PI 7.1.Gateway died with status 2

[former_member468073]

Hello !

Dear colleagues!

I have a big Problem:

We cannot start our PI 7.1 due problems with the dispatcher:

Here a small extract from a file dev_disp:

Wed Oct 1 19:34:12 2008

rdisp/http_min_wait_dia_wp : 1 -> 1

***LOG CPS=> DpLoopInit, ICU ( 3.4 3.4 4.1) [dpxxdisp.c 1603]

***LOG Q0K=> DpMsAttach, mscon ( sadcb114) [dpxxdisp.c 12274]

DpStartStopMsg: send start message (myname is >server_SID_00 <)

DpStartStopMsg: start msg sent

CCMS uses Shared Memory Key 73 for monitoring.

CCMS: SemInMgt: Semaphore Management initialized by AlAttachShm_Doublestack.

CCMS: SemInit: Semaphore 38 initialized by AlAttachShm_Doublestack.

CCMS: AlInitGlobals : alert/use_sema_lock = TRUE.

DpJ2eeLogin: j2ee state = CONNECTED

MBUF state PREPARED

MBUF component UP

DpMBufHwIdSet: set Hardware-ID

***LOG Q1C=> DpMBufHwIdSet [dpxxmbuf.c 1074]

Wed Oct 1 19:34:17 2008

MBUF state ACTIVE

DpModState: change server state from STARTING to ACTIVE

Wed Oct 1 19:34:52 2008

• ERROR => gateway (pid 17906) died [dpxxdisp.c 16133]

• DP_FATAL_ERROR => Gateway died with status 2 - I better exit now

• DISPATCHER EMERGENCY SHUTDOWN ***

increase tracelevel of WPs

One more strangeness:

If I start the server with the parameter : rdisp/ignore_dw_dead =1 , i see in the gateway setting, that the paramater service gateway is 3300 ( service gateway = 3300 ) as usual service gateway is sapgw00 ( Or something similar)

help me please !

thanks a lot!

and :

System configuration:

PI 7.1

SAP_ABA -7.10_6

Basis - 7.10_6

PI_basis --2006_1_71

ST-PI 2005_1_71

SAP_BW 71

Kernel 117

HP-UX 11.23

Oracle 10.2.0.2

java -- 1.4.2.11

[former_member408875]

This issue also happens with me .In dev_disp and dev_rd, same error was seen where gateway died written and service already in use.

This issue was resolved after shutting down r3 (app server) ,gateway process was running (u may find from command : ps -eaf| grep <SID> ) . After kill that process and start the app server , dispatcher is up.

[former_member213031]

Check if the shutdown is clean, it looks like the gateway process is not stopped cleanly..

Use ps-ef|grep Instance after the shutdown command, and kill the process which are not stopped.

This should solve the issue.

Regards

Dhiraj Wamanacharya

[Former Member]

hello,

it looks as if the tcp ip service port is already in use

NiIBindSocket: bind (226: Address already in use)

The port 226 is not free, some other application is using

it, or it is accidently locked by the operating system.

You can use command netstat -a to see what application

/ process is using port 226 and kill that process.

If no process is holding a lock, restart the server, so the operating system is releasing the locks.

kind regards,

Andreas

[Former Member]

Hi,

This is because the gateway was not able to communicate with the kernel because of patch level or kernel mismatch.

Please follow this sapnote and implement it.

https://service.sap.com/sap/support/notes/71251

Hope this helps.

Reg, Dibya

[former_member468073]

Hello

Should I reinstall the kernel ?

thanks

[Former Member]

Get the latest kernel and apply it. That should be fine.

And, then check whether dpmon & gwmon have a same version.

Regards, Dibya

[Former Member]

Hello,

is this an NT system ? If yes, do you have any

related errors in the event log ?

Could you please attach the file dev_rd (it is

the gateway trace file which should contain

more informations).

Thanks and regards,

andreas

[former_member468073]

Hello Andreas!

this is HP-UX (11.23)

dev_rd:

( 3300 is not busy. I have not any applications or something else on this port)

-

---

trc file: "dev_rd", trc level: 1, release: "710"

-

---

Thu Oct 2 18:20:33 2008

***LOG S00=> GwInitReader, gateway started ( 23261) [gwxxrd.c 1705]

systemid 274 (HP (IA-64) with HP-UX)

relno 7100

patchlevel 0

patchno 117

intno 20020600

make single threaded, Unicode, 64 bit, optimized

pid 23261

gateway runs with dp version 95000(ext=113000) (@(#) DPLIB-INT-VERSION-094000-UC)

gateway (version=710.2008.07.31)

gw/local_addr : 0.0.0.0

CCMS: initialize CCMS Monitoring for ABAP instance with J2EE addin.

CCMS uses Shared Memory Key 73 for monitoring.

CCMS: AlInitGlobals : alert/use_sema_lock = TRUE.

CCMS: Initalized shared memory of size 60000000 for monitoring segment.

CCMS: Checking Downtime Configuration of Monitoring Segment.

CCMS: AlMsUpload called by wp 1002.

Thu Oct 2 18:20:37 2008

CCMS: AlMsUpload successful for /usr/sap/SID/DVEBMGS00/log/ALMTTREE (918 MTEs).

Thu Oct 2 18:20:38 2008

***LOG Q0I=> NiIBindSocket: bind (226: Address already in use) [nixxi.cpp 3574]

• ERROR => NiIBindSocket: SiBind failed for hdl 1/sock 10

(SI_EPORT_INUSE/226; I4; ST; 0.0.0.0:3300) [nixxi.cpp 3574]

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

*

• LOCATION SAP-Gateway on host server1 / sapgw00

• ERROR service '0.0.0.0:3300' in use

*

• TIME Thu Oct 2 18:20:38 2008

• RELEASE 710

• COMPONENT NI (network interface)

• VERSION 39

• RC -4

• MODULE nixxi.cpp

• LINE 3574

• DETAIL NiIBindSocket

• SYSTEM CALL bind

• ERRNO 226

• ERRNO TEXT Address already in use

• COUNTER 1

*

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

• ERROR => GwInitService: GwListen(sapgw00) failed (NIESERV_USED) [gwxxrd.c 8755]

***LOG S10=> GwInitReader, GwInitService () [gwxxrd.c 1913]

Thu Oct 2 18:20:39 2008

***LOG S0V=> GwStopGateway, gateway stopped () [gwxxrd.c 12436]

thanks!