How to guides Performance Profiler v4.1.2 - Initial Setup for a trial by uwn15494


									Haf Saba – Attachmate | NetIQ

 How to guides: Performance Profiler v4.1.2 –
            Initial Setup for a trial



                                       Prepared by

                                         Haf Saba

                                Senior Technical Consultant

                                        Asia Pacific

Haf Saba – Attachmate | NetIQ

Executive Summary

This document will walk you through an initial setup of AppManager Performance Profiler (AMPP). It is
designed to get you up to speed rather quickly on basic level monitoring and highlighting some of the main
features of AMPP. This guide is particularly handy for those who have just downloaded the product via the
website and are eager to get it installed quickly and see some value from the product. This install document
assumes you are providing 1 server to run the Application.


This guide in no way replaces the need for NetIQ Professional Services to assist, plan, deploy, and develop a
proper monitoring solution following NetIQ Best Practices. This is purely to showcase the benefits within a
trial period.


Please follow the screenshots and descriptions to install AMPP 4.1.2.

Basic Requirements:

       1 Physical or Virtual Machine running Windows Server 2003 (including service packs)
       Microsoft SQL Server 2005 SP2
       IIS
       Port 8080 and 8005 free (otherwise a port available to use for the Tomcat web service)
       Minimum 2 GB of RAM (highly recommended to have at least 3 GB or more)
       Separate Drives for SQL Data and OS files
       A Windows account (either local or domain) that has administrator rights
       A Windows account that has Server Administrator Rights to SQL Server
       Adobe Flash Player (used as a part of the web console)
       Administrator Access to your AppManager Operator or Control Center Console

Finally, you will need an existing running version of AppManager version 7.01. AMPP uses your AppManager
agents for data collection that will then go into their own analysis. This document assumes you have basic
knowledge of how to use AppManager.

If you are not familiar with AppManager 7.01, please contact your NetIQ sales rep about receiving the sister
guide to this document on installing and configuring AM 7.01 in a trial.

Haf Saba – Attachmate | NetIQ

    Begin the installation by running the setup.exe in the Win32 folder under to bring up the installation
  program. Step through the next install screens (although the screenshots are from a Beta build, the final
                                         version is exactly the same)

           The install is quite simple and painless provided all the requirements have been met.

Haf Saba – Attachmate | NetIQ

Haf Saba – Attachmate | NetIQ

Haf Saba – Attachmate | NetIQ

Haf Saba – Attachmate | NetIQ

                   It is recommended you do *not* change the cycle time for the trial

Haf Saba – Attachmate | NetIQ

Haf Saba – Attachmate | NetIQ

                An icon in the start menu will be created that will launch the web console.

                                  This will take you to the main screen.

Haf Saba – Attachmate | NetIQ

                         Use the username admin and password admin to log in.

           At which you will arrive at the following dashboard with no details currently defined

Haf Saba – Attachmate | NetIQ

Now we have AMPP installed and ready to use, we will need to turn back to AppManager to deploy the
AMPP scripts to the agents so we can now monitor them with AMPP. Log into the Operator Console to check
in the scripts for AMPP. The scripts are located here:

\program files\netiq\appmanager\performance profiler \scripts\qdb

Note: If your AppManager Operator Console is located on a different server, you may need to copy the qml
files in the above directory locally to the where the AppManager console is before you check in the scripts
otherwise you may have an error during the script check in process.

                            Highlight all the scripts in the directory for check in

Haf Saba – Attachmate | NetIQ

   You should see the above confirmation and the following scripts appear in a new tab called analytics

Haf Saba – Attachmate | NetIQ

The next thing to do is deploy the appropriate scripts to the corresponding servers. If you have existing
monitoring policies for different server functions, you can add the scripts to the appropriate knowledge
script groups and they will be automatically deployed. If however you do not have monitoring policies or are
unfamiliar with that term, you may just want to consider running the scripts manually. Just drag the
appropriate script on the appropriate server and press OK. There is no need to change any of the values of
the scripts.

For all Windows servers with an AppManager agent, you may run the CompleteNTCounters KS

For all Exchange servers with an AppManager agent, you may run the CompleteExchangeCounters KS

For all SQL servers with an AppManager agent, you may run the CompleteSQLCounters KS

For all IIS servers with an AppManager agent, you may run the CompleteWebCounters KS

For all Domain Controllers with an AppManager agent, you may run the CompleteADCounters KS

For all Unix servers with an AppManager agent, you may run the CompleteUNIXCounters KS

New scripts and templates in version 4.1.2!

In addition to the above, AMPP now introduces the following new scripts:

For all VMware VirtualCenter servers with an AppManager agent, you may run the VMwareHost and
VMwareVirtualMachine KS

For all Network Devices being monitored by an AppManager agent, you may run the NetworkDeviceChassis,
NetworkDeviceIP, and NetworkDeviceLANLink KS

Note: AMPP comes with a 30 day trial for 25 servers. Only deploy AMPP scripts to no more than 25 servers.

For your existing Knowledge Script Groups (KSG) that are a part of a monitoring policy, simply add the AMPP
KS to the KSG as shown below. Otherwise, run the scripts manually on the servers you want to monitor.

Haf Saba – Attachmate | NetIQ

               Verify the jobs are running and data is being collected in the graph data tab

             Next, head back over to the AMPP web console and click the Administration tab

Haf Saba – Attachmate | NetIQ

           Click on             and on the next page click on NetIQ AppManager

Haf Saba – Attachmate | NetIQ

                  Next, fill in the details for the data source connection and click submit

                                The following screen will be the confirmation

Haf Saba – Attachmate | NetIQ

               Now click        and on the following screen press submit

Haf Saba – Attachmate | NetIQ

        AMPP will scan the QDB and build objects for the servers which are running the AMPP jobs.

                                      After scanning, click continue

            Click on the                   link to take you back to the dashboard homepage.

                     You will see the Enterprise icon has been populated. Click on it

Haf Saba – Attachmate | NetIQ

       This drills down to various containers for the servers. Click the Server icon that has grey boxes

                   Now you see the servers you have added to AMPP with ? on each one.

Over the course of the next 30 minutes, AMPP will start to populate “workload” data along with “alarm
index” and “availability” values. The “health” values will take up to 2 hours before they are calculated. This
is quite normal behaviour. This data will be used for ongoing analysis. By design, AMPP will not generate
any “trusted alarms” for the first 24 hours as it captures a full days’ worth of data to be used.

Just above the circles, is an option to change between Health and Workload. Click the Workload button

Haf Saba – Attachmate | NetIQ

 The screen will automatically refresh and within a short period of time (in some cases up to 45 mins), you
      will see the icons change colour based on their performance. You will see something like this:

                   You will also see the following screen if you click on the Elements page.

Next, click on the Home link to take you back to the main dashboard page. This screen also shows the latest
incidents if there are any. Certain servers may take longer to populate values such as “workload” and will
continue to have a “?” listed. Again, this is normal behaviour. If you see the “?” after a few hours, read the
Troubleshooting section at the end of this document to determine what you can do to fix the issue.

Haf Saba – Attachmate | NetIQ

                  Towards the top right of the screen is the full-screen icon >

  Clicking this icon will show the dashboard in a full screen layout if you want this display to be permanent

 Should you click on one of your servers, you will be taken to a specific page showing various components
  about that server. This is very useful in determining other performance issues on a server and you can
                      access any modifications you need to make regarding a server.

Haf Saba – Attachmate | NetIQ

     In the following screen you will see a page for a server. There are a few components to consider:

   1. Incidents
          a. These are the alarms pertaining to a server
   2. Components
          a. Here you can see the server and scoring of various metrics (also shown clearly on the right)
   3. Service Level Objectives
          a. Here you can specify SLO targets that match your environment so alarms are triggered when
              your SLO is breached
   4. Model
          a. This is a very unique feature to AMPP. Here you can see very clearly a workload KPI score
              for each component that a server is being monitored.

Haf Saba – Attachmate | NetIQ

 A closer look at the example above shows variances in the KPI values for each application installed on the

This makes it very easy to determine where the majority of the performance increase is happening. In the
example above, the server is running Windows 2003 but also SQL 2005, IIS, and is a Domain Controller. The
Active Directory portion is the busiest of the 4 components.

Clicking on the Application link will show what components go into determining that number.

Clicking on the Percentage Workload Score will show which metric is the highest contributor to that score.

           Congratulations! You have successfully installed AppManager Performance Profiler!

It is highly recommended at this time that you contact your NetIQ sales and technical representatives to
discuss the value of AMPP and how to use it in your environments. This highly sophisticated tool will
correlate alerts, update information in AppManager, and remove false-positive alarms that you may
commonly see. Also, please refer to the User Guide by clicking Help in the top-right of the web page.

Haf Saba – Attachmate | NetIQ

Troubleshooting Common Issues and Basic Usage
Installation issues

Depending on the state of the Java installation on the machine, the Tomcat website that powers AMPP fails
to start. As a result, you will see the following screen when you attempt to open the AMPP web console:

To help resolve this issue, ensure the regional settings of the machine are set to US English. Also, restart the
                        NetIQ AppManager Performance Profiler 4.1.2 Tomcat service

Haf Saba – Attachmate | NetIQ

How do I add more servers to AMPP?

To do this, you will need to deploy the AMPP scripts to the agents you want to monitor. If you have already
done this but they are not appearing in the AMPP console, you will need to Re-scan the AppManager Data
Source. To do this, click on the Administration link. You will see your existing data source.

Simply click on your data source that is listed. Select Scan Data Source on the following page and press
Submit. It will automatically synchronise with the AppManager QDB.

Haf Saba – Attachmate | NetIQ

How do I remove servers from AMPP?

Click on the Elements link. This page will show all the servers being monitored. To remove a server you
must do 2 things.

   1. Delete the server from AMPP. Check the boxes for the servers you want to delete and choose the
      option to delete the agent.

   2. Stop and remove the jobs from the AppManager Operator Console. If the jobs are part of a
      monitoring policy, remove the Analytics jobs from the KSG associated with the policy. Otherwise, if
      you manually dragged the jobs to the servers you were monitoring, simply stop the jobs and delete

Haf Saba – Attachmate | NetIQ

Some servers still appear with a “?” for the workload value

Take a look at the following screenshot taken following an install. There are two servers which still have not
completed their “workload” and “health” value analysis.

To resolve this issue, click on the Elements page. Check the boxes for the servers with question marks, and
from the pull-down menu, choose Reset Baselines

                        Wait up to another 45 minutes to build a new index of values.

Again, please contact your local NetIQ Sales representative for a demonstration of the product and how it
can benefit your environment.


To top