cancel
Showing results for 
Search instead for 
Did you mean: 

PXA_NO_SHARED_MEMORY

Former Member
0 Kudos

Several dialog, update, spool and update2 processes are continuesly stopped and cannot restart on SAP R/3 Enterprise / Windows NT (32 bit) / ORACLE 9.2.0.6.0 machine.

The user is always SAPSYS and receive the following messages:

DEV_W2:

X *** ERROR => EM context 'U3/M0' already exists (deleting old context).

A

A PXA

A PXA INITIALIZATION

A

A Tue Jan 08 15:00:38 2008

A System page size: 4kb, total admin_size: 28436kb, dir_size: 14152kb.

I *** ERROR => CreateFileMapping(6,800000 KB) failed with Err=1455

ERROR_COMMITMENT_LIMIT: The paging file is too small for this operation to complete.

I *** ERROR => ShmCreate: Create (6,819200000,1) failed

*****************************************************

  • Memory diagnostic *

*****************************************************

Systeminformation

Processor-Typ : Intel unknown

Processor-Count : 4

Operating System : NT 5.2, Build 3790

Service Pack : Service Pack 1

NT Pagefile Informations

Config. minimum size : 10477568 K

Config. maximum size : 10477568 K

Avail. maximum size : 10477568 K

Num

Pagefile

Min.Size

Max.Size

Avail.Max

Curr.Size

1

C:pagefile.sys

2097152 K

2097152 K

2097152 K

2097152 K

2

D:pagefile2pagefile.sys

4190208 K

4190208 K

4190208 K

4190208 K

3

D:pagefile3pagefile.sys

4190208 K

4190208 K

4190208 K

4190208 K

NT Task Manager Informations

Total Handles : 27104

Total Threads : 1168

Total Processes : 109

Commit Charge Total : 18127448 K

Commit Charge Limit : 18449052 K

Commit Charge Peak : 18443932 K

Phys.Memory Total : 16604440 K

Phys.Memory Available : 13747944 K

File Cache : 145832 K

Kernel Memory Total : 104304 K

Kernel Memory Paged : 62552 K

Kernel Memory Nonpaged : 41752 K

Memory usage of current process

Total virt.address space : 3145600 K

Avail.virt.address space : 795180 K

Private Pages : 0 K

Total heap size : 16413 K

Virtual memory regions : 0 K

Uncommitted heap memory : 7104 K

Allocated heap memory : 8095 K

Moveable heap memory : 0 K

DDE shared heap memory : 0 K

A Error Code PXA_NO_SHARED_MEMORY.

A Module $Id: //bas/640_REL/src/krn/runt/abload3.c#43 $ SAP.

A Function ab_pbinit Line 1754.

A ** RABAX: level LEV_RX_STDERR completed.

A ** RABAX: level LEV_RX_RFC_ERROR entered.

A ** RABAX: level LEV_RX_RFC_ERROR completed.

A ** RABAX: level LEV_RX_RFC_CLOSE entered.

A ** RABAX: level LEV_RX_RFC_CLOSE completed.

A ** RABAX: level LEV_RX_IMC_ERROR entered.

A ** RABAX: level LEV_RX_IMC_ERROR completed.

A ** RABAX: level LEV_RX_DATASET_CLOSE entered.

A ** RABAX: level LEV_RX_DATASET_CLOSE completed.

A ** RABAX: level LEV_RX_RESET_SHMLOCKS entered.

A ** RABAX: level LEV_RX_RESET_SHMLOCKS failed.

A ** RABAX: level LEV_RX_ERROR_SAVE entered.

A ** RABAX: level LEV_RX_ERROR_SAVE completed.

A ** RABAX: level LEV_RX_ERROR_TPDA entered.

A ** RABAX: level LEV_RX_ERROR_TPDA completed.

A ** RABAX: level LEV_RX_PXA_RELEASE_RUDI entered.

A ** RABAX: level LEV_RX_PXA_RELEASE_RUDI completed.

A ** RABAX: level LEV_RX_LIVE_CACHE_CLEANUP entered.

A ** RABAX: level LEV_RX_LIVE_CACHE_CLEANUP completed.

A ** RABAX: level LEV_RX_END entered.

A ** RABAX: level LEV_RX_END completed.

A ** RABAX: end RX_SUICIDE

A Shared Memory von 800000 Kilo-Bytes für PXA nicht verfügbar..

A

A *

A * Non recoverable error. Aborting process...

A *

M ***LOG R63=> ThWpHalt, halt wp ()

M in_ThErrHandle: 1

M *** ERROR => ThWpHalt: halt wp (step 1, th_errno 5, action 2, level 1)

M *****************************************************************************

M *

M * LOCATION SAP-Server ask_QAS_01 on host feronia (wp 2)

M * ERROR ThWpHalt: halt wp

M *

M * TIME Tue Jan 08 15:00:39 2008

M * RELEASE 640

M * COMPONENT Taskhandler

M * VERSION 1

M * RC 5

M * MODULE thxxhead.c

M * LINE 9806

M * COUNTER 1

M *

M *****************************************************************************

M

M PfStatDisconnect: disconnect statistics

M ThIErrHandle: action changed to 2

M Entering TH_CALLHOOKS

M ThCallHooks: call hook >SAP-Trace buffer write< for event BEFORE_DUMP

M TrThHookFunc: called for WP dump

M ThCallHooks: call hook >ThrSaveSPAFields< for event BEFORE_DUMP

M *** ERROR => ThrSaveSPAFields: no valid thr_wpadm

M *** ERROR => ThCallHooks: event handler ThrSaveSPAFields for event BEFORE_DUMP failed

M Entering ThSetStatError

M ThErrPrivHandle: set th_errno (5)

M Entering ThReadDetachMode

M *** ERROR => ThIErrHandle: bad value for th_act_em_hdl (11), detach T2/M0

DEV_DISP

Mon Jan 07 08:00:52 2008

      • ERROR => W4 (pid 6156) died

      • ERROR => W5 (pid 7756) died

      • ERROR => W7 (pid 7588) died

      • ERROR => W8 (pid 7500) died

      • ERROR => W10 (pid 6148) died

      • ERROR => W18 (pid 6932) died

      • ERROR => W19 (pid 1540) died

      • ERROR => W34 (pid 5296) died

Mon Jan 07 08:05:52 2008

      • ERROR => W4 (pid 5120) died

      • ERROR => W5 (pid 7964) died

      • ERROR => W7 (pid 5684) died

      • ERROR => W8 (pid 4872) died

      • ERROR => W10 (pid 6584) died

      • ERROR => W18 (pid 3248) died

      • ERROR => W19 (pid 4680) died

      • ERROR => W34 (pid 664) died

Mon Jan 07 08:10:52 2008

      • ERROR => W4 (pid 7360) died

      • ERROR => W5 (pid 5944) died

Adjusted the profile parameter abap/buffersize from original value 600000 to 800000, but no difference and also restarting the server does not solve the issue.

Any ideas ?

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

The value of the parameter rdisp/PG_MAXFS is increase to 262144 as explained in attached note 133909.

Added the value /USERVA=2800 to the boot.ini as recommended in Note 990538 "WSAENOBUFS error on Windows" and not /Userva=3030 as mentioned in http://support.microsoft.com/kb/810371/en-us, because no exchange server.

Rebooted SAP and windows server and in Task Manager the following values.

Before reboot:

Commit Charge Limit

18449052

Commit Charge Peak

18443932

After reboot:

Commit Charge Limit

26829460

Commit Charge Peak

18705576

Page file on server is 2048 MB and not increased.

PHYS_MEMSIZE

9008

em/initial_size_MB

13100

em/address_space_MB

512

em/blocksize_KB

4096

abap/buffersize

800000

After reboot no error messages, will monitor the system today if it will occur again.

former_member433984
Active Contributor
0 Kudos

that is it

> Before reboot:

> Commit Charge Limit

> 18449052

> Commit Charge Peak

> 18443932

>

> After reboot:

> Commit Charge Limit

> 26829460

> Commit Charge Peak

> 18705576

>

before reboot even you stopped the system probably there left some processes holding SHM and occupying the page file space. After reboot the space was freed. So far you do not need to change any SAP memory parameter.

Answers (4)

Answers (4)

Former Member
0 Kudos

Hello,

When we checked the paging memory, which was more than that was recommended at the OS.

We increased the abap/buffersize parameter, but no success.

We had nothing but last option to restart the server, restarting the server helped resolve this issue for me.

Regards,

Omkar

Former Member
0 Kudos

Monitored for a day and issue is solved.

Also found more details about the cause of the issue.

The pagefile settings were no longer good after a restore.

On 😧 drive is normally a total of about 10 Gb pagefiles. There has been a restore, where this setting was NOT restored:

The pagefile folders were not in the backupselectie. Misconceptions therefore wrong backup configuration.

The pagefiles are now well established and the backupselection been corrected.

Thanks Markus and Yaroslav for your quick response and answers.

Gidion

former_member433984
Active Contributor
0 Kudos

what about this

I *** ERROR => CreateOsShm CreateFileMapping(6,800000 KB) failed with Err=1455

ERROR_COMMITMENT_LIMIT: The paging file is too small for this operation to complete. http://shmnt.c 2130

I *** ERROR => ShmCreate: Create (6,819200000,1) failed http://shmnt.c 469

did you try to increase the page file?

What you see in Task Manager for

Commit Charge Limit

Commit Charge Peak

if SAP system is stopped (and started)?

markus_doehr2
Active Contributor
0 Kudos

You are getting this error because the system started in emergency mode.

how big are you other buffers? You may reach the 2 GB border. Do you have "/3GB /USEVA:2600" in your boot.ini?

Markus

Former Member
0 Kudos

Here's my boot.ini file:

[boot loader]

timeout=30

default=multi(0)disk(0)rdisk(0)partition(1)\WINDOWS

[operating systems]

multi(0)disk(0)rdisk(0)partition(1)\WINDOWS="Windows Server 2003, Enterprise" /noexecute=optout /3GB /fastdetect

Got Microsoft Windows Server 2003 Enterprise Edition Service Pack 1 and 15,8 GB of RAM Physical Address Extension

Gidion

markus_doehr2
Active Contributor
0 Kudos

Ok. Looks good.

How big are your other buffers? You may have run into problem described on 129813 - Windows: Problems due to address space fragmentation

Markus