Service Level Management
Service Level Management
This OBE describes how to manage services and service levels
using Oracle Enterprise Manager 10g Grid Control Release 2. The objective
of this OBE is to help you understand the concept of services, how to monitor
services and report on service availability, performance, usage and service
levels.
Approximately 1 hour
This OBE covers the following topics:
Place
the cursor over this icon to load and view all the screenshots for this tutorial.
(Caution: Because this action loads all screenshots simultaneously, response
time may be slow depending on your Internet connection.)
Note: Alternatively, you can place the cursor
over each individual icon in the following steps to load and view only the screenshot
associated with that step.
The screenshots will not reflect the specific environment
you are using. They are provided to give you an idea of where to locate specific
functionality in the Oracle Enterprise Manager 10g Grid Control console.
In today's business environments, end-users have high expectations
for the applications they depend on. They demand applications that are highly
available, and inherently, the quality of services provided has a direct impact
on the success or failure of the business itself. Thus, it is critical to effectively
manage service levels for a business to achieve optimal success.
The Service Level Management feature offers the most effective
services monitoring solution by providing tools that model many types business
services, measures key service indicators from remote sites, and generates service
level reports for both executives and administrators.
Back to Topic List
You should have Petstore Web Application deployed and managed
using Oracle Enterprise Manager 10g Grid Control Release 2. You should
also have a Petstore System with all the components that are needed to run the
Petstore Web Application.
Back to Topic List
Linda's IT team is responsible for various internal and external
applications, as well as the system infrastructure that supports these applications.
Management wants to ensure that service levels for infrastructure components,
particularly the hosts that serve these critical applications, should be available
99% of the time for the last 24 hours at any given time.
Linda was also asked to ensure that all key transactions of
their critical application, their Pet Store Web application, is up 99% of the
time during business hours. If there are any potential availability or response
time problems for any of the key transactions, she would like to be able to
quickly isolate where the bottlenecks are so that she can have the appropriate
person address the problem.
Linda's management also requires a view and reports
so that at any point in time, they can see the real-time and historical status
of all services supported by their department, and the current and past service
levels so that they can measure and continuously improve on their service delivery
to their users.
Back to Topic List
A Generic Service is a simpler form of a service that can
be modeled using one or more supported service protocols. Supported service
protocols include various infrastructure, network, database, Web application,
Web services, and email protocols, and in addition, custom scripts are also
supported. You can define one or more service models by defining service tests
that represent a critical business functionality. Perform the following steps
to create a generic service that will use the ICMP Ping protocol:
|
1.
|
Open the browser and enter the
following URL:
http://<management
service hostname>.<domain>:<port>/em/
The default port value on a clean machine is 7777. However,
if there are other instances running on the machine, then the port may
be different.
The login page will be displayed. Enter the User Name
and Password, and then click the Login button.
|
|
2.
|
Click the Targets tab.

|
| 3. |
Click the Services subtab.
|
| 4. |
To create a Generic Service, select Generic Service from the Add
drop-down list. Click the Go button.
|
| 5. |
Specify a name for the service. Click the Select System button
to choose a system.
|
| 6. |
Select any generic system and click the Select button.
|
| 7. |
Click the Next button.
|
| 8. |
Select Service Test from the Define availability based on
drop-down list and click the Next button.
|
| 9. |
From the Test Type drop-down list select Host Ping.
|
| 10. |
Specify a name for the Service Test and also the host that the agent
will ping. You can use the host name or IP address of your workstation.
Keep the Mode as ICMP Ping. It will use the Internet Control Message Protocol
(ICMP) utility to test the availability of the remote host on the network.
Click the Next button.
|
| 11. |
Click the Add button to select a beacon.
|
| 12. |
Select a beacon that will monitor your service test. Click the Select
button.
|
| 13. |
Make this beacon a 'Key' beacon so that availability is determined from
the successful execution of your 'key' transaction(s) from this beacon.
Click the Next button.
|
| 14. |
Use the Performance Metrics page to define the metrics that will be used
to measure the performance of your service. Performance metrics can be
defined using either service test metrics, or system component metrics.
You can optionally change the default 'Ping Time (ms)' metric name and
the warning and critical thresholds. You can optionally add other performance
metrics to your service. Click the Next button.
|
| 15. |
Usage metrics measure workload or the user demand of the service. Usage
metrics are collected based on the usage of the underlying system components
on which the service is hosted. You can monitor the usage of a specific
component or statistically calculate the average, minimum, maximum, and
sum value from a set of components. Use this page to add usage metrics
for the service. Click the Add button.
|
| 16. |
Select Filesystem Space Available (%) from the Metric drop-down
list. Scroll down and click the flashlight icon to select the object.
|
| 17. |
Select all mount points and click the Select button.
|
| 18. |
Click the Continue button.
|
| 19. |
Specify the Warning and Critical Thresholds as 85 and 90
respectively. Click the Next button.
|
| 20. |
Review all the parameters and click the Finish button.
|
| 21. |
Generic Service Service_nnn is successfully added. Let your service test
run for few minutes and come back later to review the monitored results
for this generic service.
|
| 22. |
To monitor the results of the service, click the service name link.
|
| 23. |
The service home page provides an overview of the health of the service.
Click the Test Performance subtab to see the performance of your
service.
|
| 24. |
The Key Tests page shows you the historical and current performance of
the service tests as measured from each of the defined beacons.
|
Back to Topic List
A Web application is a special type of service that models
a Web-based application or a Web site. By defining a Web application target,
you can monitor the availability, performance, usage, and service level of the
application. In this section, you will create and monitor the Pet Store Web
application service by following the steps below:
|
1.
|
On the Services page, select Web Application
from the Add drop-down list and click the Go button.
|
|
2.
|
Specify a name for the service. The home page URL for the Petstore Web
application, for example, http://<hostname>.<domainname>:<port>/petstore.
After specifying the name and the URL, click the Select System
button to select a system.

|
| 3. |
Select the Pet Store System and click the Select button.
|
| 4. |
Make all the components 'key' components and click the Next button.
|
| 5. |
Select Service Test from the Define availability based on
drop-down list and click the Next button.
|
| 6. |
From the Create Web Application: Service Test page, you can define one
or more tests to define the availability of your Web application. The
test can be a single URL or a transaction that consist of one or more
URLs/steps. To define a service test / Web transaction to monitor the
availability of your application, specify the name of your transaction
as Browse Petstore, select the Record a Transaction option
button and click the Go button.
|
| 7. |
Click the Record button to record a Web transaction using the
Transaction Recorder.
|
| 8. |
If you are using the Transaction Recorder for the first time, you will
be prompted to install it on your workstation. Accept the installation
prompts to install the Transaction Recorder. Click the Start button
to record the transaction. This will open a new Internet Explorer browser
window. In the browser window, enter the starting URL of the transaction
and click through the steps that you want to record.
|
| 9. |
Browse through your Pet Store application with the URL that was assigned
to you. Navigate through the Web application by browsing some products,
sign in to the site, add items to the cart, and proceed to checkout. You
should logout of the application at the end of the transaction. Return
to the Grid Control console and click the Stop button. This will
stop the recording of the transaction.
|
| 10. |
Click the Continue button to complete the recording of the transaction.
|
| 11. |
Scroll down to the Steps section to view the recorded steps. You can
optionally change the default name of each step to be more descriptive
of the step.
Scroll down to Advanced Properties and expand this section. Select the
Step option under the Data Granularity drop-down list. This
will enable you to monitor your transaction on a step by step basis. Also
note the metrics under Collected Metrics that will be generated for your
transaction. Click the Continue button.
|
| 12. |
Click the Next button.
|
| 13. |
Click the Add button to select a beacon.
|
| 14. |
Select any beacon that will monitor your Web transaction. Click the Select
button.
|
| 15. |
Make this beacon a 'Key' beacon so that availability is determined from
the successful execution of your 'key' transaction(s) from this beacon.
Click the Next button.
|
| 16. |
Add a performance metric that is based on a service test metric. Select
the Based on Service Test from the Add drop-down list and click
the Go button.
|
| 17. |
Select Total Time(ms) from the Metric drop-down list and click
the Continue button.
|
| 18. |
Specify the Warning and Critical Threshold as 6000 and 8000
respectively for the newly added metric. Experiment with defining other
performance metrics for this service by adding metrics based on either
service test or system. Select Total Time (ms) from the Charts
on Home Page drop-down list and click the Next button.
|
| 19. |
Usage metrics measure workload or the user demand of the service. Use
this page to add usage metrics for the service. Click the Add button.
|
| 20. |
Select Database Instance from the Target Type drop-down list.
Select Tablespace Space Used(%) metric from the Metric drop-down
list. Click the flashlight icon to select the tablespace.
|
| 21. |
Select the USERS tablespace and click the Select button.
|
| 22. |
Click the Continue button.
|
| 23. |
Specify the Warning and Critical Threshold as 85 and 90
respectively and click the Next button.
|
| 24. |
Review your settings and click the Finish button.
|
| 25. |
Web Application Service Petstore_nnn is successfully added.
|
Back to Topic List
For each service, you can define a Service Level Rule . A
Service Level Rule defines the assessment criteria used to measure service quality.
It enables you to specify availability and performance criteria that your service
must meet during business hours, as defined in your Service Level Agreement.
For example, an E-mail service must be 99.99% available between 8 a.m. and 8
p.m.,Monday through Friday.
You can define only one Service Level Rule for each service. This Service Level
Rule enables you to evaluate the Actual Service Level over a time period and
compare the Actual Service Level against your Expected Service Level. Perform
the following steps to edit and monitor the services and service levels that
you defined:
|
1.
|
On the services page, click the link for the Pet Store
service you created to navigate to the service home page.
|
|
2.
|
To edit the service level rule for this service, click the Edit Service
Level Rule link from the Related Links section.

|
| 3. |
A default Service Level Rule is automatically applied when you create
a service. The default Expected Service Level is 85%. Change the Expected
Service Level to 99% and Business Hours as Monday to Friday
09:00 to 17:00 hours.
In the Performance Criteria section, move one of your defined performance
metrics from Available Performance Metrics list to the Selected Performance
Metrics list. If any of the selected performance metrics reaches the critical
threshold, then it is also considered as a service level violation. Click
the OK button.
|
| 4. |
Now let Grid Control run for some time to gather performance data for
your Web application. Come back later to review the monitored results.
|
| 5. |
To monitor the results, refresh the home page after few minutes. Note
the Actual and Expected Service Levels % for the last 24 hours. Click
the Charts subtab to observe the performance and usage trends.
|
| 6. |
Click the Topology subtab to observe the overall status of your
Petstore application.
|
| 7. |
The Topology page is used to view the relationship between the Web application
and its dependencies, including other subservices and key system components.
You can also view the causes of failure, as identified by Root Cause Analysis.
Potential root cause subservices and components are highlighted with a
red box.
|
Back to Topic List
You can monitor the performance of your service tests using
the Test Performance page. This page displays historical performance data of
your service tests. From this page, you can view the performance trends for
a single test as executed from one or more beacons in a single graph. This enables
you to compare response times from different beacons for the same metric.
You can also view details for multiple components, such as transaction, step
group, or step. If the data granularity in Advanced Properties for the test
is set to Transaction, then you will not find any data for step and step group.
To see data for the step group or the step level, you must set the Data Granularity
in the Advanced Properties section for the test to 'step group' or 'step'. Perform
the following steps to create additional service tests and to monitor test performance:
|
1.
|
Click the Monitoring Configuration subtab of
your Web application.
|
|
2.
|
Click the Service Tests and Beacons link.

|
| 3. |
Select any one of your service test (transaction) and click the Verify
Service Test button.
|
| 4. |
Click the Perform Test button to interactively run the service
test from the specified beacon to verify the response time at that specific
moment.
|
| 5. |
The results of the test are displayed in the Results table. Click the
Continue button.
|
| 6. |
Now you will create a new service test of the Web transaction type where
you will group one or more URLs into a single group. Select Web Transaction
from the Test Type drop-down list. Click the Add button.
|
| 7. |
Specify a name for your new transaction and click the Record button
to record the transaction using the Transaction Recorder. Record at least
5 to 6 steps using the Petstore application.
|
| 8. |
After recording the transaction click the Create button under
the Step Groups section.
|
| 9. |
For the first group, specify the name as Group1 and select the
first 3 steps. Then click the Continue button.
|
| 10. |
To create the second group again click the Create button under
the Step Groups section.
|
| 11. |
Specify the name as Group2 and select the remaining steps. Then
click the Continue button.
|
| 12. |
By default, only transaction level metrics are collected. To enable page
level data collection for a transaction, you need to define the Data Granularity
at the Page Level. Expand the Advanced Properties section.
|
| 13. |
Set the Data Granularity as Step and click the OK button.
|
| 14. |
To enable this service test select it and click the Enable button.
|
| 15. |
To change your availability definition, click the Availability Definition
link under the Related Links section.
|
| 16. |
Make your service test a Key Service Test and click the OK button.
|
| 17.
|
Click Yes for confirmation.
|
| 18. |
Click OK to save the changes.
|
| 19. |
Click the Test Performance subtab to view the performance of your
service test.
|
| 20. |
From the Key Test drop-down list select Browse Pets and click
the Go button.
|
| 21. |
Go to the All Tests section. From here, you can view the performance
of your transactions by service test or by beacon. From here you can find
out the potential performance bottlenecks within your Web application
transactions by looking at where the transactions are spending time. Click
any Service Test to go its home page.
|
| 22. |
Click the Performance subtab from the Transaction's home page.
|
| 23. |
On this page, you can compare the performance data for each transaction
from one or more beacons, or a metric for one or more transactions, step
groups, or steps.
Scroll down to the 'Steps and Step Groups' section. Click Expand All
in the All Metrics section.
|
| 24. |
You can view all the response times at the transaction, step group, and
individual step levels. Scroll down further to the Cumulative Time section
and click Expand All.
|
| 25. |
From here, you can view the response time breakouts (connect time, redirect
time, first byte time, html time, and non-html time) for the transaction
and all step groups and individual steps. Click the Home subtab.
|
| 26. |
Click the Play button to replay your transaction interactively
which will provide you with the response time details for each step of
your transaction.
|
| 27. |
After the transaction is replayed, click the Show Results button.
|
| 28. |
The results are displayed as shown below.
By using the Play with Trace option you can perform a more detailed
interactive trace, drill down to find where the most time is spent within
the application stack and drill down to SQL statements.
|
Back to Topic List
Perform the following steps to create a service dashboard:
|
1.
|
Click the Reports tab.
|
|
2.
|
Click the Create button to generate a Service Dashboard for your
services.

|
| 3. |
Specify the title as Service Dashboard nnn where nnn is your workstation
number. Select Monitoring from the Category drop-down list and
Dashboards from the Subcategory drop-down list. Select the Use
the specified target option button but leave the target box empty.
Expand the Options section and select the Visual Style as Dashboard.
After making all these settings click the Elements subtab.
|
| 4. |
Click the Add button to add reporting elements.
|
| 5. |
Select Services Monitoring Dashboard from the list and click the
Continue button.
|
| 6. |
Click the Set Parameters icon for the Services Monitoring Dashboard
element.
|
| 7. |
From the Available Services list select the 'Service_nnn' and 'Petstore_nnn'
services you created earlier and move them to the Selected Services list.
Click the Continue button.
|
| 8. |
Click the OK button on the Elements page.
|
| 9. |
The report is successfully created.
Scroll down the list and click the report that you just created.
|
| 10. |
You should see your dashboard with all the services you created earlier.
Note the status, performance, usage and service levels for the last 24
hours, 7 days and 31 days for each of your service.
Note: Under the Reports page, there are many out-of-box reports
subcategorized under the Monitoring section, such as Service Alerts, Service
Performance and Usage, and Web Application Transaction Performance reports.
Out-of-box reports have the owner name specified as 'SYSMAN'. Review these
reports to observe the variety of these reports provided for Services.
|
Back to Topic List
In this lesson, you learned how to:
 |
Create a Generic
Service |
 |
Create a Web Application
Service |
 |
Monitor Service and
Service Levels |
 |
Monitor Test Performance |
 |
Create a Service
Dashboard |
Back to Topic List
 |
To ask a question about this OBE tutorial, post a query on the OBE
Discussion Forum. |
Back to Topic List
Place the cursor over this icon to hide all screenshots.
|