Skip to Content

Monitoring FI/CA application by use of SAP Solution Manager & the EMMA

*Introduction*  In this article I´m going to show you an example of how you can use EMMA and BPM.   *Problem* In more complex system landscapes built with SAP and nonSAP systems the application support units are challenged by monitoring and managing tasks. The units therefore need strong and web-enabled tools which enable them to face this. Monitoring of running and completed/erroneous business processes gets more and more important. I will try to give you an idea of what has to be done within the SAP *Solution* *Manager* (release 4.0) and the satellite systems to afford this. As you will see, logging is crucial for that. With *EMMA* you will be able to aggregate huge amounts of messages to *business* *areas*. At best a relation to one single *business* *object* can be established. If an *error* occurs, you need quickly the *step* inside a business process where the error occurred.  *Prerequisites* Ensure to have the newest ST A/PI add-on and ST-PI components installed in your satellite system. Before you start, ascertain that you have the current versions of the following software components established in your SAP Solution Manager system: (software levels of June 29th 2007) 0.1. ST-SER 700_2007_1 SP1 0.1. ST-A/PI 01G_CRM500 The appearance of business process monitoring in the solution manager can be seen in fig.1:  image ICON ALERT: image                           source: SAP {code:html}{code}.  This example shows a 3-system-landscape where one process starts in a CRM system (step 1), goes on in the TT4 system end ends in the WMS system. The icon “ICON_ALERT” shows that on these steps an *alert* was triggered. Many things about application monitoring objects and how to setup BPM are well described in the OSS, e.g. available monitoring objects in OSS note *945081* ( I will point out the use of *+BOMASS01+* – Mass Activity Monitor (as of Rel. 4.72) for the mass activity (FPCOPARA) and *+BOPERFMO+* – Dialog Performance Monitor for Relevant application areas: a) Cross-application and b) ST-A/PI Release is version 01G to show the number of calls for a certain function module to request correspondence. A very good idea of how to setup BPM in SOLMAN is described here (remark: you’ll need an OSS user for that): {code:html}{code} .It’s called *Business Process Monitoring – Setup Roadmap*. source: SAP AG. In the next lines I´m concentrating on how to monitor the number of entries in *application* *logs* and the therefore created number of *clarification* *cases*. (the *EMMA* stuff). The example I will use is the *+mass+* +activity+ (FPCOPARA) to “print correspondence” within a FI/CA application. *Overview of steps to be done for productive usage of BPM* | Create Solution | SOLMAN(SM) | transaction [DSWP]  | | please enter *solution name*, your *customer* *number* and a *language* used for stored documents in knowledge warehouse.| – | | Create logical component |  SM |  [SMSY] | |  for systems which are nonSAP a “blackbox” can be created to show *interfaces* between these systems and SAP. You will have to create new *products* for nonSAP systems too. Also your *productive* *systems* have to be maintained in the logical components.|   | | Create business scenarios, processes and steps |  SM |  [SOLMAN_DIRECTORY] | | set status of *scenarios* and *processes* to “production”; enter the *steps* you want to use in the monitoring. a. a *local* *RFC* *destination* will be generated b. in *satellite* system a *user* like “*CSMREG*” is needed with RFC authorities. This is described in OSS note *455356*.|   | | optional: setup interface monitoring in satellite system |  SAT |   |   |   | | optional: setup interface monitoring in solution manager system  |  SM |   |   |   | | generate monitoring customizing and specify data collector frequencies |  SM | | check it in (RZ20) on SOLMAN. | | If there occurs any dump it will be seen in (ST22) of satellite system. |   | | activate monitoring |  SM |  [DSWP] |   |   | | Use the monitoring |  SM |  [DSWP] |   |   |   Picture (fig.2) mentioned in the table above image fig.2: Scenarios, processes, steps and connections between steps Remark: step 2 (receive BAPI call) will be monitored with the *+BOPERFMO+* monitoring object whereas the processing of the dialogue step 3 will be monitored by investigating the messages in the *application* *log*. Now the configuration of the *application* *log* *monitor* is described. 1. Make sure that you know the application log object key and the sub-object´s key. Be sure to know the ID of the error messages (message id and number).  Subobject in  (SLG0) fig. 3: Subobject in (SLG0) This can be maintained in (SLG0) as shown in fig. 3.(remark: _transaction_ _names_ are always shown here with the brackets). In my example I want to throw an alert if a “parameter error” occurs. This is marked in the application log as a message of ID +ZKK_AP12+ with message no. +010+. The object key is ZKK_KS2T and the sub-object is Z_KS2T_DB. You can see an example in fig. 4 which shows some EMMA Messages.  EMMA Messages fig. 4: EMMA Messages 2. go to: (DSWP); select your solution; use “SETUP business process monitoring”. 3. select the process step in the tree view and mark “application log”: (see fig.3) image fig.3:  Process Step   4. mark “Application Log” and specify a user (* if it’s not a special one) and the keys for the application log entries. The example in fig. 4 shows an hourly aggregation (the H in column 4). image fig 4: Keys for the application log entries  5. You have to maintain the “critical application log messages” branch to react on a special message.   image fig. 5: Customizing of Alerts  In fig. 5 it is shown that we want to see a *warning* (*yellow* alert) if only 1 message of ID ZKK_AP12 with number 010 has been stored in application log. The number of critical messages for a *red* *alert* can also be maintained here. 6. Generate the monitoring *customizing* 7. Goto (DSWP) –> operations and *activate* monitoring (if it’s the first time).
8. Use the monitoring.



 fig. 6: Alerts in the monitoring view

As seen in fig. 6 a red flash (alert) is shown if at least 1 critical message has been reported. The corresponding entry in the application log looks like fig. 7:


fig. 7: Details of a message 

The details of this message can be shown by using the function button marked with the glasses. Then you see that the message id used is +ZKK_AP12+ and the number is +010+.

*Analyzing the mass activity*

Now let’s start (FPCOPARA) (this is the transaction for correspondency printing like bills etc.) and the required steps.  (–> this will lead us to the previously mentioned usage of the application monitoring object *BOMASS01*) 1. First of all get yourself informed by the *mass activity* and the *jobs* which are needed to perform the process chain. In this case I use the report “RFKK_MASS_ACT_PARAMETER” to copy the parameters for the mass activity (FPCOPARA), afterwards I make use of the report “FKJO_SCHEDULE” to start the mass activity. If the run of report “FKJO_SCHEDULE” was successful, your system shows child jobs for each interval processed. The name of each child job (see field Jobname in fig. 8) starts with +COPR+ (if you use standard customizing). 2. The mass activity can be analyzed locally via transaction (ST13). Please use the tool *+MASS_MAN_MONITORING +*for that. 3. example:   image

Be the first to leave a comment
You must be Logged on to comment or reply to a post.