center instance failed

[Former Member]

Dear Experts:

my sap system occured a lot of error and all user operation failed, so i have to restart it urgently, following are some messages in trace log file, please kindly help to investigate it, thanks!



OS: Windows Server 2003 SP1

DB: MSS 2005(9.0.3042)

SAP: ECC 5.0

Kernel: 6.40 unicode 221

in the dev_disp:

Wed Jan 14 09:07:28 2009

NiPWrite: ENOBUFS: buffers reduced to 4/4096 (errNo=10055; cnt=1; len=7208; hdl 28 / socket 824)

NiPWrite: ENOBUFS: buffers reduced to 4/1024 (errNo=10055; cnt=2; len=7208; hdl 28 / socket 824)

NiPWrite: ENOBUFS: buffers reduced to 0/4096 (errNo=10055; cnt=1; len=6184; hdl 28 / socket 824)

in the syslog:

09:11:51 DIA 003 300 GO01 ZFIR BY4 Database error 0 at FET access to table DDFTX

09:11:51 DIA 003 300 GO01 ZFIR BY0 > [10054] TCP Provider: An existing connection was f

09:11:51 DIA 003 300 GO01 ZFIR BY0 > closed by the remote host.#[10054] Communication l

09:11:51 DIA 003 300 GO01 ZFIR BY0 > failure#[-1] Session Provider: Physical connection

09:11:51 DIA 003 300 GO01 ZFIR BY0 > usable [xFFFFFFFF].#[-1] Communication link failur

09:11:51 DIA 003 300 GO01 ZFIR BY0 > Session Provider: Physical connection is not usabl

09:11:51 DIA 003 300 GO01 ZFIR BY0 > [xFFFFFFFF].#[-1] Communication link failure#[0] P

09:11:51 DIA 003 300 GO01 ZFIR BY0 > error in TDS stream

09:11:51 DIA 003 300 GO01 ZFIR D01 Transaction Canceled 00 951 ( 0 )

09:11:51 DIA 003 300 GO01 ZFIR BZY Unexpected return value 8 when calling up DbSlR

09:11:51 DIA 003 300 GO01 ZFIR BYJ Function ROLLBACK on connection R/3 failed

09:11:51 DIA 003 300 GO01 ZFIR R39 Error in DB rollback/SyFlush, return code 016384

09:11:51 DIA 003 300 GO01 ZFIR Q02 Stop Workproc 3, PID 23580

09:11:51 RD Q0I Operating system call recv failed (error no. 10054)

09:12:02 WRK 000 Q0Q Start Workproc 3, Pid 25920

In the dev_w3 dialog work process:

C Wed Jan 14 09:11:51 2009

C GetNextRows: line 20630. hr: 0x80004005 Protocol error in TDS stream

C sloledb.cpp [GetNextRows,line 20630]: Error/Message: (err 0, sev 0), Protocol error in TDS stream

C Procname: [##Y3LAX05SESr3p00000059800000000020094911]

C sloledb.cpp [GetNextRows,line 20630]: Error/Message: (err -1, sev 0), Communication link failure

C Procname: [##Y3LAX05SESr3p00000059800000000020094911]

C sloledb.cpp [GetNextRows,line 20630]: Error/Message: (err -1, sev 0), Session Provider: Physical connection is not usable [xFFFFFFFF].

C Procname: [##Y3LAX05SESr3p00000059800000000020094911]

C sloledb.cpp [GetNextRows,line 20630]: Error/Message: (err -1, sev 0), Communication link failure

C Procname: [##Y3LAX05SESr3p00000059800000000020094911]

C sloledb.cpp [GetNextRows,line 20630]: Error/Message: (err -1, sev 0), Session Provider: Physical connection is not usable [xFFFFFFFF].

C Procname: [##Y3LAX05SESr3p00000059800000000020094911]

C sloledb.cpp [GetNextRows,line 20630]: Error/Message: (err 10054, sev 0), Communication link failure

C Procname: [##Y3LAX05SESr3p00000059800000000020094911]

C sloledb.cpp [GetNextRows,line 20630]: Error/Message: (err 10054, sev 0), TCP Provider: An existing connection was forcibly closed by the remote host.

C Procname: [##Y3LAX05SESr3p00000059800000000020094911]

C DbSlExeRead - Error 99 (dbcode 10054) on get_next_row

C 1 times error (0,0) in sequence

B ***LOG BY4=> sql error 0 performing FET on table DDFTX [dbtran#10 @ 7292] [dbtran 7292 ]

B ***LOG BY0=> [10054] TCP Provider: An existing connection was forcibly closed by the remote host.

Kind Regards

Alex Cheng

[markus_doehr2 · Accepted solution]

It seems that your database has a problem.

Please use SQl Server Enterprise Manager and check if the database is corrupted.

Markus

[kurt_harding2]

Looking in SM21, I see the following:

A database operation returned a return code that indicates
that the work process is no longer connected to the database, or that
this connection was broken for a while.

This error can occur, for example, if the database was shut
down, but not the SAP System. Network problems can also cause the
connection between the application server and the database server to be
broken.

For initial information on the cause of the error, see the
database error text.

The work process with the return code has the status
Reconnect. It attempts to reconnect itself to the database status. If it
reconnects successfully, it can start sending database requests to the
database instance again. An appropriate message is written in the
syslog.

If the work process cannot reconnect itself, it remains in
Reconnect status and attempts to reconnect itself to the database instance until

it is successful.

After talking to my network guys, I get the following answer: problems reported this morning appear to be connected to an issue with one of the IBM Blade Center chassis.  They have
found that resets have occurred on one of the I/O modules.

Good luck!

[kurt_harding2]

has there been a solution found for this issue yet? I just received the same error in a BW 7.3 environment that was just upgraded from 7.0.

[Former Member]

Hello,,

i guess you have a problem with your kernel,

you will have upgrade kernel level,

[Former Member]

Anyone who knows the solution to this problem!! Help me please!! Do you mind to share?

After restarting the system this error disappear... I still didn´t figured out why this error appears sometimes and makes the SAP system stops!

　

This system is very similar to yours:

SAP System: R/3 4.7x200

Database: SQL Server 2000 SP2

Operation system: Windows 2003 Server (X32)

RAM Server = 16 Gb

[Former Member]

Hi,

Check latest "ERRORLOG" file which will give you correct information regarding database functionality.

You can find the log in the following location.

"Drive:\Program Files\Microsoft SQL Server\MSSQL.1\MSSQL\LOG"

--Kishore

[Former Member]

Hi, Kishore

Thanks for your prompt, I have checked the lastest "ERRORLOG" file ,but there is no exceptional message in that time, and the db issue disappear after a R/3 system restart.

In this system the DB is standalone, The exception accompanied by message "NiPWrite: ENOBUFS: buffers reduced to 4/4096 (errNo=10055; cnt=1; len=7208; hdl 28 / socket 824)" , is it possible the network result in this issue?

If you suffer similar issue, please give more prompt and clew, thank you!

[joo_migueldimas]

Hello Alex Cheng,

I know this post I know is old, but I have a system with exactly problem!!

This system is very similar to yours:

SAP System: ECC 5.0

Database: SQL Server 2000 SP2

Operation system: Windows 2003 Server (X32)

RAM Server = 3,5 Gb

Sometimes the system stops and when I check the dev_disp log I see in there that error:

...

Wed Jul 14 18:12:46 2010
NiPWrite: ENOBUFS: buffers reduced to 4/4096 (errNo=10055; cnt=1; len=12651; hdl 24 / socket 1132)
NiPWrite: ENOBUFS: buffers reduced to 4/1024 (errNo=10055; cnt=2; len=12651; hdl 24 / socket 1132)
NiPWrite: ENOBUFS: buffers reduced to 4/512 (errNo=10055; cnt=3; len=12651; hdl 24 / socket 1132)
NiPWrite: ENOBUFS: buffers reduced to 4/128 (errNo=10055; cnt=4; len=12651; hdl 24 / socket 1132)
NiPWrite: ENOBUFS: buffers reduced to 4/32 (errNo=10055; cnt=5; len=12651; hdl 24 / socket 1132)
*** ERROR => DpTmSend: NiBufSend failed(rc=-6)->disconnect tm: 18 [dpxxdisp.c   10981]
RM-T18, U4612, 300     JRIBEIRO, JORGE-PC, 18:12:46, M0, W0,     , 2/0

...

After restarting the system this error disappear... I still didn´t figured out why this error appears sometimes and makes the SAP system stops! You found any solution for that problem.... Could you remember???

Tell me anything please... or someone who knows the solution to this problem!! Help me please!!

Best regards,

João Dimas - Portugal