System Analysis Troubleshooting

This page contains some hints for troubleshooting when for example some dashboards or views are empty.

System Monitoring (MAI) Metrics

Most pre-defined dashboards per system type are populated with metrics from the Monitoring and Alerting infrastructure that is also providing the content for System Monitoring. As a result, if any metrics do not report data, the first check is to verify if system monitoring is working correctly and if metrics in System Monitoring report data. If System Monitoring is not working correctly, please open a support ticket with component SV-FRN-APP-SYM.

Common checks
  1. If System Analysis dashboards show no metric data for a particular system: Navigate to System Monitoring for the same system. You can use the link in the toolbar (top right) to directly select the current system in System Monitoring. 
    1. If System Monitoring does not report metric data for a particular system either, then open a support ticket for System Monitoring (SV-FRN-APP-SYM).
    2. If System Monitoring reports data properly and metric data is only missing in System Analysis in a freshly configured FRUN system: potentially System Analysis was launched before setting up System Monitoring. Up to FRUN 2.0 FP02, this generates a wrong default configuration. Double check table WEA_UI_CONFIG: if there is an entry "raw.days" it should have the value 28. If the value is 0 then correct it to 28 and relaunch System Analysis.
  2. No data for historical time ranges (age > 28 days): by default System Monitoring data is retained for 28 days. You can configure System Analysis Aggregation to retain selected metrics for selected systems for a longer period. 
     

RCA Capabilities

In general you can check table RCA_INBOUND_LOG in FRUN via SE16, MESSAGE_TYPE="E" will filter on errors and and the date ( field UTC_DATE) . Here you can see errors of data that are sent to FRUN. When changing the user parameters in SE16 (Menu: Settings-> User Parameters -> ALV Grid Display) you can see the detailed error messages and can download the table.
However there might be already problems in collecting data either in the agent application, in the managed system or on the way.

HANA Metrics

HANA Metrics are collected via an operation of the SAP Hostagent. Check in Agent Administration for the HANA system if the collectors are running:

  1. Open Agent Internals
  2. Select tab Collector Framework
  3. expand the capability HANA_*
    Check for exceptions or if the coverage is below 1.
  4. Select the tab Logs
    Choose the thread rca worker <collector_key>

ABAP_WP_OVERVIEW

ABAP work process samples (/SDF/SMON) are collected via a RFC from the SDA usually with user SDAGENT. Check in Agent Administration for the ABAP system if the collector is running:

  1. Open Agent Internal
  2. Select tab Collector Framework
  3. expand the capability ABAP_WP_OVERVIEW
    Check for exceptions or if the coverage is below 1.
  4. Select the tab Logs
    Choose the thread rca Worker <FRUN_SID>.ABAP_WP_OVERVIEW_<Managed_SID>
In the managed system you can check in /SDF/SMON if a daily Monitoring is scheduled. If not the user SDAGENT have probably not all roles (check security guide).

STATRAGG

The page ABAP Performance does not display any data which could have multiple reasons:

  • Check in the managed system in ST03 if aggregated data is available. If this is not the case follow SAP note 2369736 - Troubleshooting missing data in ST03N/ST03.
  • Check if SSI has been executed successfully (step 2 for activity Configure Statistical Records Aggregation)
  • Use the link "ABAP Performance KPI Setup" in System Analysis to open the "Statistics Records Aggregates Administration" (/sap/bc/webdynpro/sap/wd_statragg_cockpit). From here you can check:
    • Which instance collects the data
    • Status of collection
    • Jump to Agent Administration for logs and Agent Internals → Collector Framework. There should be a collector "statragg" for the system.
    • Find out which agent is relevant for collection. 
      Filter for the system. You can see from which instance the data is collected.
      If you have multiple instances you might need to add the column HOST_NAME.With this physical host name you can search then in Agent administration for further issues.
  • Agent Administration
    Open Agent internals and click on tab Logs.
    STATRAGG data are usually collected around the full hour. Enter in the the last full hour (usually in UTC) and then you should be able to select in the Threads List box the statragg collector: