Re: OEM Upgrade from 10.2.0.1 to 10.2.0.5

From: Maxim Demenko <mdemenko_at_gmail.com>
Date: Thu, 11 Jun 2009 19:59:41 +0200
Message-ID: <4A31460D.6050301_at_gmail.com>

gaelle.simonchristien_at_gmail.com schrieb:

> On 20 mai, 19:27, Maxim Demenko <mdeme..._at_gmail.com> wrote:

>> Simon schrieb:
>>
>>
>>
>>> Hi all,
>>> I am doing a fresh install of Oracle Enterprise manager on
>>> Linux 32 bit using OLE5.3
>>> using
>>> How to Install Grid Control 10.2.0.5.0 on Enterprise Linux 5 Using
>>> the
>>> Existing Database (11g) Option
>>> Doc ID: 784963.1
>>> As per the Metalink note I am doing software only install of 10.2.0.1
>>> and then 10.2.0.5 with no config.
>>> Then I manually run the follwoing
>>> ConfigureGC.pl script: where I get this error
>>> .../oms10g/perl/bin/perl .../oms10g/sysman/install/ConfigureGC.pl
>>> /u01/app/gc
>>> Any ideas/suggestions to solve this ? This is not related to sys/
>>> sysman password I think
>>> Oc4jDeploy tool completed, but with errors.
>>> '
>>> INFO: oracle.sysman.top.oms:PerformSecureCommand:runCmd:Command
>>> Output
>>> stderr:
>>> 'java.lang.reflect.InvocationTargetException
>>> at sun.reflect.NativeMethodAccessorImpl.invoke0(Native
>>> Method)
>>> at sun.reflect.NativeMethodAccessorImpl.invoke
>>> (NativeMethodAccessorImpl.java:39)
>>> at sun.reflect.DelegatingMethodAccessorImpl.invoke
>>> (DelegatingMethodAccessorImpl.java:25)
>>> at java.lang.reflect.Method.invoke(Method.java:324)
>>> at oracle.j2ee.tools.deploy.Oc4jDeploy.doDeploy(Unknown
>>> Source)
>>> at oracle.j2ee.tools.deploy.Oc4jDeploy.execute(Unknown
>>> Source)
>>> at oracle.j2ee.tools.deploy.Oc4jDeploy.deploy(Unknown Source)
>>> at oracle.j2ee.tools.deploy.Oc4jDeploy.main(Unknown Source)
>>> Caused by: java.lang.NoClassDefFoundError
>>> at
>>> com.evermind.server.rmi.RMIConnection.EXCEPTION_ORIGINATES_FROM_THE_REMOTE_­
>>> SERVER
>>> (RMIConnection.java:1634)
>>> at com.evermind.server.rmi.RMIConnection.invokeMethod
>>> (RMIConnection.java:1585)
>>> at com.evermind.server.rmi.RemoteInvocationHandler.invoke
>>> (RemoteInvocationHandler.java:55)
>>> at
>>> com.evermind.server.rmi.RecoverableRemoteInvocationHandler.invoke
>>> (RecoverableRemoteInvocationHandler.java:23)
>>> at __Proxy0.bindWebApp(Unknown Source)
>>> at com.evermind.client.orion.BindWebAppCommand.execute
>>> (BindWebAppCommand.java:28)
>>> at com.evermind.client.orion.Oc4jAdminConsole.executeCommand
>>> (Oc4jAdminConsole.java:139)
>>> ... 8 more
>>> Died at /u01/app/oracle/product/10.2.0/oms10g/sysman/install/
>>> emdeploy.pl line 154.
>>> '
>>> INFO: oracle.sysman.top.oms:OmsPlugIn:Requested Configuration Step 1
>>> have been completed with status=false
>>> INFO: oracle.sysman.top.oms:EmcpPlug:invoke:Completed EmcpPlug invoke
>>> method on an aggregate=oracle.sysman.top.oms for Action=configuration
>>> in step=2:microstep=0
>>> INFO: oracle.sysman.top.oms:The plug-in OMS Configuration has failed
>>> its perform method
>>> INFO: Cfm.save() was called
>>> INFO: Cfm.save(): 15 aggregate instances saved
>> Simon, i ran into the same issue, which in my case was easily
>> workarounded by installing libXp ( it was a minimal cent os installation
>> with no X related packages at all). There should be a Metalink Note
>> about required packages for installation on RHEL 5/OEL 5, however i
>> could not access it - slightly more informative in this regard is
>> instruction published by Jeff Hunter (he mentions the libXp library as
>> well).
>> I've spent a couple of hours trying to resolve the problem with invalid
>> username/password by creation of repository but to no avail - without
>> comprehensive understanding how OUI works (which i obviously don't
>> have), this is too time consuming task - i could see with strace the
>> passwords for sysdba and sysman sent to emrepmgr.pl, but it is unclear,
>> why "default" is used instead of specified in response file.
>> In the original SR it is stated, that the problem is in the base
>> 10.2.0.1 bundle, but i have certain doubts on this, because the 10.2.0.4
>> silent install doesn't have this problem ( of course with 10.2.0.4
>> repository database).
>> If you still have the problem with your installation despite
>> installation of libXp, i'm sure, it is some library missing - in that
>> case you can contact me offline, so i can sent you the rpm list on my
>> system and used response files if you are interested.
>>
>> Best regards
>>
>> Maxim
>>
>> --
>> Why make things difficult, when it is possible to make them cryptic
>> and totally illogical, with just a little bit more effort?
>>
>> Aksel Peter Jørgensen

> 
> 
> Maxim,
> The same issue happened to me while upgrading Grid Control 10.2.0.3 to
> 10.2.0.5.
> I used a existing database 11g for this install, I'm on a RHEL 5.3.
> I followed Jeff Hunter's recommandations and I have the libXp rpm
> installed.
> Still, the ConfigureGC.pl script failed with
> EXCEPTION_ORIGINATES_FROM_THE_REMOTE_­­
> SERVER.
> I checked the environment variables ORACLE_BASE, ORACLE_HOME and
> ORACLE_SID, everything's ok.
> I don't know what to do anymore to solve the problem.
> Do you (or anybody else) have any idea to run this script succesfully?
> Regards,
> 
> Gaelle

Gaelle, as far as i understood from metalink,google etc., the mentioned error is really generic one and seems to mask the real reason of the problem. As i stated before, in my case it was the missing library ( to verify this issue, we can compare the list of installed packages on my system - and yours), but it can be caused as well by lack of memory and inability of jvm to start new threads ( see Note. 413220.1 - here they speaks about hpux, but it may happen on other platform as well, i think). There may be dozens of other reasons as well (you can google on it) and probably, the only reliable way to recover from this error is - to log a SR and try to escalate it.

Best regards

Maxim

-- 
Why make things difficult, when it is possible to make them cryptic
and totally illogical, with just a little bit more effort?

Aksel Peter Jørgensen

Received on Thu Jun 11 2009 - 12:59:41 CDT