cancel
Showing results for 
Search instead for 
Did you mean: 

ABAP NW 7.40 Swap Space Spikes

Former Member
0 Kudos


Periodically I encounter a spike with Swap Space on an application sever in production.  It will spike to about 60%.  At the AIX level using TOPAS -P I can see many disp+work.

The work around is to pull this application server out of the Message Server/ASCS running on separate servers and go through the prep work for shut down using transaction smms, smlg, rz12, sm61, rz11, sm14.

My question is if there is a better method at resolving this?  Our kernel is 745 patch 119 and we apply kernel updates every 6 months.  We are running NW 740 Support Pack 10.  Currently applying SPS updates on six to other landscapes prior to touching production.  Other than the typical stuff of updating patches etc., I would really like to find the root cause and fix it once and for all.

Dan Mead

Accepted Solutions (0)

Answers (1)

Answers (1)

Reagan
Advisor
Advisor
0 Kudos

Assuming that the app server is running on a dedicated server I would first check the memory available on the machine and the memory configured for the app server. AIX will make use of the available memory before using the page space. In your case you should look whether the physical memory usage is high or not. If yes then you should check whether the system is correctly sized. In addition to that check the paging space configuration based on the note 1121904 - SAP on AIX: Recommendations for paging space also check if AIX is set up as recommended

973227 - AIX: Virtual Memory Management Tuning Recommendations

1972803 - SAP on AIX: Recommendations

If the issue is happening only on one of the application servers then check if the number of users using sent to that app server is huge. Also check if all the AIX servers are on the same patch level.

Without checking the system it is difficult to understand where the issue is. Start with the memory allocation and usage, system configuration like ulimits, AIX parameters etc.