholds for the metrics to avoid unwanted email notifications from unwanted alerts.
The Incident Manager chapter that I referenced above has an Advanced Topics section that covers other scenarios.   If you have other monitoring requirements, you can take a look at that section to get ideas on how these can be implemented in Enterprise Manager.     Or you can reply back to this post if you have additional questions.
Finally, you can take a look at this monitoring best practices paper that describes how you can leverage EM12c to set up monitoring in a scalable way:
http://www.oracle.com/technetwork/oem/sys-mgmt/wp-em12c-monitoring-strategies-1564964.pdf
Regards,
Ana