SAP - shutdown

[Former Member]

Hi,

We had our Production System went down suddenly. When I analyzed the log details. I found the following. I went manually using SIDADM and started the SAP. Now it is working. But i would like to know what had caused the sudden interruption. Can someone help us?

Mon Mar 1 04:01:17 2010

***LOG Q0E=> SigIGenAction, signal ( 33) [sigux.c 1067]

DpHalt: shutdown server >DBSERVER_UP1_00 < (normal)

DpModState: change server state from ACTIVE to SHUTDOWN

Stop work processes

Mon Mar 1 04:01:20 2010

Stop gateway

• ERROR => DpProcKill: kill failed (3: No such process) [dpuxdisp.c 404]

Stop icman

Terminate gui connections

wait for end of work processes

waiting for termination of work processes ...

Mon Mar 1 04:01:23 2010

waiting for termination of work processes ...

Mon Mar 1 04:01:24 2010

Mon Mar 1 04:01:27 2010

waiting for termination of work processes ...

Mon Mar 1 04:01:28 2010

waiting for termination of work processes ...

DpStartStopMsg: send stop message (myname is >DBSERVER_UP1_00

<)

***LOG Q0I=> NiIWrite: writev (32: Broken pipe) [nixxi.cpp 3948]

• ERROR => NiIWrite: SiSendV failed for hdl 3 / sock 12

(SI_ECONN_BROKEN/32; UD; ST; 192.168.102.244:3900) [nixxi.cpp 3948]

• ERROR => MsINiWrite: NiBufSend (rc=NIECONN_BROKEN) [msxxi.c 2480]

• ERROR => MsISnd2: MsINiWrite (rc=NIECONN_BROKEN) [msxxi.c 1438]

MBUF state OFF

MBUF component DOWN

• ERROR => DpStartStopMsg: MsSnd2Type failed (rc=-100) [dpxxdisp.c 14285]

• ERROR => MsIReceive: not_attached [msxxi.c 1201]

• ERROR => MsISnd2: not_attached [msxxi.c 1321]

• ERROR => DpHalt: MsSendKeepalive failed (-3) [dpxxdisp.c 10943]

detach from message server

cleanup EM

cleanup event management

cleanup shared memory/semaphores

Mon Mar 1 04:01:59 2010

removing request queue

***LOG Q05=> DpHalt, DPStop ( 618560) [dpxxdisp.c 11257]

• shutdown completed - server stopped ***

=> Stack Trace:

DpHalt() at 0x101300f34

SigIGenAction() at 0x1000c1f04

Unknown at 0x44a4

!!! Invalid stack address encountered: 0x0

!!! Stack corrupted.

[Former Member]

Increase the swap memory at os level. It was set very less at os level prior to this

[Former Member]

Hi Raj, what did you do to increase the swap memory?

[markus_doehr2]

> Mon Mar 1 04:01:17 2010

> ***LOG Q0E=> SigIGenAction, signal ( 33) [sigux.c 1067]

> DpHalt: shutdown server >DBSERVER_UP1_00 < (normal)

> DpModState: change server state from ACTIVE to SHUTDOWN

> Stop work processes

Someone issued a "normal shutdown" of the instance.

Markus

[Former Member]

Hi ,

Is it possible to findout via OS or SAP to trace out the history of activity done towards this server.

OS is : IBM AIX

Database : Oracle with ECC 6.0

thanks

raj

[Former Member]

Hi,

please check the alert log of the server so that you can find whether the database is shutdowned because of manual command or because of some other oracle errors. If you found any oracle errors (EX:ORA-XXX) search with that error in the marketplace.

Also check the dev_disp log it will give you the detailed information why the system went down.

Thanks,

Chaitanya.

[Former Member]

analyzed and problem got resolved. It was due to swap size which led to the issue