Startup of Instance failed SAP ECC 5.0

[Former Member]

Hi Gurus,

When I starting the SAP ECC 5.0 i am getting the Startup of instance is failed message. But i can able to login and work in the system.

I have 2 instance in the same system another one instance started successfully without any issues.

Below is the log file (startsap_DVEBMGS10.log)

(31059) Waiting for Child Processes to terminate.

(31059) **** 2008/07/26 12:30:00 Child 31072 terminated with Status 2 . ****

(31059) **** 2008/07/26 12:30:00 Child 31071 terminated with Status 2 . ****

(31059) **** 2008/07/26 12:30:00 Child 31069 terminated with Status 0 . ****

(31059) **** 2008/07/26 12:30:01 Child 31070 terminated with Status 1 . ****

(31059) **** No more Child Processes to wait for.

(31059) Parent Shutdown at 2008/07/26 12:30:01

Execute Post-Shutdown Commands

-

---

(31059) Exiting with Return-Code 3. (No more child processes)

Startup of Instance failed

Please guide me.

Thanks

Siva

[JPReyes · Accepted solution]

Check dev_disp and dev_w0 of the instance that is having problems.... post the errors here.

Regards

Juan

[Former Member · Accepted solution]

Dear Juan

Thank you so much.

Please find below the log for both dev_disp and dev_w0.

help me to sort out the issue.

-

---

trc file: "dev_disp", trc level: 1, release: "640"

-

---

Sat Jul 26 12:30:01 2008

kernel runs with dp version 128(ext=102) (@(#) DPLIB-INT-VERSION-128)

length of sys_adm_ext is 312 bytes

systemid 387 (Intel x86 with Linux)

relno 6400

patchlevel 0

patchno 43

intno 20020600

make: single threaded, ASCII

pid 31070

***LOG Q00=> DpSapEnvInit, DPStart (10 31070) [dpxxdisp.c 1100]

shared lib "dw_xml.so" version 43 successfully loaded

shared lib "dw_xtc.so" version 43 successfully loaded

shared lib "dw_stl.so" version 43 successfully loaded

shared lib "dw_gui.so" version 43 successfully loaded

MtxInit: -2 0 0

• ERROR => DpSysAdmExtCreate: ShmCreate (CREATE) [dpxxtool2.c 520]

• ERROR => dispatcher already running ???? [dpxxtool2.c 521]

• ERROR => I better EXIT before I do any (more) damage [dpxxtool2.c 523]

"dev_disp" 25L, 952C 25,1 Bot

-

---

trc file: "dev_w0", trc level: 1, release: "640"

-

---

*

• ACTIVE TRACE LEVEL 1

• ACTIVE TRACE COMPONENTS all, M

*

M

M Sat Jul 26 13:10:33 2008

M TRACE FILE TRUNCATED (pid = 14208 )

M systemid 387 (Intel x86 with Linux)

M relno 6400

M patchlevel 0

M patchno 43

M intno 20020600

M make: single threaded, ASCII

M pid 14208

M

M

M Sat Jul 26 14:24:48 2008

M hostaddrlist return 0

A

Sat Jul 26 14:24:49 2008

A *** ERROR => RFC ======> Could not open file rfcexec.sec

[abrfcio.c 7569]

M hostaddrlist return 0

[Former Member · Accepted solution]

Hi

Check the memory configuration and network connectivity.

Increase the shared memory

Can u mention the hardware details

Regards

Bhaskar

Edited by: bhaskar1818 on Jul 29, 2008 10:12 AM

[former_member1351727]

Hi Siva,

Try starting Oracle from SQLplus.

It is giving the error message that it is unable to determine the shared m,emory size menas it is not able to read the configuration file initSID.ora.

Try see the file is readable to the user with which you are trying to start the DB.

After the start if you need top check the log or anyerrors have a look at the alertSID.log .

Thanks.

[Former Member]

Hi Everybody,

Thanks to everyone.

I have changed file permission for ORACLE_HOME/bin/oracle to 6751

After that I am getting ORA-27123 because Shared Memory Segments and Semaphore Arrays as locked due to tried to start the SAP with out permission.

After that i have restarted the machine and problem resolved. Also you can user ipcrm command to delete the invalid entry in Shared memory segment.

Thanks.

Regards,

Siva.

[Former Member]

Hi,

ERROR => DpSysAdmExtCreate: ShmCreate (CREATE) http://dpxxtool2.c 520

ERROR => dispatcher already running ???? http://dpxxtool2.c 521

ERROR => I better EXIT before I do any (more) damage http://dpxxtool2.c 523

The error means that after the system was stopped some disp+work processes were still running.That is why the system while starting throws the error dispatcher already running.These are called zombie processes.

To solve this I would recommend the following:

1)For Linux/Unix(from the error posted by you I made out that your system is installed on a linux server)

Stop the system again.

Check for all the processes related to system using ps-ef | grep -i dw or ps -ef | grep -i <SID>

Kill any processes related to R3 which might be running.check if all the processes are killed.

Run cleanipc command.

Start the system.It should come up.

2)For Windows servers it is very simple.

Stop the R3 from MMC,once the R3 is stopped open the task manager and look for all the Processes related sidadm.You can kill them all there.And Start the system.It should come up fine.

Plz revert for any clarifications.

Regards,

Abhishek

[Former Member]

Hello,

"Dispatcher already running".

Please check if there are any zombies running.

Regards,Prasanna

[JPReyes]

Hi Kulkarni.

"Dispatcher already running".
Please check if there are any zombies running.

That info was already posted and in much more detail... avoid posting the same info again!

Regards

Juan

[Former Member]

I have killed the process and restared my machine and when i am trying to startsap

I am getting startdb: termininating with error code12.

Kinldy help me.

[Former Member]

Also when I am tring to start the DB (Oracle 9.2) via BRTOOLS

i am getting ORA-27121 unable to determine size of the shared memory segment

Linux error :13 permission denined.

Regards,

Siva

[Former Member]

Are you try to start DB from SQLPLUS? Are your listener up and running? Try execute "R3trans -d" under sidadm user and check trans.log

Regards.