cancel
Showing results for 
Search instead for 
Did you mean: 

Start_shadow_first fails, shadow system missing a hardware key

Former Member
0 Kudos

Hello All,

I have a longer posting on the enhancement package forum, but for iseries, there is no better place to ask than here.

I am using the EHPI installer. The source system had the maintenance certificate expire sometime during or before the shadow system was created. In the start_shadow_first step, the process fails with the rfc log on. I can log in the shadow system as SAP* and install the licenses but the hardware key is missing. SAP* is not allowed to run the "determine hardware key" function. No other user can log on to do it.

I believe that this is the reason I am stuck. This is my first test run so I expect alot of issues.

Thanks,

Pat

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Hi Pat,

for sure, you do need a license, but I never had this issue !

Especially as the maintenance certificate is not of interest during EHPI AND the "real license" is missing on your site ...

But: As you will have both instances on the same host, you could apply the same license to the shadow ...

BUT: SAP* IS authorized to get the hw key and install licenses ... it is just not allowed to delete old licenses ...

=> if you do not see a HW key, then the message server might have issues ... most likely an issue in your tcp/ip config or in this case I would start to hunt at WRKSRVTBLE ... there the MSG-Srv entry for the shadow needs to be unique ! (I thing it is called sapmsSIDSHD or similar)

Regards

Volker Gueldenpfennig, consolut international ag

http://www.consolut.net http://www.4soi.de http://www.easymarketplace.de

Former Member
0 Kudos

Thanks for the reply Volker,

We noticed that the shadow message server starts and ends immediately. The service table and other settings seem to be correct. Several logs are referring to the message server not connecting but I cannot find where it is wrong.

The dev_ms log is as follows

trc file: "dev_ms", trc level: 1, release: "701"

[Thr 01] Thu Aug 5 08:33:02 2010

[Thr 01] MsSOsPrivInit

iSeries with OS400) relno 7010 patchlevel 0 patchno 63

ASCII, 64 bit, optimized pid 110612 [Thr 01]

***LOG Q01=> MsSInit, MSStart (Msg Server 1 110612)

[msxxserv_mt. 1853] [Thr 01] SigISetDefaultAction :

default handling for signal 20

***LOG Q0I=> NiIBindSocket: bind (67: Address already in use) [nixxi.cpp 3239] [Thr 01] *** ERROR => NiIBindSocket: SiBind failed for hdl 0 / sock 5 (SI_EPORT_INUSE/67; I4; ST; 0.0.0.0:3614) [nixxi.cpp 3239]

[Thr 01] *** ERROR => MsSCommInit: NiBufListen(sapmsTEH) (rc=NIESERV_USED) [msxxserv_mt. 10222] [Thr 01] *** ERROR =>

MsSInit: MsSCommInit [msxxserv_mt. 1888]

[Thr 01] *** ERROR => main: MsSInit [msxxserv_mt. 5983] [Thr 01] ***LOG Q02=> MsSHalt, MSStop (Msg Server 110612) [msxxserv_mt. 6030]

System running without ICM - check rdisp/start_icman [dpxxdisp.c 13279] and this:

***LOG Q0I=> NiPConnect: connect (79: Connection refused) [nixxi.cpp 2519] *** ERROR => NiPConnect: SiConnect failed for hdl 2 / sock 9 (SI_ECONN_REFUSE/79; I4; ST; 10.127.0.100:3615) [nixxi.cpp 2519] *** ERROR => MsIAttachEx: NiBufConnect to omni/3615 failed (rc=NIECONN_REFUSED) [msxxi.c 652] ***LOG Q0L=> DpLoopInit, nomscon ()

SysLog file.. /usr/sap/TEH/EHPI/abap/exe/sapmscsa: finished. *** *** MSG_SERVER died => main: MsSInit ***and syslog Initialization complete Failed to send a request to the message server Run-time error "LOAD_PROGRAM_NOT_FOUND" occurred > Short dump "100805 083441 omni_TEH_15 SAPSYS " created Msg server hardware ID missing (HWID_PENDING)

The load program error has to do with note 1382020 a spool process error.

My next step will be to ipl the system but this has to wait till the weekend.

I am using note 525677 qp2term to start and stop the shadow system , I assume that everything for that is in place. If the ipl does not fix the problem, I start to clean out the upgrade and start again.

Regards, Pat

Former Member
0 Kudos

Hi Pat,

in my eyes, the following is the issue:

(67: Address already in use)

There is somewhere a conflict with the ports, that one is blocked ... I would guess, that it might be used twice from either both of the SAP instances or even from one of them. I would check 80xx, 81xx, 36xx & 39xx ports ...

Regards

Volker Gueldenpfennig, consolut international ag

http://www.consolut.net http://www.4soi.de http://www.easymarketplace.de

Answers (1)

Answers (1)

Former Member
0 Kudos

Hi All,

As I start the process from the begining I found that the server I am using for this test did not have QUSRWRK subsystem started, therefore no SAPSTRSRV was running. This may have been part of the problem.

On another note: Service Marketplace "Notes Search" has stopped providing the side panel restriction for the os400 as an operating system, and the database is missing too.

These limits were very helpful when the number of results are large.