cancel
Showing results for 
Search instead for 
Did you mean: 

DBIF_RSQL_NO_MEMORY or DBIF_REPO_NO_MEMORY on bakcgorund job

Former Member
0 Kudos

No free memory available in class PERM . Memory block ID "DBSL ", 10932 bytes.

this is the error in Sm21

i am using NW7.0 On 32-bit Windows with SQL server

any idea how to correct this?

this happens when i run language sumplementation

strange thing is this was run before with no problems ()

it seems to happen on TRDIR table

Edited by: Yavor Markov on Jan 30, 2009 9:57 AM

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Hai,

This is swap space problem in the system.

Check the below link.....

http://help.sap.com/saphelp_nw04/helpdata/en/fb/cbb96cd9ee11d188be0000e83539c3/frameset.htm

Regards,

Yoganand.V

Former Member
0 Kudos

well my swap space is 20 G

and this system is hardly ever used as it is not configured yet

and this is windows- extended memory is used before heap right? so will any heap parameters have effect on this one?

abap_heap_area_total now is 2000000000

how shall i change it ?

i have patched the DBSL library- no success

Edited by: Yavor Markov on Jan 30, 2009 11:17 AM

Answers (1)

Answers (1)

JPReyes
Active Contributor
0 Kudos

That simply happens when your select statement is too large.

I don't know what the background job is doing but most likely the variant is selecting too much data. Simply reduce the scope of the job to avoid the dump.

Regards

Juan

Former Member
0 Kudos

this is the additional language supplementation in smlt

can i increase the parameter to double just for this job and the get it back to 2G

JPReyes
Active Contributor
0 Kudos

Why not... worthy to give it a shot.

regards

Juan

Former Member
0 Kudos

btw anything wrong with these

ztta/roll_area 2000000 Byte Roll area per workprocess (total)

ztta/roll_first 1 Byte First amount of roll area used in a dialog WP

ztta/short_area 3200000 Byte Short area per workprocess

rdisp/ROLL_SHM 32768 8 kB Part of roll file in shared memory

rdisp/PG_SHM 16384 8 kB Part of paging file in shared memory

rdisp/PG_LOCAL 150 8 kB Paging buffer per workprocess

em/initial_size_MB 4095 MB Initial size of extended memory

em/blocksize_KB 1024 kB Size of one extended memory block

em/address_space_MB 512 MB Address space reserved for ext. mem. (NT only

ztta/roll_extension 2000000000 Byte Max. extended mem. per session (external mode

abap/heap_area_dia 2000000000 Byte Max. heap memory for dialog workprocesses

abap/heap_area_nondia 2000000000 Byte Max. heap memory for non-dialog workprocesses

abap/heap_area_total 2000000000 Byte Max. usable heap memory

abap/heaplimit 40000000 Byte Workprocess restart limit of heap memory

abap/use_paging 0 Paging for flat tables used (1) or not (0)

JPReyes
Active Contributor
0 Kudos

Can't really tell without knowing your hardware and system configuration.

Regards

Juan

Former Member
0 Kudos

no luck with the 4G heap

i fact when i checked the memory usage in ST02-mode list

i can see the processes doesn't occupy heap at all

no idea at all what to do

JPReyes
Active Contributor
0 Kudos

increase your extended memory and you abap/heap_area_nondia and see how it goes.

regards

Juan

Former Member
0 Kudos

Hai,

Try to reduce the file cache size.

Check the below link.....

http://help.sap.com/saphelp_nw04/helpdata/en/fb/cbb96cd9ee11d188be0000e83539c3/frameset.htm

Regards,

Yoganand.V

Former Member
0 Kudos

em/initial_size_MB - i set it from 4096 to 8192

abap_heap_non_dia from 2 to 4 G

abap_heap_total - 4 GB

still nothing

the way i see it in sm21 it is the other message that is important

No free memory available in class PERM . Memory block ID "DBSL ", 10932 bytes.

Storage request (??? bytes) failed

btw the cache size was reduced upon the installation

Former Member
0 Kudos

Hai,

Please post the contents of dev_w files, or check for any hints what really happens.

It will show the real problem.

Regards,

Yoganand.V

Former Member
0 Kudos

this is the result of the memlimits.exe

Check the maximum data size per process (malloc)

allocating 1964 MB

Maximum process heap size = 1964MB

----


Result

----


Maximum heap size per process........: 1964 MB

this value is probably limited by swap space

Total available swap space...........: 1964 MB

main memory size x 3 recommended , minimum 1 GB

swap space probably even bigger, but cannot be

totally allocated by only one process

em/initial_size_MB 8190 MB Initial size of extended memory

em/blocksize_KB 1024 kB Size of one extended memory block

em/address_space_MB 512 MB Address space reserved for ext. mem. (NT only

ztta/roll_extension 2000000000 Byte Max. extended mem. per session (external mode

abap/heap_area_dia 2000000000 Byte Max. heap memory for dialog workprocesses

abap/heap_area_nondia 4000000000 Byte Max. heap memory for non-dialog workprocesses

abap/heap_area_total 4000000000 Byte Max. usable heap memory

abap/heaplimit 40000000 Byte Workprocess restart limit of heap memory

could it be the abap/heatlimit is little?

here is the dev_w* file

B Fri Jan 30 14:57:31 2009

B table logging switched on for client '400'

B

B Fri Jan 30 14:59:43 2009

B ***LOG BXF=> table logging switched off in program RSTLAN_SUP by user YMARKOV [dbacfun#11 @ 1087] [dba

M

M Fri Jan 30 15:01:43 2009

*M **LOG ST0=> PERM DBSL 10932 [saprstg.c 5165]

**Y *** ERROR => malloc failed [saprstg.c 5169]**

*Y *** ERROR => size 10932 [saprstg.c 5170]*

*B **LOG BZ7=> memory allocation (??? bytes) failed [dbrepo#5 @ 2722] [dbrepo 2722 ]

A TH VERBOSE LEVEL FULL

Edited by: Yavor Markov on Jan 30, 2009 2:08 PM

Former Member
0 Kudos

Hai,

The value of abap/heaplimit should be smaller than abap/heap_area_dia or abap/heap_area_nondia, so that the dialog step that is running can still be executed.

Default is 40000000 (40 MB) you can increase to 80MB or 100MB, just to check whether it helps or not.

The problem is that a conflict arises in the address space between the shared memory i.e. extended memory, buffer memory, etc and the heap memory.

Try rebooting the server or upgrade your kernel patch to the latest.

Refer SAP Note:314004.

Regards,

Yoganand.V

Edited by: Yoganand Vedagiri on Jan 30, 2009 2:38 PM

Former Member
0 Kudos

i have patched before i posted the topic

and rebooting the server doesn't help

only thing i can think of is to use the workaround-

doing expert supplementation and diving all the tables in alphabetical order to let's say 3 or 4 groups

BTW one last thing

can this be due to 32 bit Windows and 4 GB of memory-

there were some things with the /3G option

anyone knows about it?

Former Member
0 Kudos

Before The PF gets about 90 % full, which means the first run was successful as the system has been freshly installed, and the process memory has been low like 100-150 MB

But when i run it at later points the disP+work in task manager reaches like 1 GB and after the job is done the process remains at 1 Gb, i suppose on the next run it crashes the limit for paging file.

the other possible solution is like stated in one of the notes to kill the WP and restart it

Anyway i have enabled the /3GB option

you might have been right about the swap

F:\usr\sap\ELP\SYS\exe>cd uc\NTI386

F:\usr\sap\ELP\SYS\exe\uc\NTI386>memlimits

SAP R/3 address space configuration test tool V3.2 (98/01/14)

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

Check the maximum data size per process (malloc)

allocating 2972 MB

Maximum process heap size = 2972MB

----


Result

----


Maximum heap size per process........: 2972 MB

this value is probably limited by swap space

Total available swap space...........: 2972 MB

main memory size x 3 recommended , minimum 1 GB

swap space probably even bigger, but cannot be

totally allocated by only one process

Now the WP can allocate one more GB of ram, so seems that's why it finishes ok

Edited by: Yavor Markov on Feb 2, 2009 9:41 AM