How-to: Create Custom Introscope Metrics and

Alerts
With SAP Solution Manager 7.1

Introduction: Even though the content of the SAP delivered templates in the Monitoring and
Alerting Infrastructure is based on proven best practices for technical monitoring, from time to time
customers miss metrics they consider necessary for their customer specific monitoring. Also the
integration of 3rd party monitoring can be based on the metrics for the 3rd party system in Wily
Introscope Enterprise Manager.
This guide describes how you can create own metrics and alerts out of metrics that are available in
Wily Introscope Enterprise Manager but not in the SAP delivered templates of the Monitoring and
Alerting Infrastructure.

Contents
Example Metrics used in the Guide .......................................................................................................... 2
Step 1: Find the Path to the Metric in Wily Introscope Enterprise Manager ........................................... 4
Step 2: Determine the Data Collector in Solution Manager...................................................................... 7
Example 1: Average Response Time (ms) of the SOAP Servlet ................................................................... 8
Example 2: Accumulated Response Time per interval (ms) for the VMC Top Method Statistics ...................10
Step 1: Find out the correct variables for the managed object specific values ..............................................10
Step 2: Find a SAP data collector template as copy master .........................................................................12
Step 3: Adjust the variables of the new data collector template ....................................................................13
Step 3: Create the custom Metrics and Alerts .........................................................................................19
Example 1: Average Response Time (ms) of the SOAP Servlet ..................................................................19
Example 2: Accumulated Response Time per interval (ms) for the VMC Top Method Statistics ...................26
Step 4: Appliying and Activating the Monitoring Template ....................................................................31
Assign Custom Templates to Managed Object ............................................................................................31
Find the new Metrics in System Monitoring .................................................................................................32
Appendix: Further Examples for Custom Metrics ...................................................................................34
Example 3: Processor Utilization % (aggregate) ..........................................................................................34
Example 4: Waiting tasks by Work Process type .........................................................................................38
Troubleshooting .......................................................................................................................................42
Your custom metrics don’t appear in System Monitoring after activating the custom template ......................42
Your custom metric appears in System Monitoring but stays grey................................................................44
EXAMPLE METRICS USED IN THE GUIDE

In this document we will show the creation of custom Wily Introscope metric using two examples.
Example 1: Average Response Time (ms) of the SOAP Servlet
The metric measures the average response time for the SOAP servlet per java server.

This metric is very easy and straight forward to integrate. Its Introscope path contains no variables. We can
reuse the SAP data collector template for generic J2EE server node data. The metric will be created as
metric group (only distributed by server node) in the MAI template on J2EE instance level.

Copyright/Trademark
Example 2: Accumulated Response Time per interval (ms) for the VMC Top Method Statistics1
This metric grouping displays the "most expensive" in VMC methods by accumulated runtime. The
accumulated time is reset each interval of measurement.

This metric is much more complicated to include. It’s IS path contains the SID and the instance name of the
SAP system, which is different for each system. Furthermore, we want to create a metric group that contains
the response time for the VMC Top methods, no matter which ones these are. So we have to keep the
method name variable too by introducing an additional parameter to our IS path. So we cannot re-use an
existing data collector but have to create an own one, specific for this metric.

1
You will only find this metric on instances that run a VMC (The Virtual Machine Container is a component
integrated into the SAP Web AS ABAP that enables Java functions that comply with the Java Standard J2SE
1.4 to be executed in AS ABAP.)

Copyright/Trademark
STEP 1: FIND THE PATH TO THE METRIC IN WILY INTROSCOPE ENTERPRISE MANAGER
To include you metric in the system monitoring you have to know the resource path of the metric in Wily
Introscope Enterprise Manager. The metric path for every metric is shown when you select the metric in the
EM Investigator.

You cannot use this path directly. To use it in the Monitoring and Alerting Infrastructure you have to translate
it into a regular expression. Additionally you will not use the fixed host name and agent name for the metric,
since you want to re-use the template for more than one specific system and collect the metric for several
systems.
The path to a metric in Wily Introscope always follows the following structure:
*Super Domain*|<Coll hosts>|<IS process name>|<IS agent name>|<resource>|<sub-resource>…:metric
COLL_HOSTS = the host name on which the Introscope agents runs
IS_PROCESS_NAME = the process context of the metric
IS_AGENT_NAME = the name of the agent that collects the metric
IS_PATH = the regular expression that describes the path to the metric containing of resources, sub-
resources and the metric itself.

Copyright/Trademark
For your regular expression in technical monitoring you are only interested in the part of the IS path after the
IS agent name.
The regular expressions used in Wily Introscope EM follows the Perl 5 regular expressions language. To
translate the path in a regular expression you need to know the following rules for regular expressions.
You need to use an escape character (backslash) for separators and parentheses \| and \(and \).
o E.g. Servlets\|SoapServlet:Average Response Time \(ms\)
If there is no resource between the agent name and the metric you can enter the metric name
directly. Otherwise you have to separate the resource folders by \| and precede the metric with a
colon (:)
o metric
o resource:metric
o resource\|resource\|resource:metric
If you want to collect more than one metric from a resource or sub-resource follow the following rules:
You can include lists of things between parentheses using pipe characters
o E.g. Servlets\|Servlet(1|14|18):Average Response Time \(ms\)
Use ([^\|:]*) to represent one Resource segment
o E.g. Servlets\|([^\|:]*):Average Response Time \(ms\)
Use (.*) to represent "any."
o E.g. Servlets\|SoapServlet:(.*)

Copyright/Trademark
After translating your metric path in a regular expression, you can test if your regular expression is working,
using the “Search” tab on IS agent level. But be aware that this search only works for metrics that return
integer values. You will not be able to find text value metrics with it.

To search by regular expression, make sure you checked the “Use Regular Expression” checkbox.
Enter your regular expression and hit the “Go” button.

Check that your metric is returned by the regular expression.

Copyright/Trademark
STEP 2: DETERMINE THE DATA COLLECTOR IN SOLUTION MANAGER
In the step before you created and tested a regular expression for your Introscope metric path in Wily
Introscope. Now you have to select the correct data collector in SAP Solution Manager to add your metric to
the MAI templates. The data collector templates delivered by SAP can be found in the Data Collector
Templates view under:
http://<solman-host>:<solman-port>/sap/bc/webdynpro/sap/wd_mai_dpc_main

SAP delivers several generic Introscope data collector templates. If your metric is located in one of the SAP
trees and does not contain any additional custom variables in the IS path the chance is good that you can
find a predefined template to re-use.
Every IS data collector template contains at least the four parameters COLL_HOSTS, IS_AGENT,
IS_PROCESS_NAME and IS_PATH. Most of the parameters are predefined in the different data collector
templates, since they contain values that are determined during runtime. So the predefined value of the
parameters basically determines which template works for which metric.
E.g. the data collector template “Generic HostAgent data (Push)” contains the following parameters:

The variables in the parameters make sure it can be used for every system for single metrics located under
the SAP IS Hostagent process:

The following templates are available for the most common metrics:
Data Collector Template Metric Type
Generic HostAgent data (Push) Metrics collected by the IS HostAgent
which path does not contain SID or
instance
Generic Inst Data WilyHost Metrics collected by the IS HostAgent
(Push) which path does contain SID and
instance
Metric Location
*SuperDomain*|<host name>|SAP
HostAgent Process|SAP HostAgent
<SMD Agent Instance>
*SuperDomain*|<host name>|SAP
HostAgent Process|SAP HostAgent
<SMD Agent Instance>|<SID>|
<Instance>

Copyright/Trademark
Generic J2EE server node data J2EE metrics collected by the IS byte
(Push) code agent by java node
Generic J2EE instance data J2EE instance metrics that are
(Push) aggregated from metrics on different java
nodes
Generic J2EE system data J2EE system-level metrics that are
(Push) aggregated from metrics on different
instances and java nodes
*SuperDomain*|<host name>|SAP
Netweaver|<IS BC agent name>
*SuperDomain*|Custom Metric Host
(Virtual)|Custom Metric Process
(Virtual)|<Instance>
*SuperDomain*|Custom Metric Host
(Virtual)|Custom Metric Process
(Virtual)|<SID>

Example 1: Average Response Time (ms) of the SOAP Servlet

The first step after locating the metric in Wily Introscope EM is to check if you can re-use one of the SAP
delivered data collector templates.
The metric is located on the path *SuperDomain*|ldcisi7|SAP Netweaver|SI7_DVEBMGS88_server0 under
the IS Process “SAP Netweaver”. This fits the metric path for the collector “Generic J2EE server node data
(Push)”. Also the IS_PATH of our metric Servlets|SoapServlet: Average Response Time (ms) doesn’t contain
any additional variables. So we can go ahead with the SAP delivered data collector template.

The NODE_NAME variable in the template makes sure, that we catch the metric for all possible Java server
nodes. It is an additional grouping parameter. This additional parameter makes sure, that the incoming
values can be identified by MAI and that they are presented in a reasonable way in the System Monitoring.
Please proceed to

Copyright/Trademark
Step 3: Create the custom Metrics and Alerts to learn how to set up this custom metric in the MAI template.

Copyright/Trademark
Example 2: Accumulated Response Time per interval (ms) for the VMC Top Method Statistics
The second example metric contains variables in the metric path and has to be created as a metric group,
since the top methods contain more than one metric. Furthermore the name of the top methods can change
from time to time. These higher requirements are not fulfilled by one of the default data collectors.
So we have to create an own data collector template.
Step 1: Find out the correct variables for the managed object specific values
The metric path for the metric group we want to create is:

In this IS_PATH the red highlighted parts are additionally variable and determined at runtime.
For these parts of the IS_PATH we need to use variables in the data collector template.
Every managed object in the monitoring and alerting infrastructure has several attributes that describe it. The
values for these attributes are determined during runtime and accessible via variables. We need to find out
which placeholder to use, to replace the variable in the path with the correct attribute value.
To find the attributes for a managed object you first have to find the context ID for the managed object. Call
transaction SE16 and open table ACCONTEXTDIR. Enter the name of your managed object in the field
CONTEXT_NAME. Please be aware that the field is case-sensitive.

The result is a list of all managed objects that contain the search string in their context name.

Copyright/Trademark
Remark: If you don’t find your system in this table, please try the following: Go to SOLMAN_SETUP
Technical Monitoring System Monitoring Step “Define Scope” and select your system from the list.
Then change to the step “Setup Monitoring”. When entering this step for the first time for a system, the
landscape inquiry is executed and the system attributes are added to the table.
Copy the value in the field CONTEXT_ID of the correct managed object. In our case we are interested in the
instance running on ldcisi7.
Now go back to transaction SE16 and call the table ACCONTEXTDESCR. Enter the copied value in the field
CONTEXT_ID. Change the output list width to 990 to be able to see all information in the table.

The result gives you an overview over all attributes for the managed object and the variable to be used to
replace the value at runtime

So the IS_PATH for our metric would be

Copyright/Trademark
Step 2: Find a SAP data collector template as copy master
Now we check the parameters of the available data collector templates to find one we can use as copy
master. Since our metric is located under the IS process “SAP HostAgent Process” the data collector
template “Generic HostAgent data (Push)” is a good candidate as copy master. It contains the four
necessary variables and also leads us at least to the IS HostAgent node under which our metric is located.

But we will have to change the IS_PATH variable to fit in our variable system attributes and the additional
wildcard for the report name.
At first we create a new copy of the data collector template.
Select the template you want to copy and click the copy button.

Enter a meaningful Collector Description and a Data Provider ID. Make sure you data provider ID starts with
ZZ_ to make sure it is protected against monitoring content updates! Enter a meaningful name in the Data
Provider Name. This is the name you will see in the template maintenance later on. Click “Next step”.

Copyright/Trademark
Step 3: Adjust the variables of the new data collector template
Now you have to adjust the parameters of the data collector. With most of the parameters we are fine, but
some changes are necessary …

At first we delete the parameter PATH_NAME. It is not needed anymore since we will create a data collector
with a fixed path for one specific metric. Select the parameter and click the Delete button.

In the next step we have to adjust the parameter IS_PATH. We have to add our variable system attributes
and a variable for the VMC method name.
Select the parameter and click the “Entry one page up” button.

Now the parameter is transferred to the upper adjustment area.

Copyright/Trademark
At first change the Usage to PRECONFIG. Then change the predefined value. Delete the old entry
$PATH_NAME$. Instead add the variables for the system-specific attributes. Make sure you use the regular
expression notation \| to separate the singe resources.

Add the rest of the path until you reach the method name. The method name has to be a variable again, as
we want to collect the metric for all methods. We cannot use a predefined attribute variable, because there
doesn’t exist one. So we introduce the new variable METHOD_NAME. Make sure you put them into ($ … $).
This is the notation for free variables to be replaced at runtime.

Finish the rest of the IS_PATH variable. Make sure you don’t forget the escape character \ in front of pipe
and parentheses. Click the “Entry one page down” button to transfer your changes.
Hint: Don’t forget to test you PATH_NAME string in the Wily Introscope EM search as described in Step 1:
Find the Path to the Metric in Wily Introscope Enterprise Manager of this document. This makes sure your
path is correct and helps to identify typos. Replace the variables in the Introscope search by the values that
match the tree you are searching in. Test also the value you plan to use for the variable METHOD_NAME.
Using a metric group with a too unspecific regular expression (e.g. by using wildcards like .* for the variable)
can lead to a lot of results and to metric flooding.

Copyright/Trademark
Now you should have this entry in the lower table.

After we introduced the parameter METHOD_NAME we now have to define it. To define the new parameter,
just change the Technical Name to METHOD_NAME in the Parameter Definition area. Enter a meaningful
display name and description. Make sure you select “Extended Context”. By using the extended context you
make sure the multiple received values can be assigned to the correct method and are displayed in the
System Monitoring application separately. As a result an own sub-tree is built in the System Monitoring
application, for each value the variable METHOD_NAME has during runtime.

Enter the number 1 for the Sequence2. Change the usage to CONFIG and enter .* as predefined value. You
can overwrite the predefined value later on during the metric setup, by creating specific variants of the
metric.

2
The Sequence number defines which free parameter ($...$) in the IS_PATH value is addressed, as you can
have more than one free parameter in your IS_PATH parameter.

Copyright/Trademark
Then press the “Entry one page down” button to create your new parameter.
Now you should see the parameter in the lower table.

Click the “Next step” button.

In the last step you could add a value help for your configurable parameters. In our case this doesn’t make
much sense since we don’t know the method name before anyway. So click the Save button.

Copyright/Trademark
You are prompted to provide a transport request for your new data collector template.

After saving, you see a change report that tells you which SAP tables were changed.

You might now leave the Create Data Collector wizard by clicking on cancel.

Copyright/Trademark
You will be prompted for unsaved changes. But since you saved everything you can confirm the popup and
leave the wizard. Reopen the Push Introscope data collectors and you will find your newly created data
collector …

… with your adjusted parameters.

Copyright/Trademark
STEP 3: CREATE THE CUSTOM METRICS AND ALERTS
After identifying or creating an appropriate data collector template you now have to add the metric and the
alert to the MAI template.
Example 1: Average Response Time (ms) of the SOAP Servlet
Create a custom template for the managed object type you want to use the metric for. If you want to create a
metric for a non-SAP system, please use the “Generic Product Version” template for the copy. For metrics
related to the host or the database, please use the corresponding host or database template for the copy.
Most of the Introscope metrics are instance-level metrics. To figure out if a metric is system-level or instance-
level you can again use the location of the metric. If the SAP system instance name is part of the metric path
(even as part of the agent name) it is likely that the metric is on instance-level.
Open your custom template in edit mode and switch to expert mode:

Choose Create Metric

A new wizard opens. Enter a metric name, a category and a unit. Make sure you select “Metric Group” for
the metric class.

Copyright/Trademark
On the Data Collection tab select a collection interval and select “Introscope EM (push)” as data collector.
Select the “Generic J2EE server node data” as data provider. You see that the parameters NODE_NAME
and PATH_NAME are missing.

NODE_NAME works as extended context, which makes sure the metric (which is collected for every server
node) can be interpreted and presented by the System Monitoring correctly.
PATH_NAME is the path to our desired metric.
Press “Add Variant” to maintain the values for NODE_NAME and PATH_NAME. Make sure that you
maintain the path name as regular expression as described in this document before.

Hint: Don’t forget to test you PATH_NAME string in the Wily Introscope EM search as described in Step 1:
Find the Path to the Metric in Wily Introscope Enterprise Manager of this document. This makes sure your
path is correct and helps to identify typos.

Copyright/Trademark
Don’t change anything on the tab “Data Usage”, except you know exactly what you are doing.
On the tab Threshold maintain the thresholds for the metric. We decided to use numeric thresholds for the
metric.

The following threshold types are available in the templates:

• Already Rated & Already Rated (Best rating of last n measurements): The metric is already rated by
the data provider
• Counter Threshold: The metric rating is only changed if the rating value occurred n times in a defined
time interval. (e.g. 3 times red in the last 60 minutes). Please note that a red rating always also
counts as a yellow rating. This means if you have 1 yellow and 1 red rating in the defined time frame,
it is counted like 2 yellow and 1 red rating for the calculation of the metric rating.
• Delta Threshold: The rating is adjusted if the value of the metric changes in a certain amount of the
metric unit (e.g. the value increased more than 150 s from the current value)
• Info Only: Only information are transferred, no rating
• Numeric Threshold (Green/Red) & Numeric Threshold (Green/Yellow/Red) & Numeric Threshold
(GtoY/YtoR/RtoY/YtoG): The threshold is based on a numeric number, you have different threshold
types depending on how fine granular you want to get the metric rated
• Numeric Threshold (High watermark): A “best value” is provided, the rating depends on the
percentage of the best value (e.g. red if only 80% of the best value reached)
• Range Threshold: The thresholds define a range (e.g. green only between 50 and 100, yellow and
red ranges below 50 or above 100)
• Text Threshold & Text Threshold (Regular Expression): The threshold is based on a text pattern.
Text patterns can be defined as regular expression or as SQL select pattern. (e.g. green if
“available” is in the text)

Copyright/Trademark
Don’t change anything on the tab “Validity”, except you know exactly what you are doing.
On the tab others you have to assign a technical name for your metric.

Click “Next”. There are no custom alerts available yet, to which we could assign the metric. Click “Finish” to
close the wizard.

Make sure you save you template to persist the changes.

Now you can find you metric on the Metrics tab in the template.

Copyright/Trademark
The metric is marked as Custom-created.
To actually use the metric you need to create a custom alert for the metric as well.
Select Create Alert.

On the first tab enter a name, a category and a severity for the alert. Make sure that the category matches
the category of the metric since you cannot assign it to the alert otherwise. You can also add a custom
description.

You don’t have to maintain the tabs “Incidents” to “Auto Reaction” this can be done after the alert is created.
On the tab “Rule” you select the rule for your alert. The rule determines when an alert will be raised
depending on the values measured for the metrics in the alert. The following rules are available:
• Bestcase Rule: The metric with the best rating wins. This means Green is stronger than Yellow and
Yellow is stronger than Red.
• Worstcase Rule: The metric with the worst rating wins. The means Red is stronger than Yellow and
Yellow is stronger than Green.
• Average Case Rule: The average is calculated from all metrics. You can consider Green = 1, Yellow
= 2 and Red = 3. All metrics are added and the average is calculated. If the average is closer to 1 is
it Green, if it is closer to 2 it is Yellow and if it is closer to 3 it is Red.

Copyright/Trademark
On the last tab you provide a technical name for your alert.

Click next to assign your metric to the alert. Select the metric from the list of the unassigned metrics and click
“Finish”.

Save your template.

You can now see the alert and the assigned metric on the tab “Metrics, Alerts, Events Hierarchy”.

To improve the presentation of the metric in the system monitoring you can change the name of the event.
When the event is initially created it has the same name as the alert. But the events also build the parent
notes of the metrics in the System Monitoring application. So it makes more sense to assign a more neutral
name to the event. Select the event from the Metrics, Alerts, Events hierarchy and change the event name.

Copyright/Trademark
Save your template to persist the changes.

Copyright/Trademark
Example 2: Accumulated Response Time per interval (ms) for the VMC Top Method Statistics
The procedure for this metric is the same as for the metric before. Make sure you are in Expert mode and
create a new metric by clicking on Create Metric.

On the “Data Collection” tab you now select the data collector we created in the chapter before.

Copyright/Trademark
After selecting the data collector the values we maintained before are prefilled. You can now create variants
for different VMC methods.

We will create a variant that covers all methods. As mentioned before you should be careful when using very
unrestrictive wildcards like .* as this can lead to metric flooding.

Don’t change anything on the tab “Data Usage” except you know exactly what you do.

Copyright/Trademark
On the tab “Threshold” you maintain a threshold for your metric. We will again use a numeric threshold. The
different threshold types are described in example 1.

Don’t change anything on the tab “Validity” except you know exactly what you are doing.
On the last tab you have to give the metric a technical name.

Click next. On the next screen you see all custom alerts that are available for the metrics category.
Unfortunately none of them fits. So just click the “Finish” button.

Don’t forget to save your template to persist your changes.

Copyright/Trademark
You can find your metric group on the “Metrics” tab.

Now we create an alert for the metric group.

Select Create Alert.
Enter a name, a category and a severity for the alert. Make sure the category matches the metric you want
to assign to the alert.

You don’t have to maintain anything on the tabs “Incidents” to “Auto Reactions”. This can be changed later
on.
You can change the rule on the tab “Rule”. Which rules are available is described in example 1.

Copyright/Trademark
On the tab “Others” you add a technical name for your alert.

Click “Next”. In the next step you will see all unassigned metrics available for the alert’s category. Select the
correct metric from the list and click “Finish”.

Save your template to persist your changes.

The next step is to apply and activate your custom template to the desired managed objects.

Copyright/Trademark
STEP 4: APPLIYING AND ACTIVATING THE MONITORING TEMPLATE
The last step is to apply your custom monitoring template to the managed object.
Assign Custom Templates to Managed Object
In SOLMAN_SETUP Technical Monitoring System Monitoring go to Step “Select Scope”. Select your
system in the system list.

Click Next. In the next step you have to make sure that you template containing the custom Introscope
metrics is assigned to your system. Click the “Assign Templates” link.

And assign the correct custom template.

Copyright/Trademark
Repeat this for all managed objects of this type (e.g. all instances) if necessary. After assigning the correct
template click on the “Apply and Activate” “All managed objects” button to activate your template.

Click Next and activate the reporting for the system if necessary.
Find the new Metrics in System Monitoring
Enter the System Monitoring application via transaction SM_WORKCENTER Technical Monitoring
System Monitoring. Select your managed system from the list and open the System Monitoring application.

Click on the system name to open the system hierarchy.

In the right area you find the metric viewer. Here you can navigate to your metric. Since we have created our
metrics in the J2EE instance template, we have to navigate to the instance and from there to the category of
out metric. We created two Performance metrics.

Copyright/Trademark
Example 1: Average Response Time (ms) of the SOAP Servlet

Example 2: Accumulated Response Time per interval (ms) for the VMC Top Method Statistics

Copyright/Trademark
APPENDIX: FURTHER EXAMPLES FOR CUSTOM METRICS
The following examples might seem to be a little bit made up, but they mainly intend to show what
possibilities you have by using different wildcard options in regular expressions.
Example 3: Processor Utilization % (aggregate)
The CPU utilization of processes (Java VMs) instrumented with the Introscope Netweaver agent.

At first we have to create a new data collector as a copy from the SAP data collector “Generic J2EE server
node data (Push)” as the metric is locked under the IS process SAP Netweaver.

Copyright/Trademark
Then we create a new metric and alert in the template management as described in this document before.

On the data collection tab you select your new data collector.

Copyright/Trademark
As PROC_NO you can enter the regular expression 1|11|13 to select only the processors 1, 11 and 13.

Enter thresholds and a technical name on the corresponding tabs.

And assign it to an alert (may you have to create a new one as described in this document before).

Copyright/Trademark
The result can be found in System Monitoring:

Copyright/Trademark
Example 4: Waiting tasks by Work Process type
The dispatcher queue reflects the number of tasks waiting for a work process. This number may increase
significantly during work process bottleneck situation.

This metric is very similar to the VMC Top Methods metric we created before. But since the VMC is not
active in all systems and hence the metric is not necessary available we decided to create another example,
which you can recreate in your own system.

Copyright/Trademark
The first step is again to create an own data collector. We can reuse the data collector “Generic Inst Data
WilyHost (Push)” which already contains the variables for SID and instance. Again we have to adjust the
parameters of the data collector.

Now we again create a new metric using our new data collector.

Copyright/Trademark
Assign a technical name to the metric and don’t forget to assign it to a custom alert.

Copyright/Trademark
The result can be found in the System Monitoring application.

Copyright/Trademark
TROUBLESHOOTING
Your custom metrics don’t appear in System Monitoring after activating the custom template
Solution: Most likely the Diagnostics agent in the Wily Introscope EM host is not running or connected to
Solution Manager. Hence the monitoring configuration cannot be transferred to Wily Introscope.
Check Diagnostics Agent on Wily Introscope EM host is up and running
Please make sure the Diagnostics agent running on the Introscope Enterprise Manager host is up and
running. If you run a MOM3 cluster make sure the Diagnostics agents on all collector hosts run as well.
To check to which Wily Introscope EM your system reports, go to the Managed System Configuration for
your system to Step “Enter System Parameters” and check the field “Introscope EM”

To check which Diagnostics agent is assigned to the Wily Introscope EM, go to Basic Configuration of SAP
Solution Manager to Step 3.1 “Configure CA Wily Introscope”, select the Wily Introscope from the list and
check the field “SMD Agent” and “Server Name” in the table below.

3
MOM = Manager of Managers. It’s a scalability solution for Wily Introscope EM, if the number of metrics
collected are too much for one stand-alone Introscope EM

Copyright/Trademark
To check if the Diagnostics agent is up and running and connected to SAP Solution Manager, please go to
the Diagnostics agent administration (http://<solman-host>:<solman-port>/smd/AgentAdmin) and search for
the agent (e.g. by server name).

Copyright/Trademark
Your custom metric appears in System Monitoring but stays grey
The first step is to check if the metric is available with a reasonable value in Wily Introscope. If not it’s highly
likely that the Diagnostics agents for the managed system are shut down. They have to be restarted.
If the metrics are valid in Wily Introscope but do not appear in system monitoring you should check if the
Diagnostics agents for the managed system are also properly connected to SAP Solution Manager.
You can do this in the Agent Administration application.

In some rare cases the agents could be running, but cannot connect to Solution Manager because of invalid
credentials. As of SP 05 there is a new tab in the Agent Administration for these agents, the tab Non-
authenticated Agents.

Reconnect the diagnostics agents of the managed system properly.

After reconnecting you have to rerun the step “Apply and activate” the custom templates4.

4
Especially if you add metrics which are located under the IS process SAP HostAgent you need the SMD
agents of the managed system up and connected during the activation of the custom template. The SMD
agent instance name is used in the IS_PROCESS_NAME variable and this one is generated on activation.
When the agent is down or not connected to Solution Manager the wrong path is generated and stored in the
configuration file and the metrics will not be delivered.

Copyright/Trademark