NiPeek returns -5 NIETIMEOUT

Former Member · ‎03-30-2005

Sometimes when I do an orderly shutdown of my WebAS engine (from MMC, right-click J2E and choose STOP), then when I try to start it back up, the jcontrol.exe process stays yellow. When I check it's developer stack, it says "JControlMSReadMessage: NiPeek() returns -5 NIETIMEOUT.

Sometimes I can just try to start up the server a few times, and after few tries it manages to start up.

Today it seems that it will not startup. I hate to reinstall because that takes a LONG time, plus I since have installed JDI and I hate to lose all THAT install time as well.

Does anyone know what causes the NIETIMEOUT and how to avoid it?

Former Member · ‎04-04-2005

I am going to copy the cluster directory to an archive, grab a backup of the database, and then reinstall. That way I'll retain the logs et. al. for further analysis.

miker

Former Member · ‎04-01-2005

Okay, my last post has all the LOG file excerpts.

I'd love to solve this (with a little help from my forum friends), but it's just a test system and I need to get on with it. It is a Sneak Preview install (and yes I am well under the 90-day license limit) so I don't think I can call SAP Tech Support, I think the forum is my support option.

I'll check responses tomorrow morning, if that doesn't yield anything and I don't have any brainstorms then I am going to do the reinstall so I can get on with my work. Academically I'd love to keep pursuing this but I don't have another machine to jump to and preserve this one as-is.

If anyone sees something in the previous post's log excerpts by all means I'll check into it. Happy Hunting!

And Ivaylo, I didn't mean to truncate your name on the previous post, I messed up the cut&paste, sorry!

Thanks everyone,

Michael Rains

Former Member · ‎04-01-2005

aylo,

Here is what I find in the logs you asked about. I've checked all the other per the troubleshooting guide in SAP Library.

I could attach them all but this posting would be huge of course, so I am excerpting entries:

In all this detail, do notice

A)the very last of this posting has some ERRORS exceprted from the Cluster Server0 log and

B) the bootstrap has "exit hook called (rc=0)"

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

From C:\usr\sap\J2E\JC00\work\dev_bootstrap_ID1443000

1 this warning, I think because in instance.properties there is the entry "JLaunchParameters=".

My working test WebAS has this entry and the same warnings, no problem, so I think this can be ignored:

[Thr 2764] INFO: Invalid property value [JLaunchParameters/]

[Thr 2764] [Node: dispatcher bootstrap] java home is set by profile parameter

Java Home: C:\j2sdk1.4.2_06

2 The file ENDS with this entry. I have not yet shutdown the server, so this exit would seem suspect. From the SAP Library troubleshooting, if it had rc=66 instead of rc=0 then I should check jvm_bootstrap.out for the message "Illegal options for the bootstrap main class". I checked anyway and it is clean.

[Thr 3200] Thu Mar 31 15:26:41 2005

[Thr 3200] JLaunchIExitJava: exit hook is called (rc=0)

[Thr 3200] JLaunchCloseProgram: good bye (exitcode=0)

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

From C:\usr\sap\J2E\JC00\work\dev_bootstrap_ID1443050

-- has the same entries as ...ID1443000

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

From C:\usr\sap\J2E\JC00\work\dev_server0

1 The file Ends with the following; Maybe it has meaning to reveal exactly where it lasted processed and got stuck?

You can see where I issued command "DumpStackTrace" from JCMon at 3:11pm today too...

[Thr 564] Thu Mar 31 15:26:48 2005

[Thr 564] JHVM_LoadJavaVM: Java VM created OK.

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

JHVM_BuildArgumentList: main method arguments of node [server0]

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

[Thr 3852] Thu Mar 31 15:26:59 2005

[Thr 3852] JHVM_RegisterNatives: registering methods in com.sap.bc.proj.jstartup.JStartupFramework

[Thr 3852] JLaunchISetClusterId: set cluster id 1443050

[Thr 3852] Thu Mar 31 15:27:00 2005

[Thr 3852] JLaunchISetState: change state from [Initial (0)] to [Waiting for start (1)]

[Thr 3852] JLaunchISetState: change state from [Waiting for start (1)] to [Starting (2)]

[Thr 3852] Thu Mar 31 15:32:25 2005

[Thr 3852] JLaunchISetState: change state from [Starting (2)] to [Starting applications (10)]

[Thr 2116] Thu Mar 31 16:51:18 2005

[Thr 2116] JHVM_RegisterNatives: registering methods in com.sap.mw.rfc.driver.CpicDriver

[Thr 4252] Fri Apr 01 15:11:04 2005

[Thr 4252] JLaunchRequestFunc: receive command:10, argument:0 from pid:5828

[Thr 4252] JLaunchProcessCommand: caught dump stack trace

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

From C:\usr\sap\J2E\SCS01\work\dev_ms

1 This file looks clean. Here it is since it's short I'll just paste it here...

-

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

-

[Thr 5852] Thu Mar 31 15:26:25 2005

[Thr 5852] systemid 560 (PC with Windows NT)

[Thr 5852] relno 6400

[Thr 5852] patchlevel 0

[Thr 5852] patchno 34

[Thr 5852] intno 20020600

[Thr 5852] pid 5820

[Thr 5852] ***LOG Q01=> MsSInit, MSStart (Msg Server 1 5820) [msxxserv.c 1655]

[Thr 5852] MsHttpInit: full qualified hostname = PHI21811.phibred.com

[Thr 5852] HTTP logging is switch off

[Thr 5852] set HTTP state to LISTEN

[Thr 5852] MsHttpOwnDomain: own domain[1] = phibred.com

[Thr 5852] *** I listen to port 3601 (3601) ***

[Thr 5852] *** HTTP port 8101 state LISTEN ***

[Thr 5852] CUSTOMER KEY: >R2072291472<

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

I searched for all files modified recently to mine for error messages. Here is what I have found:

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

From C:\usr\sap\J2E\JC00\j2ee\cluster\dispatcher\log\system\server.0.log

1

#1.5#000F1F930C2400210000000900000C5C0003F3A238E7D912#1112302995098#/System/Server##com.sap.engine.core.cluster.impl6.ms.MSRawConnection.run

()#######Thread[Thread-36,5,SAPEngine_System_Thread[impl:6]_Group]##0#0#Error#1#com.sap.engine.core.cluster.impl6.ms.MSRawConnection#Java

###Reconnect to message server failed ().#1#Connection refused: connect# ###Encomi: failed to connect to PHI21811/3201(Connection refused: connect)# 2 #1.5#000F1F930C24001900000000000010DC0003F3A28E8DC84B#1112304432040#/System/Server##com.sap.engine.core.licensing.impl0.LicensingManagerImpl.checkLicense(String) #######Thread[Thread-14,5,main]##0#0#Info#1#com.sap.engine.core.licensing.impl0.LicensingManagerImpl#Java ###License checked successfully for software product "".#1#J2EE-Engine_ADA#

#1.5#000F1F930C24001A00000000000010DC0003F3A3BB183B55#1112309474273#/System/Server##com.sap.engine.services.jmx.MBeanServerInvoker

#######SAPEngine_System_Thread[impl:6]_50##0#0#Error#1#/System/Server#Plain###JMX connector exception occurred while processing external JMX request

[ JMX request (java) v1.0 len: 314 | src: cluster target-node: 1443000 req: invoke params-number: 4 params-bytes: 0 |

:name=com.sap.portal.prt.bridge.service.mbeans.PRTMBeanRuntime,j2eeType=PRTBridge_JMX_SECTION,SAP_J2EEClusterNode=1443000,SAP_J2EECluster="" null null null ]#

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

THIS ONE LOOKS SUSPICIOUS. a ClassNotFoundException

From: C:\usr\sap\J2E\JC00\j2ee\cluster\server0\log\system\server.0.log

1

Operation startApp on application sap.com/tc_secwssec~app finished on current cluster node for 3875 ms.#

#1.5#000F1F930C2400210000005500000C940003EF9FE7581D9B#1107894990553#/System/Server##com.sap.engine.services.jndi

######ff4752b07a1011d9af09000f1f930c24#SAPEngine_System_Thread[impl:5]_27##0#0#Info#1#com.sap.engine.services.jndi#Plain

###Path to object does not exist at java:webContainer, the whole lookup name is java:webContainer/applications/sap.com/SQLTrace.#

System/Server#sap.com/SQLTrace#com.sap.engine.services.servlets_jsp.server.runtime.context.WebComponents#Guest#2

####1d66cf517a1111d9c55a000f1f930c24#SAPEngine_Application_Thread[impl:3]_30

##0#0#Warning#1#com.sap.engine.services.servlets_jsp.server.runtime.context.WebComponents#Plain

###Error can't load Servlet com.sap.sql.ui.tabstat.TableStatisticsMonitorServlet The error is: com.sap.sql.ui.tabstat.TableStatisticsMonitorServlet#

#SAPEngine_System_Thread[impl:5]_27##0#0#Info#1#com.sap.engine.services.servlets_jsp.server.container.WebContainer#Plain

###Start of application [sap.com/SQLTrace] finished successfully in web container.#

#1.5#000F1F930C2400210000005700000C940003EF9FE76071E9#1107894991105#/System/Server##com.sap.engine.services.deploy

######ff4752b07a1011d9af09000f1f930c24#SAPEngine_System_Thread[impl:5]_27

##0#0#Warning#1#com.sap.engine.services.deploy#Plain###

Warning occurred on server 1443050 during startApp sap.com/SQLTrace : Cannot load servlet [ com.sap.sql.ui.tabstat.TableStatisticsMonitorServlet ].

Error is: java.lang.ClassNotFoundException: com.sap.sql.ui.tabstat.TableStatisticsMonitorServlet

-

Loader Info -

ClassLoader name: [sap.com/SQLTrace]

Parent loader name: [Frame ClassLoader]

References:

2 THESE ENTRIES FOLLOW LATER IN THE FILE (Searching for keyword 'error')

Operation stopApp on application sap.com/com.sap.rprof.remoteProfile finished on current cluster node for 10 ms.#

servlets_jsp.server.runtime.context.WebComponents#Plain###Error while destroing servlet < bcbsupport > The error is: JmxFrame not completely initialized, innerMBeanServer is null#

.WebContainer#######SAPEngine_System_Thread[impl:5]_38##0#0#Info#1#com.sap.engine.services.servlets_jsp.server.container.WebContainer#Plain###Stop of application [sap.com/tc_bcbici] finished successfully in web container.#

#1.5#000F1F930C24001B0000001500000C940003EFA0489BACAC#1107896622094#/System/Server##com.sap.engine.services.deploy#######SAPEngine_System_Thread[impl:5]_38##0#0#Warning#1#com.sap.engine.services.deploy#Plain###

Warning occurred on server 1443050 during stopApp sap.com/tc_bcbici : Obligatory container EJBContainer is not active at the moment.#

Warning occurred on server 1443050 during stopApp sap.com/tc_bcbici : Optional container Monitoring Configurator is not active at the moment.#

Operation stopApp over application sap.com/tc_bcbici finished successfully on server 1443050#

Operation stopApp on application sap.com/tc_bcbici finished on current cluster node for 1312 ms.#

Warning occurred on server 1443050 during stopApp sap.com/com.sap.engine.heartbeat : Obligatory container EJBContainer is not active at the moment.#

Former Member · ‎04-01-2005

Hey Michael,

In the jcmon tool, are all your processes in the "Running" status? If not what are they in?

I would look at the logs of any process that is not in the running status (i.e. default.trc file or process specific logfile (i.e. server logfile))

I suspect it's the server process hung somewhere (the server logfile will reveal where it's hung).

If you think it's the SDM process, you can disable it altogether...

Former Member · ‎03-31-2005

I really don't want to reinstall again. With EP6 and JDI et. al. it takes hours.

If I do have to reinstall, I am going to try to get a back-up afterwards so next time at least I can restore.

But surely someone out there has one more suggestion? There must be some way to determine where it is stopping at and why?

thanks,

miker

Former Member · ‎03-31-2005

I do not yet have this solved.

Somckit's response got me to looking, and I did find a great troubleshooting information in SAP Library/NetWeaver04/Java Technology/Administration/StartUpAndShutDown/Troubleshooting. I checked for all of the messages in dev_jcontrol, jvm_bootstrap.out, etc etc. In fact I grepped the whole 'work' directory for error, exit, abort.

When I compare dev_jcontrol to that of another PC that does start up, I see a subtle difference at the end. The difference is that the working WebAS has TWO sections of:

JControlMSMessageFunc: receive command: 6 argument....

Whereas on the WebAS that gets stuck on MMC jcontrol "some processes running" ends like this with just ONE set of these messages and, after 5 minutes, begins getting the NIETIMEOUT.

The last JLaunch process to start is sdm, and it does print it's "started" message; here is the tail of dev_jcontrol from my non-working system:

[Thr 800] JControlICheckProcessList: process server0 started (PID:1324)

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

JStartupStartJLaunch: program = C:\usr\sap\J2E\JC00/j2ee/os_libs/jlaunch.exe

-> arg[00] = C:\usr\sap\J2E\JC00/j2ee/os_libs/jlaunch.exe

-> arg[01] = pf=C:\usr\sap\J2E\SYS\profile\J2E_JC00_PHI21811

-> arg[02] = -DSAPINFO=J2E_00_sdm

-> arg[03] = -file=C:\usr\sap\J2E\JC00\SDM\program\config\sdm_jstartup.properties

-> arg[04] = -nodeName=sdm

-> arg[05] = -nodeId=2

-> arg[06] = -syncSem=JSTARTUP_WAIT_ON_3844

-> arg[07] = -jvmOutFile=C:\usr\sap\J2E\JC00\work\jvm_sdm.out

-> arg[08] = -stdOutFile=C:\usr\sap\J2E\JC00\work\std_sdm.out

-> arg[09] = -locOutFile=C:\usr\sap\J2E\JC00\work\dev_sdm

-> arg[10] = -mode=JCONTROL

-> arg[11] = pf=C:\usr\sap\J2E\SYS\profile\J2E_JC00_PHI21811

-> lib path = PATH=C:\j2sdk1.4.2_06\jre\bin\server;C:\j2sdk1.4.2_06\jre\bin;c:\sapdb\programs\bin;c:\sapdb\programs\pgm;C:\j2sdk1.4.2_06\bin;C:\WINDOWS\system32;C:\WINDOWS;C:\WINDOWS\System32\Wbem;w:\;w:\dll;v:\mqclient\bin;C:\Program Files\Microsoft SQL Server\80\Tools\Binn\;v:\Oracle_Client\bin;C:\usr\sap\J2E\SCS01\exe;C:\usr\sap\J2E\JC00\exe;C:\usr\sap\J2E\SYS\exe\run

-> exe path = PATH=C:\j2sdk1.4.2_06\bin;C:\usr\sap\J2E\JC00\j2ee\os_libs;c:\sapdb\programs\bin;c:\sapdb\programs\pgm;C:\j2sdk1.4.2_06\bin;C:\WINDOWS\system32;C:\WINDOWS;C:\WINDOWS\System32\Wbem;w:\;w:\dll;v:\mqclient\bin;C:\Program Files\Microsoft SQL Server\80\Tools\Binn\;v:\Oracle_Client\bin;C:\usr\sap\J2E\SCS01\exe;C:\usr\sap\J2E\JC00\exe;C:\usr\sap\J2E\SYS\exe\run

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

[Thr 800] JControlICheckProcessList: process SDM started (PID:2344)

[Thr 2400] Thu Mar 31 10:28:20 2005

[Thr 2400] JControlMSMessageFunc: receive command:6, argument:1213679940 from Message Server

[Thr 2400] Thu Mar 31 10:33:45 2005

[Thr 2400] JControlMSReadMessage: NiPeek() returns -5 NIETIMEOUT

-

I did recently install the JDI and configured per JDI Installation Guide version 1.39. I have since restarted the J2EE Engine (via MMC) without problem.

-

I also ran the JCMon utility. I do not see a whole lot of documentation on it (three pages in the Troubleshooting section of SAP Library that I mentioned earlier in this posting). Using it I did Increment Trace Level, I did Enable Debugging and Start a Debug session of my Server0 process. I did not see any more detail anywhere to help me though. I did the "Dump stacktrace" of server0, but it seemed to just display a truncated "11 : Process List". I did not find a DUMP file anywhere...?

-

My guess is that the J2EE Engine is trying to start up an application like Portal or JDI's CBS or CMS or something. But I do not see where I could monitor what it's doing at enough detail to really know where it's getting stuck.

-

Anyone have any thoughts?

thanks,

miker

Former Member · ‎03-31-2005

Jcontrol attempts to contact the message server upon startup -- that's probably what the error message JControlMSReadMessage indicates, that the M.S. and Jcontrol aren't able to communicate.

The nettimeout probably indicates that a response was not received in time by Jcontrol -- can you check the trace file for the M.S.?

NiPeek returns -5 NIETIMEOUT

Accepted Solutions (0)

Answers (7)

Answers (7)

Re: SAP Cloud ALM - Granular Authorization to Scop...

Delete Class assignment Button Not Visible

Re: SAP Analytics Cloud for planning - Inverse For...

EML action - passing all fields of a selected row ...

Re: Unable to write select query in BW transformat...