cancel
Showing results for 
Search instead for 
Did you mean: 

Issue with Memory in SAP(R/3 3.1I )/OS (AIX 5.2)

0 Kudos

Hi

<p>

We have problem with the memory settings in SAP(R/3 3.1I )/OS (AIX 5.2). One of the customizing ABAP program failing at background but works fine when we run with same variant in forgeground.

</p>

<p>

Below are the current parameters values in system

</p>

em/initial_size_MB 1024 <br>

ztta/roll_first 1024<br>

ztta/roll_area 7000064<br>

rdisp/ROLL_SHM 6528<br>

rdisp/ROLL_MAXFS 23040<br>

rdisp/PG_SHM 5120<br>

rdisp/PG_MAXFS 34560<br>

abap/heap_area_dia 188743680<br>

abap/heap_area_nondia 400556032<br>

abap/heap_area_total 838860800<br>

abap/heaplimit 52428800<br>

abap/swap_reserve 20971520<br>

ztta/roll_extension 200068300<br>

em/blocksize_KB 1024<br>

em/stat_log_size_MB 40<br>

em/stat_log_timeout 600<br>

ipc/shm_psize_01 -10<br>

ipc/shm_psize_02 -40<br>

ipc/shm_psize_03 -10<br>

ipc/shm_psize_04 -10<br>

ipc/shm_psize_06 -40<br>

ipc/shm_psize_07 -10<br>

ipc/shm_psize_08 -10<br>

ipc/shm_psize_09 -10<br>

ipc/shm_psize_10 276000000<br>

ipc/shm_psize_31 -10<br>

ipc/shm_psize_33 -10<br>

ipc/shm_psize_34 -10<br>

ipc/shm_psize_40 688000000<br>

ipc/shm_psize_51 -10<br>

ipc/shm_psize_52 -10<br>

ipc/shm_psize_54 -10<br>

<p>

I have checked at os level with below command, Its shows there is problem at ipc/shm_psize_10 :276000000

ipc/shm_psize_40 : 688000000 but the current values are higher then what the below recomands. Can you plz suggest me the correct values for the above parameters. If any problems with above parameters.

</P>

<p>

witco-12:ncpadm 1> sappfpar check pf=NCP_DVEBMGS00_WITCO-12

</p>

<p><p>sapparam(2): fopen("NCP_DVEBMGS00_WITCO-12","r"): No such file or directory</p>

<p>No Profile used.</p>

<p>sapparam: SAPSYSTEMNAME neither in Profile nor in Commandline</p>

<p>================================================================================

==<p/> <p>Checking profile:</p>

<p>================================================================================</p>

<p>

***ERROR: Size of shared memory pool 10 too small</p>

<p>================================================================================</p>

SOLUTION: Increase size of shared memory pool 10<br>

with parameter: ipc/shm_psize_10 =36000000<br>

***ERROR: Size of shared memory pool 40 too small

<p>================================================================================</p>

SOLUTION: Increase size of shared memory pool 40<br>

with parameter: ipc/shm_psize_40 =50000000<br>

Shared memory disposition overview<br>

<p>================================================================================</p>

Shared memory pools<br>

Key: 10 Pool<br>

Size configured.....: 12000000 ( 11.4 MB)<br>

Size min. estimated.: 31988070 ( 30.5 MB)<br>

Advised Size........: 36000000 ( 34.3 MB)<br>

Key: 40 Pool for database buffers<br>

Size configured.....: 14250000 ( 13.6 MB)<br>

Size min. estimated.: 47166184 ( 45.0 MB)<br>

Advised Size........: 50000000 ( 47.7 MB)<br>

Shared memories inside of pool 10<br>

Key: 1 Size: 2000 ( 0.0 MB) System administration<br>

Key: 3 Size: 13100000 ( 12.5 MB) Disp. communication areas<br>

Key: 4 Size: 100000 ( 0.1 MB) statistic area<br>

Key: 7 Size: 14838 ( 0.0 MB) Update task administration<br>

Key: 11 Size: 500000 ( 0.5 MB) Factory calender buffer<br>

Key: 12 Size: 120000 ( 0.1 MB) TemSe Char-Code convert Buf.<br>

Key: 13 Size: 500000 ( 0.5 MB) Alert Area<br>

Key: 14 Size: 4400000 ( 4.2 MB) Presentation buffer<br>

Key: 16 Size: 22400 ( 0.0 MB) Semaphore activity monitoring<br>

Key: 17 Size: 65636 ( 0.1 MB) Roll administration<br>

Key: 18 Size: 65636 ( 0.1 MB) Paging adminitration<br>

Key: 19 Size: 6000000 ( 5.7 MB) Table-buffer<br>

Key: 31 Size: 1806000 ( 1.7 MB) Dispatcher request queue<br>

Key: 33 Size: 2048000 ( 2.0 MB) Table buffer, part.buffering<br>

Key: 51 Size: 3200000 ( 3.1 MB) Extended memory admin.<br>

Key: 52 Size: 20000 ( 0.0 MB) Message Server buffer<br>

Key: 54 Size: 16384 ( 0.0 MB) Export/Import buffer<br>

<br>

Shared memories inside of pool 40<br>

Key: 2 Size: 961296 ( 0.9 MB) Disp. administration tables<br>

Key: 6 Size: 26112000 ( 24.9 MB) ABAP program buffer<br>

Key: 41 Size: 6010000 ( 5.7 MB) DB statistics buffer<br>

Key: 42 Size: 596968 ( 0.6 MB) DB TTAB buffer<br>

Key: 43 Size: 3898344 ( 3.7 MB) DB FTAB buffer<br>

Key: 44 Size: 1850344 ( 1.8 MB) DB IREC buffer<br>

Key: 45 Size: 1338344 ( 1.3 MB) DB short nametab buffer<br>

Key: 46 Size: 20480 ( 0.0 MB) DB sync table<br>

Key: 47 Size: 3072000 ( 2.9 MB) DB CUA buffer<br>

Key: 48 Size: 300000 ( 0.3 MB) Number range buffer<br>

Key: 49 Size: 3000000 ( 2.9 MB) Spool admin (SpoolWP+DiaWP)<br>

<br>

Shared memories outside of pools<br>

Key: 1002 Size: 400000 ( 0.4 MB) Performance monitoring V01.0<br>

Key: 58900100 Size: 4096 ( 0.0 MB) SCSA area<br>

Nr of operating system shared memory segments: 4<br>

Shared memory resource requirements estimated<br>

================================================================<br>

Nr of shared memory descriptors required for<br>

Extended Memory Management (unnamed mapped file).: 1<br>

<br>

Total Nr of shared segments required.....: 5<br>

System-imposed number of shared memories.: 11<br>

Shared memory segment size required min..: 36000000 ( 34.3 MB)<br>

System-imposed maximum segment size......: 3489660928 (3328.0 MB)<br>

R/3-imposed maximum segment size.........: 2147483647 (2048.0 MB)<br>

<br>

Swap space requirements estimated<br>

================================================<br>

Shared memory....................: 86.2 MB<br>

..in pool 10 11.4 MB, 266% used !!<br>

..in pool 40 13.6 MB, 330% used !!<br>

..not in pool 0.4 MB<br>

Processes........................: 30.9 MB<br>

Extended Memory .................: 128.0 MB<br>

-


<br>

Total, minimum requirement.......: 245.1 MB<br>

Process local heaps, worst case..: 762.9 MB<br>

Total, worst case requirement....: 1008.1 MB<br>

<br>

Errors detected..................: 2<br>

Warnings detected................: 0<br>

</p>

Edited by: Sambireddy Authu on Jan 19, 2011 1:47 PM

Edited by: Sambireddy Authu on Jan 19, 2011 1:49 PM

Accepted Solutions (0)

Answers (5)

Answers (5)

0 Kudos

Hello Sambi,

on AIX 32bit the data segment which contains the heap is limited to 256M. With your current setting:

abap/heap_area_nondia = 400M

the data segment get exhausted and collidates with the stack. In the most cases you see a signal 4.

To solve the problem you should just set abap/heap_area_nondia to the same value like abap/heap_area_dia (180M).

Best regards,

Klaus

Former Member
0 Kudos

Sambi,

In SM04 -> Goto -> Memory are you able to see how much memory the process/report is using?

Former Member
0 Kudos
***ERROR: Size of shared memory pool 10 too small

================================================================================

SOLUTION: Increase size of shared memory pool 10
with parameter: ipc/shm_psize_10 =36000000
***ERROR: Size of shared memory pool 40 too small 
================================================================================

SOLUTION: Increase size of shared memory pool 40
with parameter: ipc/shm_psize_40 =50000000

Did you adjust above parameters as per the sappfpar suggestion? If not , please adjust and check.

former_member227281
Active Participant
0 Kudos

Hi

What is the error message or error log for background job terminated for that particular custom report.

Thanks and Regards

APR

0 Kudos

Hi APR,<br>

Thanks for the update.<br>

Job Log: <br>

1/18/2011 15:09:13 00516 Job started<br>

1/18/2011 15:15:40 00554 Job also cancelled due to process termination. See the system log<br>

Thanks<br>

Sambi<br>

former_member227281
Active Participant
0 Kudos

Hi Sambireddy,

Please check system log in tx code sm21 and corresponfing short dump st22.

Also please check the developer trace file located in /usr/sap/<SID>/DVEBMGS<nn>/work or enter transaction code al11 i.e. dev_w??. For example background job was running in background processing . Check the transaction code sm50. Note down the work process number say w8 . Then in developer trace file i.e.for example open file dev_w08 .

j This will help us to identify why the job was cancelled.

Thanks and Regards

APR

0 Kudos

Hi APR,<br>

I have checked sm21, St22 and dev_wp<no> file at OS level but I am not finding the exact error .<br>

Tue Jan 18 15:15:29 2011<br>

A WP has reached abap/heaplimit = 52428800 bytes<br>

Tue Jan 18 15:15:37 2011<br>

M -


C-STACK -


<br>

NONAME at 0x103804cc<br>

NONAME at 0x103806e4<br>

NONAME at 0x1046a02c<br>

Signal 0 in M -


<br>

M ***LOG Q0E=> DpSigGenHandler, signal ( 4) [dpuxtool 0501]<br>

M ThSigHandler: signal (step 4)<br>

M Modeinfo for User T17/M0<br>

M tm state = 2<br>

M uid = 18<br>

M term type = 0x0<br>

M display = 0x0<br>

M cpic_no = 0<br>

M cpic_idx = -1<br>

M usr = >GONUGRA <<br>

M terminal = > <<br>

M client = >088<<br>

M convid = > <<br>

M appc_tm_conv_idx = -1<br>

M imode = 2<br>

M mode state = 0x42<br>

M th_errno = 0<br>

M async_receives = 0<br>

M cpic_receive = 0<br>

M em handle = 4<br>

M roll state = 2<br>

M abap state = 2<br>

M em state = 3<br>

M eg state = 1<br>

M spa state = 3<br>

M enq state = 0<br>

M next hook = T-1/U-1/M255<br>

M master hook = T-1/U-1/M255<br>

M slave hook = T-1/U-1/M255<br>

M debug_tid = -1<br>

M mode type = 0x1<br><br>

M debug = 0<br>

M tcode = > <<br>

M client convid = > <<br>

M server convid = > <<br>

M lock = 0<br>

M max enq infos = 0<br>

M act enq infos = 0<br>

M

M *****************************************************************************<br>

M *<br>

M * LOCATION SAP-Server witco-10_NCQ_00 on host witco-10 (wp 17)<br>

M * ERROR ThSigHandler: signal<br>

M *

M * TIME Tue Jan 18 15:15:37 2011<br>

M * RELEASE 31I<br>

M * COMPONENT Taskhandler<br>

M * VERSION 1<br>

M * RC 11<br>

M * MODULE thxxhead.c<br>

M * LINE 7209<br>

M * COUNTER 103<br>

M *

M *****************************************************************************<br>

M

M ***LOG R68=> ThIRollBack, roll back () [thxxhead 9123]<br>

M ***LOG R68=> ThIRollBack, roll back () [thxxhead 9123]<br>

X <ES> RecoverAll<br><br>

X <ES> RecoverFreeList<br>

X <ES> RecoverAll<br><br>

X <ES> RecoverFreeList<br>

A<br>

A ** RABAX: could not save part LEV_SN_DATA_ENV of dump.<br>

A<br>

A Tue Jan 18 15:15:38 2011<br>

A<br>

A ** RABAX: could not save part LEV_SN_ITABS of dump.<br>

M ***LOG Q02=> wp_halt, WPStop (Workproc17 60272) [dpuxtool 0773]<br>

-


<br>

trc file: "dev_w17", trc level: 1, release: "31I"<br>

-


<br>

*<br>

  • ACTIVE TRACE LEVEL 1<br>

  • ACTIVE TRACE COMPONENTS all, M<br>

*<br>

M ***LOG Q01=> tskh_init, WPStart (Workproc17 2 19746) [thxxhead 0898]<br>

M calling db_connect ...<br>

B Try to connect as default user<br>

B Using SQL-Net V2 with tnsname = 'NCQ'<br>

B No TNS_ADMIN in environment<br>

B Setting TNS_ADMIN to '/usr/sap/trans' (profile-parameter DIR_TRANS)<br>

B Got 'NCQ' for SID from profile-parameter 'rsdb/oracle_sid'<br>

B Connecting via TNS_ADMIN=/usr/sap/trans, tnsname=NCQ<br>

B Got NLS_LANG=AMERICAN_AMERICA.US7ASCII from environment<br>

B<br>

B Tue Jan 18 15:15:40 2011<br>

B Now I'm connected to ORACLE<br>

B Database instance NCQ is running on witco-10 with ORACLE version 8.1.7.4.0<br>

M db_connect o.k.<br>

X <ES> client 17 initializing ....<br>

X ES initialized.<br><br>

A Attached to PXA (address 63026e00, size 621972496)<br>

S frontend print server enabled<br>

S Form-Textpool enabled (TSP06T has 4 fields)<br>

S OMS support enabled (TSPLOMS has 13 fields, TSPROMS has 8 field)<br>

M ThCheckPrevUser: clean previous user T17/U18/M0<br>

M Modeinfo for User T17/M0<br>

M tm state = 2<br>

M uid = 18<br>

M term type = 0x0<br>

M display = 0x0<br>

M cpic_no = 0<br>

M cpic_idx = -1<br>

M usr = >GONUGRA <<br>

M terminal = > <<br>

M client = >088<<br>

M convid = > <<br>

M appc_tm_conv_idx = -1<br>

M imode = 2<br>

M mode state = 0x42<br>

M th_errno = 11<br>

M async_receives = 0<br>

M cpic_receive = 0<br>

M em handle = 4<br>

M roll state = 2<br>

M abap state = 2<br>

M em state = 3<br>

M eg state = 1<br>

M spa state = 3<br>

M enq state = 0<br>

M next hook = T-1/U-1/M255<br>

M master hook = T-1/U-1/M255<br>

M slave hook = T-1/U-1/M255<br>

M debug_tid = -1<br>

M mode type = 0x1<br>

M debug = 0<br>

M tcode = > <<br>

M client convid = > <<br>

M server convid = > <<br>

M lock = 0<br>

M max enq infos = 0<br>

M act enq infos = 0<br>

M ***LOG R47=> ThResFree, delete (001023) [thxxmode 0780]<br>

B

B Tue Jan 18 15:22:09 2011<br>

B dbtran INFO (initialize_tran):<br>

B maxblocking_factor = 25,<br>

B minblocking_factor = 10,<br>

B maxin_blocking_factor = 254,<br>

B minin_blocking_factor = 10,<br>

B preferunion_all = 0,<br>

B preferunion_for_views = 0,<br>

B preferin_itab_opt = 0,<br>

B preferfix_blocking = 0,<br>

B usehints = 0,<br>

B optgele_to_between = 0,<br>

B convafter_aggregat = 0,<br>

B allowNULL_for_agg = 0,<br>

B rownumfor_count = 1<br>

volker_borowski2
Active Contributor
0 Kudos

Hi,

this sounds like the parameters have already been adjusted in RZ10, but the system has not

yet been restarted. Check your R3 instance uptime in ST02 or in SYSTEM->STATUS against

the timestamp on that profile.

Check the parameters in charge with RZ11 and/or with RSPFPAR to find out, what values

the parameters have in the currently active runnning instance.

Hope this helps

Volker

0 Kudos

Hi Volker,<br>

Thanks for reply.<br>

The Paramters was changed and instance was restarted.<br>

I have check rspfpar and the current values are same what I have given.<br>

<br>

Parameter Current Value Default Value<br>

pc/shm_psize_10 276000000 12000000<br>

ipc/shm_psize_40 688000000 14250000<br>

Thanks

Sambi