Clustering in OC4J 9.0.2

Given that EJB clustering in 9.0.3 is known to be broken, what's the status of it in 9.0.2?
Is there a list of know issues for 9.0.2 as there is for 9.0.3?
And does anyone have any idea when a production version of 9.0.3 is expected?
Cheers,
Nathan Messer

EJB Clustering is working in OC4J 9.0.2 production version. We are planning to release an updated version of OC4J 9.0.3 by end of this month.
regards
Debu Panda
Oracle

Similar Messages

Oracle Application Server 10.1.3.3.0 OC4J Restarting

Oracle ADF application is deployed to the AS from JDeveolper 10.1.3.3.0 and runs fine initially but after continual use the OC4J returns a runtime exception and restarts.
1. This instance is not associated with an infrastructure but it is on a server with 10.1.2.2.0 Infra and middle tier instances.
2.The error occurs while using a web app. It all works fine at first, can nav pages, view records from the db just fine. Then after say 15mins the page returns Internal Server Error message at which time I check the Oracle app server ctrl to find the error below. That error is displayed in the actual app server ctrl page.
After a few minutes I can refresh the web app page and it returns to the home page working fine again until the next exception and restart. Sometimes as a result of this issue the server will not OC4J will not restart and the page will stop working altogether. At this point I cannot ever restart the OC4J. I have to Restart the entire server machine.
An error has occurred.
Caught exception calling invoke on oc4j:name=Singleton,type=OpmnConfig
Exception thrown in RequiredModelMBean while trying to invoke operation invoke
Object reference no longer valid (disconnected)
Object reference no longer valid (disconnected)
Hide Additional Trace Information
oracle.j2ee.admin.jmx.exception.config.ConfigMBeanException:
Caught exception calling invoke on oc4j:name=Singleton,type=OpmnConfig
at oracle.oc4j.admin.management.farm.util.OpmnProxyUtil.invoke(OpmnProxyUtil.java:204)
at oracle.oc4j.admin.management.farm.util.OpmnProxyUtil.invokeSingle(OpmnProxyUtil.java:129)
at oracle.oc4j.admin.management.farm.util.OpmnProxyUtil.invokeSingle(OpmnProxyUtil.java:99)
at oracle.oc4j.admin.management.farm.mbeans.OpmnConfigAgr.getNotificationServerAttrs(OpmnConfigAgr.java:200)
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:585)
at com.sun.jmx.interceptor.DefaultMBeanServerInterceptor.invoke(DefaultMBeanServerInterceptor.java:815)
at com.sun.jmx.mbeanserver.JmxMBeanServer.invoke(JmxMBeanServer.java:784)
at oracle.oc4j.admin.jmx.server.state.LocalizationFilterMBeanServer.invoke(LocalizationFilterMBeanServer.java:579)
at oracle.oc4j.admin.management.farm.util.OpmnProxyUtil.invoke(OpmnProxyUtil.java:202) ... 50 more
Caused by: java.io.IOException: Object reference no longer valid (disconnected)
at oracle.oc4j.admin.jmx.client.RemoteMBeanServer.invoke(RemoteMBeanServer.java:820)
at oracle.oc4j.admin.management.farm.mbeans.JVMAgr.invoke(JVMAgr.java:1481)
at sun.reflect.GeneratedMethodAccessor40.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:585)
at sun.reflect.misc.Trampoline.invoke(MethodUtil.java:36)
at sun.reflect.GeneratedMethodAccessor18.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:585) at sun.reflect.misc.MethodUtil.invoke(MethodUtil.java:239)
at javax.management.modelmbean.RequiredModelMBean.invokeMethod(RequiredModelMBean.java:1071) ... 58
more Caused by: com.evermind.server.rmi.RMIConnectionException: Object reference no longer valid (disconnected)
at com.evermind.server.rmi.RemoteInvocationHandler.invoke(RemoteInvocationHandler.java:60)
at com.evermind.server.rmi.RecoverableRemoteInvocationHandler.invoke(RecoverableRemoteInvocationHandler.java:28)
at com.evermind.server.ejb.StatefulSessionRemoteInvocationHandler.invoke(StatefulSessionRemoteInvocationHandler.java:38)
at __Proxy8.invoke(Unknown Source) at oracle.oc4j.admin.jmx.client.MBeanServerEjbRemoteSynchronizer.invoke(MBeanServerEjbRemoteSynchronizer.java:144)
at oracle.oc4j.admin.jmx.client.RemoteMBeanServer.invoke(RemoteMBeanServer.java:816) ... 68 more

Greetings,
First I would recommend turning up your logging level to FINEST to glean more informatin regarding the error that is occurring. Please post these data along with any related Messaging service configuration information. It is possible that two services are attempting to make the same connection and it is causing one or the other to fail. You might also review the information in your opmn config file. Does the 10.1.2.2 instance remain running during the OC4j restart?
OPMN Configuration:
10.1.20 How to Specify Ports for State Replication in OPMN
When you deploy an application utilizing state replication in a managed Oracle Application Server, OPMN dynamically allocates the ports that are used to propagate state across the cluster. You can restrict this allocation to a range of ports for an application that has peer-to-peer replication enabled. Specifying ports for state replication might be necessary in an installation with a firewall or network that uses a well-defined port range.
To specify a range of ports for peer-to-peer state replication
Add a <port> element to an OC4J instance configuration in the opmn.xml file.
Specify the name of an application that has peer-to-peer replication enabled as the value of the id attribute of the <port> element.
Specify a range of ports in the range attribute of the <port> element.
For example, for deployment of an application named rac-web that is set up for peer-to-peer replication, the line labeled <port id=rac-web .../> in the following OC4J instance configuration tells OPMN to use ports 15213 to 15214 for state replication:
<port id="default-web-site" range="80-100" protocol="http"/>
<port id="rmi" range="12401-12500"/>
<port id="rmis" range="12701-12800"/>
<port id="jms" range="12601-12700"/>
<port id="rac-web" range="15213-15214"/>
For more information about state replication, see Chapter 9, "Application Clustering in OC4J," in the Oracle Containers for J2EE Configuration and Administration Guide.Quoted from:
http://download.oracle.com/docs/cd/B32110_01/relnotes.1013/b32199/oc4j.htm#CEGICDEF-Michael

Load Balance on OC4J

Hi,
I want to run same application on n number of servers, how can i redirect request to another server ( How can I implement Load Balancing on OC4J?), kindly, write me the proper syntext/command of loadbalancing and content of the .xml file.
Thanks!
K.G.Sharma

If you are using OC4J 1.0.2.2.x then you can follow the instructions in the Getting Started Guide at http://otn.oracle.com/docs/tech/java/oc4j/htdocs/getstart.htm#1015479 to configure loadbalancing and clustering in OC4J.
In Oracle9iAS Release 2, load balancing and clustering is performed using OHS and mod_OC4J. Please look at http://otn.oracle.com/docs/products/ias/doc_library/90200doc_otn/core.902/a92171/cluster.htm#1006053
Oracle9iAS Release 2 on Windows plaform will be released very soon within 2-3 weeks.
regards
Debu

Oc4j islands

Hello all,
when I create few more islands other than juts one default_isalnd for my oc4j , the http requests to my application are being load-balanced automatically with all the available island processes.Is this the default behavior when I create more islands within a oc4j container?
If this is the case then why do i need creating clusters of oc4j's on different machines which host the same application so that my OHS will automatically load-balance with oc4j's on different machines in a cluster topology if I have got the option of creating more islands in my oc4j in my own single server,requests will anyways load-balanced among islands with my oc4j...
Plz help..i hv lil knowledge of oc4j and its islands..:)
Edited by: anandk on Nov 19, 2009 1:15 PM
Edited by: anandk on Nov 19, 2009 1:22 PM

Thanks for correcting that Roberto - I was confusing myself between versions there ;-)
anandk:
Yes, with 10.1.2 you had to specify which OC4J instances where handling which locations (context roots) in mod_oc4j.conf, e.g. in it's simplest form it might be:
<IfModule mod_oc4j.c>
Oc4jMount /app1/* app1
</IfModule>
(see [http://download.oracle.com/docs/cd/B14099_19/web.1012/b14007/confmods.htm#CIHJBGFB] )
With 10.1.3, you can still configure the routing statically if you want, but by default OC4J automatically registers its context roots with OHS/mod_oc4j via ONS
(see [http://download.oracle.com/docs/cd/B25221_04/web.1013/b25211/confmods.htm#CIHJBGFB] ).
On the wider point about JVMs etc, typically I would have different OC4J instances for the bigger applications to give me better control (e.g. restarting the container). For your example 10 applications, maybe only 2 or 3 of them are really important so would have their own OC4J instances and the remainder would be installed together in another instance. The number of JVMs per OC4J instance depends on the demands of the application but of course you need to consider how many other JVMs are running on the same server, how much physical memory there is available, how many cores you have, etc.

Clustering app server

Is Oracle Infrastructure required for clustering OHS/OC4J app server? If not, how is the cluster setup and managed?

Sorry for the beginner questions, but I've never used Oracle before. I am evaluating porting a J2EE app that currently runs in Weblogic to 9iAS.
I have my application working in OC4J standalone, but need to support clustering. The docs say that only OC4J instances in a Farm can be clustered, but I'm not sure how to create a farm. I would like to be able to run OHS on one machine and have mod_oc4j load balance between multiple OC4J instances on multiple machines if possible. As a starting point, I've installed "Oracle9iAS J2EE and Web Cache v9.0.3".
Do I need to create a farm, or can I specify different host machines for each OC4J instance without one?

Load Balancing & Fault tolerance in Oracle Apps

Hello
I am pretty new to Oracle Apps and Oracle 9iAS
We are trying to asked to find out the ways of deplying load balancing and fault tolerance in Oracle Apps
I have gone thru the following articles of metalink and some more from google
Configuring Web Cache as a Load Balancer for Application Servers
Create new middle tier node in existing Apps 11i environment using cloning and then load balance
Integrating and using Web Cache with Forms 9i for Load Balancing
Running Multiple OC4J Instances From a Single Install of 9iAS 1.0.2.2
LOAD BALANCING ORACLE APPLICATIONS ON UNIX
Load Balancing in 11i
OC4J Clustering Setup
OC4J Load Balancing for Forms 9i
Setting up 11i E-Business suite using a hardware load balancer
Sharing an APPL_TOP in Oracle Applications 11i
In Oracle Apps Concepts pdf, found that
Load balancing occurs when there are multiple installations of web server, forms server, reports server, concurrent manager server etc
Lets consider forms server component of middle tier
Can there be multiple INSTANCES of forms server within SINGLE INSTALLATION ???
If it has to be MULTIPLE INSTALLATIONS, then it will require multiple physical machines
What is software load balancing then ???
Also read that
Oracle 9i AS instance is combition of Oracle HTTP Server (OHS) and one or more instances of Oracle9iAS Container for J2EE (OC4J)
Thus software load balancing can be implemented for Web server component of middle tier using multiple instances of OC4J ???
Forms metrix server configuration is hardware load balancing ??
What about Oracle9iAS Web Cache clustering ??
Please clarify the doubts and suggest me the way
Thanks a lot

dear all can any one help me on the following;
We want to install Oracle Apps 11.5.9 on 4 IBM AIX 5L boxes, the first node
will hold the database , concurrent and Admin Tiers. as for the other 3 nodes
it will hold the Forms/Web Tiers and we need to use forms metrics for load
balancing taking into consideration that we can't use Shared APPL_TOP. we tried
to find the way how to install and configure this solution but all the
documents are talking aboout the concept only and this concept is applicable.
We need the exact steps on how do the installation either using rapidwiz or any
thing else
fadi

First patch for the non-released 9iAS 9.0.4

Today I mentioned the first patch is released for 9ias 9.0.4 (patch 3089193). Strange that Oracle already releases patches for a product that itself has not been released yet. Is it that difficult to implement that patch into the base version? I heard that we may expect 9iAS 9.0.4 around september/october so there should be time enough.

Well -- to respond to the patches...
9.0.2.1 core patchset breaks my Jdev/BC4J application with an error indicating the BC4J tables couldn't be updated
So, I try 9.0.2.2 patch. Can't get past installation instructions (posted message yesterday about this -- no response yet).
javax.naming.CommunicationException: webdev1.softmatic.esoftmatic.com:4031. Root exception is java.lang.IllegalArgumentException: object is not an instance of declaring class
Patches have definitely not helped here.
To be constructive:
1) Looking for 9.0.4 release to allow clustering of OC4J processes even if installed with reports/forms (i.e. allow me to use EM to cluster, even if EM can't cluster the reports servers). It would be nicer to have EM cluster reports services too, but probably need smart way of updating CGICMD.DAT, for example. I could deal with just clustering the OC4J portions, but to prevent me from doing it because I've installed the "business intelligence" is frustrating because I know the capability is there -- I just can't get there because we use the reports too.
2) Looking for text/xml/configuration files on NT to use Windows CRLF (cgicmd.dat, opmn.xml, etc). Annoying, especially since I have to document the installation procedures for disaster recovery and need to include unix to dos type utilities, etc. Getting a real editor isn't my problem, it's the disaster recovery plan where you can't count on people knowing how to deal with it themselves.
3) Looking for way to deploy 9.0.3 JDeveloper BC4J to 9.0.2 safely (docs say you can't)...As a generalization, allow deployment of newer jdev/bc4j libraries with the applications. The "tightness" of BC4J within 9iAS causes issues. Development environments are updated more regularly than production -- hence the ability to fully deploy newer applications to the older app servers is key.
There are probably others... but those topics seem to not come up in the forums.
Jeff

Cluster BPEL

How can I confirm my BPEL install(s) are clustered?
I KNOW my application servers are clusterd because I see all of them through the em tool.
The reason I am asking is because when I create a domain through BPELAdmin on host 1, it is not propogated to host 2.
What I've done so far:
-set soapServerURL and soapCallbakUrl to loadbalancer URL (both hosts)
-set enableCluster to "true" and clusterName to loadbalancerURL in the collaxa-config.xml file.
-added the cluster element to bpel's orion-application.xml
-left jgroups-protocol.xml to use multicast.
-set clusterName
Background
..O/S
$ uname -a
SunOS ucs6034 5.10 Generic_127111-09 sun4u sparc SUNW,Sun-Fire-V89
..Two Hosts
host1 & host2
..Installation Type (10.1.3.0)
.Advanced Install
J2EE Server and Web Server
..NEW OC4J called oc4j_soa
Installed BPEL into both oc4j_soa's on host1 and host2

I am using 7+ different resources.
My problem is when I create a domain, the new domain doesn't show on Node 2 until a restart.
Here are a few examples of my resources:
+++++++++++++++++++++++++++++++++
Oracle® Application Server Enterprise Deployment Guide
10g Release 3 (10.1.3.1.0)
Part Number B28939-03
+++++++++++++++++++++++++++++++++
Oracle® Application Server Enterprise Deployment Guide
10g Release 3 (10.1.3.3.0)
Part Number E10294-02
+++++++++++++++++++++++++++++++++
Oracle SOA Suite High Availability FAQ and Troublshooting Guide
An Oracle White Paper
April 2008
+++++++++++++++++++++++++++++++++
Oracle® Application Server High Availability Guide
10g Release 3 (10.1.3.1.0)
Part Number B28941-03
Esp:CH 5 High Availability for Oracle SOA Suite
+++++++++++++++++++++++++++++++++
Architecting BPEL Systems for High Availability
An Oracle White Paper
October 2007
++++++++++++++++++++++++++++++++
Oracle® Containers for J2EE Configuration and Administration Guide
10g (10.1.3.1.0)
Part Number B28950-01
Esp: CH 8 Configuring and Managing Clusters and OC4J Groups
+++++++++++++++++++++++++++++++
Oracle® BPEL Process Manager Installation Guide
10g Release 3 (10.1.3.1.0) for UNIX and Microsoft Windows
Part Number B28980-03
Esp. CH 5 Oracle BPEL Process Manager Clustering
++++++++++++++++++++++++++++++

Redirecting HTTPS to HTTP in IIS 8

Hi,I have a slight problem. We have a web service running on an internal server, using HTTP. This server has a DNS name like internal.domain.com. Externally, we allow access through a reverse proxy server, that also uses HTTPS externally. This server has the external DNS name of service.domain.com. Internally (we're running split horizon DNS), service.domain.com points to internal.domain.com. Now the problem becomes that you need to use https://service.domain.com externally, and http://service.domain.com internally to access the same resource. What I would like to do is to have any incoming internal traffic to service.domain.com using HTTPS redirected to the same URL but using HTTP.I tried the following global rewriter rule in the applicationhost.config file.TextThat doesn't work for some reason.Then I tried;TextDidn't work either.What...
This topic first appeared in the Spiceworks Community

Thanks for your answer Roberto,
I have to say that i have one OC4J instance with 3 processes (virtual machines)[production environment], the problem arises when my application switches from https to http, https runs, let´s say, on process 1 and then http runs on process 3, that´s the reason why any object created on https is lost when switching to http.
I´ve checked some documentation about this subject but they talk about oc4j clusters (multiple oc4j with multiple processes within),*my case is one oc4j with multiple processes* i just need to know if there is some way to keep session between multiple virtual machines or to keep application execution on one virtual machine using https and http.
Greetings....

Clustered Oracle 9i AS and OC4J Problem

Hi,
I am working with two Oracle 9i AS (9.0.2.0.1), serverA and serverB, in a clustered environment. I have published a servlet to the cluster under the URL of /MyApp and the main servlet is MyServlet. I have noticed that when I go to either URL:
http://serverA/MyApp/MyServlet
or
http://serverB/MyApp/MyServlet
my servlet will show up fine but sometimes when going to serverA it will actually use the servlet running on serverB, and vice versa. I am able to verify this by looking at the log files. At first I wasn't sure why it would do this but it makes sense that in a clustered environment it would pick either of the servers to load balance the application. After reading some Oracle documentation it re-enforced the way the AS works.
Recently I have noticed some problems with the cluster though but can't seem to pin-point the problem. When I go to either URL it will sometimes bring up the servlet web page but other times it will give an error of:
No Response from Application Web Server
Upon further investigation I noticed that the page would come up fine any time it tried to use the servlet on serverA but it seems that when it tries to use the servlet on serverB it returns the error. The problem is I am having a hard time verifying that serverB is causing problems or if it is something with the communication between the two servers. As far as I can tell in the Oracle EM it shows both OC4J containers running fine and all the other standard containers are running as well. Some testing that I have done is turning off the OC4J container on serverB. When I do this I stop getting any errors. Checking the opmn log for the containers doesn't give any information on the errors.
Has anyone had similar problems? Might someone know how to properly test / troubleshoot the problem to see where things are messing up? Anyone have a solution?
Thanks

This is the way it works in 903 as well.
There is a more general forum for 9iAS questions on this site, where other product managers who look after other areas of the product answer questions. It might be worth your while to post your question there and one of the PMs from the management team might be able to shed some light on the issue with respect to a future release.
-steve-

WARNING: workaround for critical clustering bug in OC4J

I found a workaround for the HTTP clustering bug described in my previous posting of August 16, 2001 "Urgent: serious HTTP clustering bug in OC4J?" (http://technet.oracle.com:89/ubb/Forum99/HTML/000261.html)
The workaround:
When adding the <cluster config/> tag in orion-web.xml specify and ID (numeric) with no more than 9 figures. The autogenerated ID is longer (14 figures on my test system)
and will cause the error.
Luciano
null

burricall wrote:
..Did you find a solution?See [this post|http://forums.sun.com/thread.jspa?threadID=5426321&messageID=11001580#11001580].

Does OC4J 9.0.4 standalone support HTTP clustering?

Does OC4J 9.0.4 standalone support HTTP clustering? If so, could someone point me to the correct documentation for setting this up?

Yes the container supports it. You need to configure it manually using the XML files. The Oc4J Users Guide should document it for you. There's no load balancer capability so you need to use some 3rd party solution such as Web cache, Distributor, etc.
I can't get to the doc site right now so I can't give you the full URL to it. But have a hunt around and you should be able to find it.
cheers
-steve-

Standalone oc4j in 3 different machine with clustering enabled

Hello,
I just want to know if clustering is possible in my situation.
I have 3 different machine/server with a load balancer, I've installed a standalone oc4j in each machine, I deploy my application to each of the standalone oc4j, I have enable the clustering of each oc4j using peer to peer configuration.
Machine 1 pointing to node of Machine 2
Machine 2 pointing to node of Machine 3
Machine 3 pointing to node of Machine 1
Then I test my application, for my first try I was pointed to machine 1 by my load balancer, created a session, etc..etc.., after that I stop my application to oc4j machine 1, then after I refresh my page, the load balancer pointed me to machine 2, and there I see that my session is lost. Clustering is not working.
Can anyone guide my if clustering is possible using oc4j's only. Thanks.

I have solved this issue.
It turns out that wls 10.3 does not always delete an application cleanly.
I found 3 copies of the application remaining in the server.
I deleted and reinstalled the wls and the problem was solved.

Session not replicated in clustered oc4j instances

The application is deployed to two clustered oc4j instances(A and B) and is configured to use PEER replication protocol. After logging into the application(say A handled the request, created and stored a user object in http session), we shutdown A, and click on application link, the request is routed to B, but B shows no user object in http session is found, this happens in a servlet filter. Could somebody provide some suggestions why?

Hmm, OK. Then we need to dig in a little further.
Can you post what the orion-application.xml file looks like for these applications?
It has the <cluster> tag set in it.
You can find it in $ORACLE_HOME/j2ee/home/application-deployments/<app-name>
Are these clustered OC4J instances on the same server -- if so, did you set allow-colocation="false", which would prevent the replica from being stored on the same physical host, and thus not be avaialble when you need it?
You can enable logging for the clsuter session manager -- you can acess the logging propeties page in ASC and then set the "oracle.j2ee.cluster" to the FINEST level to start pumping some log messages into the log file.
Note: in 10.1.3.0 the settings are not persisted when made via ASC -- you have to enter them manually in the j2ee/<instance-name>/config/j2ee-logging.xml file. In 10.1.3.1 the settings are persisted automatically into this file.
The easiest way to see the log messages in the file is using the Log Viewer function in ASC as well. Look for the "Diagnostics" log in the particular Oc4J instance where the app is deployed.
cheers
-steve-

Clustering OC4J in a full installation OracleAS

I have a full OracleAS Infrastructure and 2 MiddleTier installation with OC4J,
Apache, Web Caché and Portal.
If I create an OC4J on each MiddleTier installation for my Java Application.
Can I create a cluster with the 2 OC4Js?
I'm asking this because in 9.0.2 the only way to create OC4J cluster was to
install a Java only OracleAS Instance.

If by "clustering" you mean using dcm to create a dcm-managed cluster and mantain the configuration of multiple instances syncrhonized, then the answer is NO FOR PORTAL (you can't use dcm-managed clusters with OracleAS instances of type "Portal") If by clustering you mean enabling session replication for two OC4J instances that are created inside OracleAS instances of type "Portal", then the answer is YES; you can use sesison replication inside OC4J containers that are part of a Portal installation
HTH

Clustering in OC4J 9.0.2

Similar Messages

Maybe you are looking for