cancel
Showing results for 
Search instead for 
Did you mean: 

SUBMOD_SWITCH_OLDREL/SETUPGPAR_UPG_DOWN RFC Logon Failure

Former Member
0 Kudos

Hello Friends,

I am performing a NW BI 7.0  ABAP [ Java stack is already split ] Upgrade to  NW 7.40 using the SUM Tool.

The Upgrade has crossed most of the major hurdles and is now in execution phase [downtime phase] and we are now facing the following error RFC_LOGON_FAILURE

System Details

OS : Windows 2008 R2 server

DB: Oracle 11.2.3

NW 7.0  >>  NW 7.4  Upgrade running   [ Source system BI 7.0 ABAP stack after Dual stack split ]

During the Execution phase, we are getting the RFC_LOGON_FAILURE  (error in license check)  error in the SUM Tool.

I also see that the SAP message_server process is stopped  and not starting up in MMC. The process of determining the correct license requires the SAP instance to obtain a hardware ID from the message server. The error message indicates that this could not be obtained  , this is confirmed in the system logs [SAP BASIS system : Msg Server hardware ID missing ( HWID_PENDING)  ] , due to which the SAP is giving License check errors. 

I already checked via saplikey  we have the right license keys in place, to be sure I also deleted and reimported the permanent License keys again ,so its is definitely not related to  the license problem.

When I check the dev_disp logs, I can see the below errors

==================================================

Mon Mar 09 05:18:02 2015

*** WARNING => DpNetCheck: NiAddrToHost(1.0.0.0) took 5 seconds

***LOG GZZ=> 1 possible network problems detected - check tracefile and adjust the DNS settings [dpxxtool2.c  6525]

MtxInit: 30000 0 0

DpSysAdmExtInit: ABAP is active

DpSysAdmExtInit: VMC (JAVA VM in WP) is not active

display_tcp_ip: 0

Mon Mar 09 10:48:37 2015

***LOG Q0I=> NiPConnect2: 10.1.101.159:3900: connect (10061: WSAECONNREFUSED: Connection refused) [nixxi.cpp 3283]

*** ERROR => NiPConnect2: SiPeekPendConn failed for hdl 24/sock 504

    (SI_ECONN_REFUSE/10061; I4; ST; 10.1.101.159:3900) [nixxi.cpp    3283]

*** ERROR => MsIAttachEx: NiBufConnect to orsapbisbx01/3900 failed (rc=NIECONN_REFUSED) [msxxi.c      718]

The dev_w*   logs

=================

M  ***LOG Q16=> ThGetMsHwId, DpMBufHwIdGet (HWID_PENDING) [thxxmbuf.c   474]

M  *** ERROR => SlicLikeyGetHardwareKey: no hwid from MSG server: -1 [sliclikey.c  1299]

M  {root-id=00221982BAFF1ED4B1CACB253DF01767}_{conn-id=00000000000000000000000000000000}_0

M  *** ERROR => 11 wlikey_check_webas: Error getting the hardware key. You might find some information in file "dev_ms" on the message server. [sliclikey.c  1318]

M  {root-id=00221982BAFF1ED4B1CACB253DF01767}_{conn-id=00000000000000000000000000000000}_0

M  ***LOG Q16=> ThGetMsHwId, DpMBufHwIdGet (HWID_PENDING) [thxxmbuf.c   474]

M  *** ERROR => SlicKCheckLicense: did not get hwid from message server [slicch.c     323]

M  {root-id=00221982BAFF1ED4B1CACB253DF01767}_{conn-id=00000000000000000000000000000000}_0

The dev_ms  logs

==================

trc file: "dev_ms", trc level: 1, release: "721"

---------------------------------------------------

[Thr 7028] ***LOG Q01=> MsSInit, MSStart (Msg Server 1 3184) [msxxserv.c   2305]

[Thr 7028] load acl file = D:\usr\sap\BD2\SYS\global\ms_acl_info.DAT

[Thr 7028] MsGetOwnIpAddr: my host addresses are :

[Thr 7028]   1 : [10.1.101.159] orsapbisbx01.radisys.com (HOSTNAME)

[Thr 7028]   2 : [127.0.0.1] orsapbisbx01.radisys.com (LOCALHOST)

[Thr 7028] MsHttpInit: full qualified hostname = orsapbisbx01.radisys.com

[Thr 7028] HTTP logging is switch off

[Thr 7028] MsHttpOwnDomain: own domain[1] = radisys.com

[Thr 7028] *** I listen to port 3600 (3600) ***

[Thr 7028] *** I listen to internal port 3900 (3900) ***

[Thr 7028] CUSTOMER KEY: >P1199735526<

[Thr 7028] build version=721.2014.07.17

[Thr 6880] Mon Mar 09 04:28:25 2015

[Thr 6880] MsSExit: received SIGINT (2)

[Thr 6880] ***LOG Q02=> MsSHalt, MSStop (Msg Server 3184) [msxxserv.c   7804]

Could this be a compatibility issue of the old msg_server and the new 740 Kernel.

How do i get my message server & the SAP system fully up and running in order to be able to finish the Exchange process.

Any suggestions are welcome!!!

Thank you!

Prashant

Accepted Solutions (1)

Accepted Solutions (1)

manumohandas82
Active Contributor
0 Kudos

Hi Prashanth ,

[Thr 7148] CUSTOMER KEY: >P1199735526<

[Thr 7148] build version=721.2014.07.17

Are you in 721 version of the kernel ( Source system ) ?

Can you paste the disp+work from /usr/sap/<SID><Instance>/exe

Thanks ,

Manu

Former Member
0 Kudos

Hi Prashant,

AS suggested by manu do verify the kernel version under the exe directory.

Also as you mentioned earlier you have updated the kernel.

so do verify the kernel binaries under /usr/sap/SID/<Inst Nr>/exe & /sapmnt/SID/exe, /usr/sap/SID/SYS/exe/run.

And you should have all the version matching and should be 742.

You can run the sapcpe to sync the binaries in all locations.

Let us know the update.

Regards,

Ram

Former Member
0 Kudos

Hello Manu / Ram,

I finally found the problem, since our host name was too long , we had to shorten our profile names for the SUM tool to accept them..

But looks like in this process, we also shortened the host name inside the START profile for the parameter  _MS [msg_server], so no matter what other things we tried, the message server would not start  up , the trace file was not initialized , which led us to read other logs, but the as I was checking the sapstart logs, noticed that the profile name was incomplete.

This led to the message server start up failure. Once I identified this and rectified this, the message server came up.

Thank you for all your help and inputs and for boosting my morale when it was down.

Cheers

Prashant

Answers (4)

Answers (4)

manumohandas82
Active Contributor
0 Kudos

Hi Prashanth ,

Stop the SUM tool .

Shut down all the SAP system and processes . ( Reboot the server )

Manually switch the kernel to the latest available for 740  and try once again   ( Backup the current kernel and profiles )

Thanks ,

Manu

Former Member
0 Kudos

Hi Manu,

We tried this, stopped the SUM tool & stopped SAP system and restarted the server.

Then manually switched the kernel to the latest 740 Kernel which was already running in the shadow Instance 742_REL [after backing up the kernel & profiles]

The message server started up for a brief few seconds ( this happens with old Kernel as well after a reboot)  and then shutdown .Message server doesnt start up on subsequent attempts [as the message server only starts once.

Our SAP system is still in the same state and without the Message sever process running, our upgrade is still stuck.

Regards

Prashant.

Former Member
0 Kudos

Hi Prashant,

Can you please paste the dev_ms log in detail here. So we all can have a look.

Also Are your sapstartsrv services running correctly, As SUM tool uses sap control to get all the required instance information.

Do restart them once and try to start the system using sapcontrol.

Regards,

Ram

Former Member
0 Kudos

Hi Ram,

Please find the dev_ms log below

---------------------------------------------------

trc file: "dev_ms", trc level: 1, release: "721"

---------------------------------------------------

[Thr 7148] Mon Mar 09 11:43:58 2015

[Thr 7148] ms/http_max_clients = 500 -> 500

[Thr 7148] MsSSetTrcLog: trc logging active, max size = 52428800 bytes

systemid   562 (PC with Windows NT)

relno      7210

patchlevel 0

patchno    327

intno      20020600

make       multithreaded, Unicode, 64 bit, optimized

pid        2728

[Thr 7148] ***LOG Q01=> MsSInit, MSStart (Msg Server 1 2728) [msxxserv.c   2305]

[Thr 7148] load acl file = D:\usr\sap\BD2\SYS\global\ms_acl_info.DAT

[Thr 7148] MsGetOwnIpAddr: my host addresses are :

[Thr 7148]   1 : [10.1.101.159] orsapbisbx01.radisys.com (HOSTNAME)

[Thr 7148]   2 : [127.0.0.1] orsapbisbx01.radisys.com (LOCALHOST)

[Thr 7148] MsHttpInit: full qualified hostname = orsapbisbx01.radisys.com

[Thr 7148] HTTP logging is switch off

[Thr 7148] MsHttpOwnDomain: own domain[1] = radisys.com

[Thr 7148] *** I listen to port 3600 (3600) ***

[Thr 7148] *** I listen to internal port 3901 (3901) ***

[Thr 7148] CUSTOMER KEY: >P1199735526<

[Thr 7148] build version=721.2014.07.17

[Thr 3936] Mon Mar 09 11:50:53 2015

[Thr 3936] MsSExit: received SIGINT (2)

[Thr 3936] ***LOG Q02=> MsSHalt, MSStop (Msg Server 2728) [msxxserv.c   7804]

Also I really tried to restart , delete and re register the sapstartsrv services from the MMC as I suspected that the message server was affected due to the start trigger not recieved correctly , but this didn't solve the problem.

Regards

Prashant

Former Member
0 Kudos

Looks like its not taking your license.

Follow these steps from OS level and install your license and then it will start.

Also make sure your DVBMSG work folder is not empty. It should have links. If no links - recreate them.

Follow this to install your license from OS level.

Install SAP License from OS Level (SAPLIKEY)&lt;/title&gt;&lt;script type=&quot;text/javascript&quo...

Former Member
0 Kudos

Hi Manu  / Bevin ,

When I saw that the HW_Key was not being recognized by the message , I too deleted the permanent license via saplikey and re-applied the permanent license at the OS level and the system confirmed that permanent license has been applied , but still login wasnt possible , which confirmed that the problem is not with the license , but rather the message server which have got changed during the SUM tool ugprade.

Kindly keep in mind that we are already in the Execution/ downtime phase " SUBMOD_SWITCH_OLDREL/SETUPGPAR_UPG " of  the NW 7.0 to NW 7.4 Upgrade, and I read a note somewhere which forbids to change kernel executables manually during the Execution phase [ consistency reasons].

some Background info -

It looks like the Message server is unable to connect ever since the SUM tool performed the Kernel Switch during the Execution phase [ Main system was brought down  , once earlier , during the downtime  after which it came up during the STARTSAP_TRANS  phase and also carried some steps and later got stuck in current phase.

Regards

Prashant

manumohandas82
Active Contributor
0 Kudos

Hi Prashanth ,

Seems your issue is similar to the one detailed in the following note

1562079 - Temporary license overwrites valid permanent license


Try patching your "current" kernel to the latest available and try once again


Thanks ,

Manu

manumohandas82
Active Contributor
0 Kudos

Hi ,

Seems the message server has not restarted properly

Mon Mar 09 10:48:37 2015

***LOG Q0I=> NiPConnect2: 10.1.101.159:3900: connect (10061: WSAECONNREFUSED: Connection refused) [nixxi.cpp 3283]

*** ERROR => NiPConnect2: SiPeekPendConn failed for hdl 24/sock 504

    (SI_ECONN_REFUSE/10061; I4; ST; 10.1.101.159:3900) [nixxi.cpp    3283]

*** ERROR => MsIAttachEx: NiBufConnect to orsapbisbx01/3900 failed (rc=NIE

Please do a complete restart of the server . Check whether any process is occupying the port 3900 ,

After that start sap and try once again

Thanks ,

Manu

Former Member
0 Kudos

Hi Manu,

I had already tried this earlier and tried it again, stopped everything and took a clean restart of the server i.e. OS + DB + SAP.

The message server is still down and not starting up with same errors.

The SAP system was running till before the Downtime stage without problems.

Any suggestions would be appreciated.

Thanks

Prashant