cancel
Showing results for 
Search instead for 
Did you mean: 

Performance issues on LiveCache

Former Member
0 Kudos

Hello,

We are experiencing some performance issues in our LiveCache. So i have checked the logs. Since one week, i keep on having the following warning :

Low Number of cached history pages (160313) exceeds number of cached OMS pages (106651)

And when i check the caches in LC10, i have the following info :

OMS Data 853 200 106 650 2,84%

History/Undo 1 282 504 160 313 4,28%

So it means i have more history versions than active versions in the OMS cache. How do i fix this issue ?

We don't have any space problems as we have:

cache accesses for OMS Data : 99,99%

and:

Entire Data Cache 2 147 176 268 397 7,16%

Could you please advice on what I should check ?

Thanks in advance for your help.

Y.

Accepted Solutions (1)

Accepted Solutions (1)

lbreddemann
Active Contributor
0 Kudos

Hi Youssef,

you may want to open a support call for this.

It's much easier to look into this issue via a service connection then by this forum thread.

What you can do is (of course): check whether you get any warnings for the CHECKS page in /SAPAPO/OM13.

Maybe there is some job not scheduled correctly.

Regarding the warning message you posted, you may also want to review the liveCache parameters (e.g. retention time...).

regards,

Lars

Former Member
0 Kudos

Hello Lars,

thanks for your answer ... All the jobs are scheduled. But we don't use the DB13 to schedule them. We use an external scheduler for that (controlm). So we have all the technical jobs programmed in the external scheduler on both the QA and PRD systems. Only the dev system is not.

When i launch the /sapapo/om13 on all 3 systems, i have realized that everything is green on the dev system, but on the prd and qa, i usually (but not everyday) have two warnings :

error om13

and

/sapapo/om_reorg_daily (but only the /sapapo/om_delete_old_simsess)

I have checked all the jobs scheduled by controlm and all the necessary jobs (compared to the dev system) are scheduled. So i was wondering whether the system checks the DB13 when we run the /sapapo/om13 or just if the jobs are successfully programmed and run ...

former_member229109
Active Contributor
0 Kudos

Hi Youssef,

1) Concerning warnings in //om13 -> checks, please check if you have errors listed in LCA logging folder < or //om11 >. Please check that the recommended LCA jobs scheduled to run in the liveCache relevant clients.

Please also have a look at Information on how to avoid the red lights, it can be found via the blue info buttons.

2) "Low Number of cached history pages (160313) exceeds number of cached

OMS pages (106651)"

As you know, this message can occur for liveCache instances only.

In the data cache, the number of history pages is higher than the number of OMS data pages.

=>>

Please check, if you had the old OMS versions and transactions that prevent the history pages from being deleted in the data cache.

Also please check whether the reports /SAPAPO/OM_REORG_DAILY and /SAPAPO/OM_DELETE_OLD_SIMSESS are being executed at regular intervals; see SAP notes 800927, 954261 & 679118.

3) You could check/monitor if you have too old OMS versions & long running transactions in LC10.

4) Do you have Performance problems in liveCache?

What is the vesion of your system? What is the version of the liveCache? LCA build? SCM SP? LCAPPS SP?

What applications are you running on your system? Do you have the APO and liveCache installed on the same server?

Do you run the liveCache applications in parallel? How much data do you have in liveCache?

For SAP liveCache documentation please see the SAP note 767598.

5) As you know, that as SAP customer you have the option to create the SAP ticket to the component 'BC-DB-LVC' or 'BC-DB-LCA' , and we could login to your system via OSS and check the reported issues on the system.

Thank you & best regards, Natalia Khlopina

Answers (0)