본문 바로가기

Database

[HANA] 2491748 - How-To: Analyzing Runtime Dumps with SAP HANADumpViewer

출처 : https://launchpad.support.sap.com/#/notes/2491748

 

https://launchpad.support.sap.com/#/notes/2491748

 

launchpad.support.sap.com

 

2491748 - How-To: Analyzing Runtime Dumps with SAP HANADumpViewer

Symptom

You are interested in analyzing SAP HANA runtime dumps

Environment

SAP HANA

Cause

Attention: The SAP HANADumpViewer functionality is now fully integrated in the more comprehensive SAP HANA Dump Analyzer (SAP Note 2498739). It is strongly recommended to use the SAP HANA Dump Analyzer for evaluating dumps in SAP HANA contexts.

SAP HANA runtime dumps (SAP Note 2400007) are used for various analysis purposes and contain a lot of information in text format. The SAP HANADumpViewer can be used to display certain important pieces of information in a different way in order to simplify the analysis. For example, it is able to aggregate the thread call stacks graphically in order to see quickly what the most dominant thread activities were.

This script is an expert tool designed by SAP support. You are allowed to use it, but SAP doesn't take over any responsibility for problems originating from the use of this tool.

Resolution

Here you can find the SAP HANADumpViewer and a presentation that provides further information how to use this tool.

SAP HANADumpViewer is a Python script that can be executed directly on a SAP HANA database server (where Python is available per default) with the <sid>adm user or on any other host (including Windows machines) where Python is installed. It takes the runtime dump file as input and generates a file in dot format as output. SAP HANADumpViewer doesn't connect to the SAP HANA database, so neither credentials nor a running SAP HANA database are required when using it.

The following tasks can be performed with the SAP HANADumpViewer:

Task Option Details
Generation of a graphical call stack hierarchy representation  

The following command can be used to generate a graphical call stack hierarchy representation:

The output file is generated in DOT format and can be displayed with GraphViz tools like WebGraphviz. The following color conventions are used:

  • Shades of red for frequent modules (more red -> more frequent)
  • Blue frame for final modules (i.e. modules where threads are currently active)

Example:

 

Extraction of [INDEXMANAGER_WAITGRAPH] -mw

The [INDEXMANAGER_WAITGRAPH] section of a runtime dump may already contain a wait graph in DOT format. This section can be written to a dedicated file using the "-mw" option.

Extraction of monitoring view data -mv

The section [STATISTICS] of runtime dumps contains raw data of specific monitoring views. The "-mv" option can be used to extract the different outputs to files in CSV format for easy evaluation in Excel.

For feedback and suggestions you can contact Christian Hansen.

Keywords

SAP HANA dump viewer HANADumpViewer runtime dump rtedump graphviz dot call stack

'Database' 카테고리의 다른 글

[HANA] Load 측정 KPI - Current Load  (0) 2020.03.12
[HANA] Load 측정 KPI - Load History  (0) 2020.03.11
[HANA] Scale-out Workshop with SAP (3/3)  (0) 2020.03.06
[HANA] Group지정 및 Table 위치 지정  (0) 2020.03.05
[HANA] Table Replication  (0) 2020.03.05