This content has been marked as final. Show 10 replies
Creating Incident rules is a very simple process as long as you know what event you wanna create your rule one. In your case specifically, its even more simple since you already know your target type and metric name.1 person found this helpful
1) Create a rule set.
2) On the same page hit "Targets" tab and select "All target types" and select "SOA Composite" from the target type list.
3) Now hit "Rules" tab and then hot "Create" and then select first radio button - "Incoming events" since you wanna wrap SOA composite fault metric alert event into a rule.
4) Now select "Type" and select "Metric Alert"
5) Now since you are interested in a "specific event" - SOA composite fault metric alert event, select that radio button and hit "Add"
6) From the "Target Type" dropdown, select "SOA Composite" target type and type "%Fault%" in the Metric name textbox and hit Search.
7) Select you Metric name and select you severity as you wish.
8) Add Actions - create notification, emails etc. (Refer docs to understand how best you can use these options as needed)
9) Save your rule and save your Rule set.
You have a rule (Make sure its enabled) that will execute your specified actions when any SOA Composite will breach your the selected metric (Fault metric in your case)
Thanks a lot!!!
Could you plesae confirm that if we have to set the metric threshold for the composite before firing this rule. Please guide!!
The metric alert event must get fired for this rule to get triggered. This is true for any event type and not just metric alert event type.
For metric alert event to get fired, you will need to set threshold on your selected metric.
Can you please tell me threshold for what metrci needs to be set. Actually I want to get a notification with the fault mesage details i-e- with the fault type, fault reason and fault message.
Can you atleast tell me what metrics are used for what purpose. Is there any list regarding that. Also which view in management repository contains information about all the faulted instances.
Please help!!! its really urgent!!!
Metrics definitions are defines in the online doc: http://docs.oracle.com/cd/E24628_01/em.121/e25158/soacomp.htm#sthref3843
You can get more understanding of SOA composites by using the following query:
select * from gc_metric_values_latest where metric_group_name like '%component_rollup_by_revision_tc%' and metric_column_name like '%totalFaults.count%';
Now I am able to create an incident whena fault occurs but in the summary message I am getting the following message
Total number of faults since server start is %value%
I think instead of this *%value%* I shoud get the exact count. Can you please tell me how can I resolve this problem. Yoi have guided me so well and now I am almost close to the desired result. Please get me something to solve this issue too
Also can you tell me how can I fetch the error messages. Through ehich metric I can fetch the error messages getting generated for the fault. What if I want to notify my administrator about the fault messages.
I suspect this is a bug in the SOA composite target type metadata. The message should probably have been:
Total number of faults since server start is %totalFaults.count%.
You may want to follow up with support to file a bug or check if one has been filed.
Can anyone please tell me in OEM 12c Cloud control, in which table/view of management repository it stores the SOA composite instance information i.e. instance id, instance state etc. or does it not store these information about a SOA composite at all in any of the table or view in management repository. But then from where it fetches these information for a particular SOA composite and shows on the SOA composite home page.
Thanks in Advance!!