Registration

Dear SAP Community Member,
In order to fully benefit from what the SAP Community has to offer, please register at:
http://scn.sap.com
Thank you,
The SAP Community team.
Skip to end of metadata
Go to start of metadata

Purpose

This page is to explain the steps on how to perform checks if the SAP HANA instance is hanging.

 

General Things To Collect

  • Take RTE Dumps as per SAP Note 1813020  - How to generate a runtime dump on SAP HANA
  • On the HANA OS using SSH, take an output of the “top –H” command
  • Take screenshot of sub tabs of the performance tab in HANA studio
  • Take screenshot of SM66 output if it’s BW system
  • Additional ways to collect SM66 output:  /SDF/MON and SAP Note 42074 - Using the R/3 dispatcher monitor 'dpmon'
  • Diagnostic Queries can be run when system hang

Transparent Huge Page:

 

SAP Note 2131662 - Transparent Huge Pages (THP) on SAP HANA Servers

SAP Note 1824819 - SAP HANA DB: Recommended OS settings for SLES 11 / SLES for SAP Applications 11 SP2

SAP Note 1954788 - SAP HANA DB: Recommended OS settings for SLES 11 / SLES for SAP Applications 11 SP3

 

Power Saving:

 

SAP Note 1890444 - Slow HANA system due to CPU power save mode

 

Extremely High Memory consumption

 

  • When you run into system hang Situation please do not restart the Database.
  • Instead please take a look into SAP Note:  1813020 - How to generate a runtime dump on SAP HANA
  • Generate 3 to 5 dumps on about 30 seconds part.
  • Take a look into these dumps to find out which allocator is taking up the most memory. 
  • Look at SAP Note  1999997  - FAQ: SAP HANA Memory under Q13 for a list of allocator list and their purpose.

 

SAP Note: 1858357- HANA DB: Analysis of wait situations

SAP Note:  1989031 - Scheduling the shell script "thrloop.sh"

CPU maxed out

 

All CPU cores are completely used system wide

  • Check to see if it's HANA process that is consuming most of the CPU.
  • If it is, trigger runtime dump from hdbcons, or use OS tool gstack to trigger thread dump.
  • If it is not, consult with DB admin to see if that process can be terminated or restarted.

 

SAP Note: 1999020  - SAP HANA: Troubleshooting when database is no longer reachable

SAP Note: 1999670  - Operating system load analysis using "sar"

File System errors

 

SAP Note: 1641148  - HANA server hang caused by GPFS issue

HANA internal thread deadlock

 

  • It is possible due to programming error, that HANA's internal threads deadlocked (request handlers), which makes the system appear to hanging and not receiving any request.
  • This is not the deadlock situation caused by transactions locks, which can be resolved automatically by the database.
  • Run time dumps need to be collected.
     

SAP Note 1999998 - FAQ: SAP HANA Lock Analysis

 

 

 

 

 

Related Documents

 

 

 

 

Related Videos:

 

 

Related SAP Notes/KBAs

SAP Note 1732157 - Collecting diagnosis information for SAP HANA

SAP Note 2083715 - Analyzing log volume full situations

SAP Note 1788692 - Index Server crash due to missing LogSegment file

SAP Note 1813020 - How to generate a runtime dump on SAP HANA

SAP Note 1858357 - HANA DB: Analysis of wait situations

SAP Note 1989031 - Scheduling the shell script "thrloop.sh"

SAP Note 1999020 - SAP HANA: Troubleshooting when database is no longer reachable

SAP Note 1999670 - Operating system load analysis using "sar"

SAP Note 1999998 - FAQ: SAP HANA Lock Analysis

 

 

  • No labels