cancel
Showing results for 
Search instead for 
Did you mean: 
Subscribe

Hi SAP-xi guru,

Can please let me know what are production issues that commonly come in SAP-PI system and how to resolve them.

Thanks & Regards

Vijaya

0 Likes
View Entire Topic
Former Member
0 Likes

Hi,

Well certainly there are NO free-defined production issue but there are no of area that you might require proactive monitor at regular intervals.

Tcode: SXMB_MONI

1. Process XML messages u2013 Standard and Process (check for message failures, determine if ABAP or Java related)

2. Job Overview

3. Persistence Layer Analysis (Observe growth day to day, no sudden jumps).

4. Tcode: SMQ2, SMQ1 (check for blocking)

5. Tcode: SM66 (no long running DIA)

6. Tcode: SM21, ST22 (look for unusual errors)

7. Tcode: ST04 (Overview - hit ratios)

8. Tcode: ST06

1. CPU (no overloads)

2. Memory

9. Tcode: ST03 (Workload overview. Response times, special attention on DIA, HTTP)

10. Tcode: SMICM (for each server: look for Current/Peak/Maximum values and make its not Peak is not hitting MAX)

RWB:

1. Message Monitoring u2013 Database (overview) -Integration Engine and Adapter engine (Look for unusual errors)

2. Performance Monitoring u2013 last 7 days/Daily (look for volume jumps)

More inportantly if you have Wily INTROSCOPE in your environment look for trends in messages, garbage collections, etc.

And more in detail documentation on 'Process Integration PI 7.1' Trobleshooting Guide

https://websmp104.sap-ag.de/operationsNWpi71

-> Process Integration

-> Troubleshooting Guide - SAP NetWeaver PI 7.1

I hope this helps you.

Regards

Sekhar

Edited by: sekhar on Jan 4, 2010 3:43 PM

Former Member
0 Likes

Hi Sekhar

When do we use transaction ST04 ? Why do we need to use it ?

Also how do we check the Current/Peak/Maximum values in SMICM and why should we do that ?

Ajith

Former Member
0 Likes

st04 gives the number of times the cache is being hit.

Just going to tcode smicm gives Current/Peak/Maximum values.

Both are used to monitor the performance of the system. For instance if the system is unusually slow, these monitoring features can give you good idea of whats happening behind the scenes.