Code synchronization across the cluster

Do the App Servers take care of synchronizing webapp java code across multiple nodes of a cluster.
or does synchronizing work only on a single JVM
Thanks,
Sunil.

When you talk of synchronization do you mean more than one thread accessing mutable object state or synchronizing serialized session data per node such as data stored in the HTTP session?
For the thread synchronization which is what I think you mean, a thread on machine A running on JVM 1 cannot access an object running on machine B under JVM 2 so there is no need to worry about synchronization.
In the second instance, there is nothing concrete in the Java EE spec therefore it is up to the vendor, but typically this is handled.

Similar Messages

Synchronization of memberLeaving event across the cluster

Hi,
Let's assume that we have 2 nodes and one of them is leaving the cluster (but will continue to operate alone). Both nodes use a MemberListener to react to the memberLeaving(MemberEvent) event. Moreover, in our case as part of the leaving event logic we need to send a lot of messages from one node to the other node. That is why we need to have both nodes be part of the cluster until the leaving event is done in all nodes.
So my question is: Will be node leaving the cluster wait to actually leave the cluster once all the other nodes (and itself too) finished processing the memberLeaving event?
In case the answer is no I was thinking that the way to ensure a proper clean up would be to do all the clean up logic in the node leaving the cluster but that would imply bringing all the shared/cached data to the leaving node and that wouldn't be a scalable operation.
Ideas are welcome. :)
Thanks,
-- Gato

Hi Gato,
The memberLeaving event is an asynchronous notification raised on a node B when another node A issues a programmatic Service.shutdown() or Cluster.shutdown() call. There is no way for code running on B to block or delay the shutdown process on the node A.
If you want to "negotiate" the conditions of a clean shutdown between cluster nodes, you would need to use a "satellite" InvocationService to communicate all necessary information and actions across the cluster before initiating the shutdown sequence.
Regards,
Gene

Will the Application Scope be shared across the cluster in a multi-node OC4

Hi,
I have the following requirement:
Users of the application can only have single (browser) session. When a user who already has a session connects again, he should no longer be allowed to access the older session.
My proposed implementation is:
- After successful login – possibly using a Session Listener - an entry is made in a HashMap UserSessions that lives in the application scope. Key is the username, value is the session id (HttpSession.getId()).
- For every request, using a ServletFilter, we check whether the session is still in the UserSessions HashMap for the current user. If a new session has been created for the same user, the session id for that new session is in the UserSessions map and the servletfilter will not find the session. In that case, the filter should invalidate the session and forward to the user to an error page.
However, the application will run on a multi-node OC4J cluster. I am starting to wonder:
Will the Application Scope be shared across the cluster in a multi-node OC4J environment?
I know session state can be shared. But what application state/scope?
Does anyone know? Do I have to do anything special in the cluster or the application to get this to work?
Thanks for your help.
Lucas

gday Lucas --
Application scope is not replicated across JVM boundaries with OC4J.
I'm sure this used to be described in the doc, but I can't find it now from a quick scan.
If you wanted to use this type of pattern, you could look to use a Coherence cache as distribution mechanism to share objects across multiple JVMs/nodes.
-steve-

Common memory place across the cluster nodes

Hi All,
I am a websphere application server v6.1 user. I am running an application that uses a HashMap to store common information in the form of key value pairs. The application works fine in a single server environment but the same application fails in a cluster environment. This happens because the HashMap information will not be available for the cluster environment nodes which were running on a different JVM�s.
Could anybody suggest a good design where in I can use a common place to store the HashMap information like queue, database or any common memory area which is available across the cluster nodes? I am not really familiar with the memory facilities offered by websphere server. (The use of a central database is the worst case I prefer as the application makes several calls to the database resulting in a deadlock and 100% CPU utilization)
And also the values to the HashMap were added dynamically so the memory place should allow me to add my values dynamically during the runtime.
Please suggest is there any other way or any links to refer to achieve the above situation.
Thanks in advance
-Sandeep
Message was edited by:
km-sandeep

For a similar scenario we maintain a version flag in the DB based on which we would reload the hashmap.I'm too interested in finding out a design without DB.

Code hints across the timeline

Hi all,
If I define a variable in the first frame of my flash timeline (strongly typed).
     var strMyString:String = "My String";
then stop the playhead on the 10th frame and access my variable in actionscript, I lose code hints.
     stop();
     strMyString.
Is it possible to have code hints available across the whole timeline or are they limited to the frame in which the variable was declared?
Many thanks..
Simon.

You have to follow naming procedure for variables for proper coding hint.
In frame 1 itself if you type strMyString. nothing happens.
For a check I have typed strMyString_txt. in frame 10 then code hint appeared.
I hope I have succeed in explaining what I meant for proper naming convention.

Keeping the toplink cache insync across a cluster

We would like to deploy our application across a cluster of 10g application servers and we are using toplink as our persistence framework using POJO's. I have been looking around and I have been unable to find anything very helpful on how to setup toplink to be able to keep the cache in sync across the cluster. Has anyone does this on 10g with POJO's? If so can you please point me to an example or a good piece of documentation?
Thanks
Marc

The documentation on Cache synch can be found starting on page 8-3 on the 10g (9.0.4) release.

The Cluster not failover when i shutdown one managed server?

Hello, I created one cluster whit two managed servers, and deployed an application across the cluster, but the weblogic server gave me two url and two different port for access to this application.
http://server1:7003/App_name
http://server1:7005/App_name
When I shutdown immediate one managed server i lost the connection whit the application from this managed server, My question is, the failover and de load balancer not work, why??
Why two diferent address?
thank any help

Well you have two different addresses (URL) because those are two physical managed servers. By creating a cluster you are not automatically going to have a virtual address (URL) that will load balance requests for that application between those two managed servers.
If you want one URL to access this application, you will have to have some kind of web server in front of your WebLogic. You can install and configure Oracle HTTP Server to route requests to WebLogic cluster. Refer this:
http://download.oracle.com/docs/cd/E12839_01/web.1111/e10144/intro_ohs.htm#i1008837
And this for details on how to configure mod_wl_ohs to route requests from OHS to WLS:
http://download.oracle.com/docs/cd/E12839_01/web.1111/e10144/under_mods.htm#BABGCGHJ
Hope this helps.
Thanks
Shail

Managed server not able to join the cluster

Hi
I have two storage node enabled coherence servers on two different machines.These two are able to form the cluster without any problem. I also have two Managed servers. When I start one, will join the cluster without any issue but when I start the fourth one which does not join the cluster. Only one Managed server joins the cluster. I am getting the following error.
2011-12-22 15:39:26.940/356.798 Oracle Coherence GE 3.6.0.4 <Info> (thread=[ACTIVE] ExecuteThread: '2' for queue: 'weblogic.kernel.Default (self-tuning)', member=n/a): Loaded cache configuration from "file:/u02/oracle/admin/atddomain/atdcluster/ATD/config/atd-client-cache-config.xml"
2011-12-22 15:39:26.943/356.801 Oracle Coherence GE 3.6.0.4 <D4> (thread=[ACTIVE] ExecuteThread: '2' for queue: 'weblogic.kernel.Default (self-tuning)', member=n/a): TCMP bound to /172.23.34.91:8190 using SystemSocketProvider
2011-12-22 15:39:57.909/387.767 Oracle Coherence GE 3.6.0.4 <Warning> (thread=Cluster, member=n/a): This Member(Id=0, Timestamp=2011-12-22 15:39:26.944, Address=172.23.34.91:8190, MachineId=39242, Location=site:dev.icd,machine:appsoad2-web2,process:24613, Role=WeblogicServer) has been attempting to join the cluster at address 231.1.1.50:7777 with TTL 4 for 30 seconds without success; this could indicate a mis-configured TTL value, or it may simply be the result of a busy cluster or active failover.
2011-12-22 15:39:57.909/387.767 Oracle Coherence GE 3.6.0.4 <Warning> (thread=Cluster, member=n/a): Received a discovery message that indicates the presence of an existing cluster:
Message "NewMemberAnnounceWait"
FromMember=Member(Id=2, Timestamp=2011-12-22 15:22:56.607, Address=172.23.34.74:8090, MachineId=39242, Location=site:dev.icd,machine:appsoad4,process:23937,member:CoherenceServer2, Role=WeblogicWeblogicCacheServer)
FromMessageId=0
Internal=false
MessagePartCount=1
PendingCount=0
MessageType=9
ToPollId=0
Poll=null
Packets
[000]=Broadcast{PacketType=0x0DDF00D2, ToId=0, FromId=2, Direction=Incoming, ReceivedMillis=15:39:57.909, MessageType=9, ServiceId=0, MessagePartCount=1, MessagePartIndex=0, Body=0}
Service=ClusterService{Name=Cluster, State=(SERVICE_STARTED, STATE_ANNOUNCE), Id=0, Version=3.6}
ToMemberSet=null
NotifySent=false
ToMember=Member(Id=0, Timestamp=2011-12-22 15:39:26.944, Address=172.23.34.91:8190, MachineId=39242, Location=site:dev.icd,machine:appsoad2-web2,process:24613, Role=WeblogicServer)
SeniorMember=Member(Id=1, Timestamp=2011-12-22 15:22:53.032, Address=172.23.34.73:8090, MachineId=39241, Location=site:dev.icd,machine:appsoad3,process:19339,member:CoherenceServer1, Role=WeblogicWeblogicCacheServer)
2011-12-22 15:40:02.915/392.773 Oracle Coherence GE 3.6.0.4 <Warning> (thread=Cluster, member=n/a): Received a discovery message that indicates the presence of an existing cluster:
Message "NewMemberAnnounceWait"
FromMember=Member(Id=2, Timestamp=2011-12-22 15:22:56.607, Address=172.23.34.74:8090, MachineId=39242, Location=site:dev.icd,machine:appsoad4,process:23937,member:CoherenceServer2, Role=WeblogicWeblogicCacheServer)
FromMessageId=0
Internal=false
MessagePartCount=1
PendingCount=0
MessageType=9
ToPollId=0
Poll=null
Packets
{

Hi,
By default Coherence uses a multicast protocol to discover other nodes when forming a cluster. Since you are having difficulties in establishing a cluster via multicast, Can you please perform a multicast test and see if multicast is configured properly.
http://wiki.tangosol.com/display/COH32UG/Multicast+Test
Hope you are using same configuration files across the cluster members and all members of the cluster must specify the same cluster name in order to be allowed to join the cluster.
<cluster-name system-property="tangosol.coherence.cluster";>xxx</cluster-name>
I would suggest, try using the unicast-listener with the well-known-addresses instead of muticast-listener.
http://wiki.tangosol.com/display/COH32UG/well-known-addresses
Add similar entries like below in your tangosol override xml..
<well-known-addresses>
<socket-address id="1">
<address> 172.23.34.91<;/address>
<port>8190</port>
</socket-address>
<socket-address id="2">
<address> 172.23.34.74<;/address>
<port> 8090</port>
</socket-address>
</well-known-addresses>
This list is used by all other nodes to find their way into the cluster without the use of multicast, thus at least one well known node must be running for other nodes to be able to join.
Hope this helps!!
Thanks,
Ashok.
<div id="isChromeWebToolbarDiv" style="display:none"></div>

How is index data managed within the cluster?

Hello
We are in the process of evaluating Coherence.
With regard to indexes, am I correct in thinking that the data for the index is stored in an (internally managed) cache on the same node as the attendant indexed values? I'm not interested in getting at the index data, but rather just understanding how the index data itself is treated/stored within the cluster.
So in a 2 node cluster where an index is applied to a distributed cache that only has 2 entries, will there be an index cache on each node that has 1 {Binary,key} entry where the key points to the value that is stored on that same node? That is, I trust that the index data is spread out across the cluster in the case of a partitioned topology?
Regards
Peter

user11279467 wrote:
Hello
We are in the process of evaluating Coherence.
With regard to indexes, am I correct in thinking that the data for the index is stored in an (internally managed) cache on the same node as the attendant indexed values? I'm not interested in getting at the index data, but rather just understanding how the index data itself is treated/stored within the cluster.
So in a 2 node cluster where an index is applied to a distributed cache that only has 2 entries, will there be an index cache on each node that has 1 {Binary,key} entry where the key points to the value that is stored on that same node? That is, I trust that the index data is spread out across the cluster in the case of a partitioned topology?
Regards
PeterHi Peter,
up to 3.4.x indexes were supported only on partitioned topology, and yes, indexes on each node contain only data corresponding to entries which have their primary copy on that node (referred to as local entries from now on).
The indexes are not stored in a cache but they are managed as part of the information Coherence maintains about the backing map (the map which contains local entries).
The indexes are made of 2 parts:
- forward index: this is a mapping from the cache key in an internal representation (backing map keys from now on) to the value extracted with the extractor used to create the index from the entry belonging to the cache key (extracted value from now on)
- reverse index (aka. reverse map): this is a mapping from the extracted value to a set of backing map keys of entries from which the reverse index key was extracted with the extractor used to create the index
The index can be sorted which means that the reverse index will be a SortedMap (so sorting order is the ordering of extracted values)
Let's see an example. Assuming you have the following cache content within a particular storage-enabled node :
A --> Object (getX=5 ,...)
B --> Object (getA=3 ,...)
C --> Object (getA=3 ,...)
D --> Object (getA=4 ,...)
The forward index in that node will contain:
Binary(A) --> 5
Binary(B) --> 3
Binary(C) --> 3
Binary(D) --> 4
The reverse index will contain:
3 --> { Binary(B), Binary(C) }
4 --> { Binary(D) }
5 --> { Binary(A) }
Where Binary(...) means the internal (binary) representation of the ... object.
If the index is not ordered, then the order of iteration on entries in the reverse index are not deterministic.
If the index is ordered, then the iteration of entries in the reverse index will be as shown above. Also, you can cast the reverse index to SortedMap so you have the very useful headMap and tailMap and firstKey and lastKey methods.
Hope this helps.
Best regards,
Robert

The Cluster Service function call 'ClusterResourceControl' failed with error code '1008(An attempt was made to reference a token that does not exist.)' while verifying the file path. Verify that your failover cluster is configured properly.

I am experiencing this error with one of our cluster environment. Can anyone help me in this issue.
The Cluster Service function call 'ClusterResourceControl' failed with error code '1008(An attempt was made to reference a token that does not exist.)' while verifying the file path. Verify that your failover cluster is configured properly.
Thanks,
Venu S.
Venugopal S ----------------------------------------------------------- Please click the Mark as Answer button if a post solves your problem!

Hi Venu S,
Based on my research, you might encounter a known issue, please try the hotfix in this KB:
http://support.microsoft.com/kb/928385
Meanwhile since there is less information about this issue, before further investigation, please provide us the following information:
The version of Windows Server you are using
The result of SELECT @@VERSION
The scenario when you get this error
If anything is unclear, please let me know.
Regards,
Tom Li

Windows could not start the Cluster Service on Local computer. For more information, review the System Event Log. If this is a non-Microsoft service, contact the service vendor, and refer to service-specific error code 2.

Dear Technet,
Windows could not start the Cluster Service on Local computer. For more information, review the System Event Log. If this is a non-Microsoft service, contact the service vendor, and refer to service-specific error code 2.
My cluster suddenly went disappear. and tried to restart the cluster service. When trying to restart service this above mention error comes up.
even i tried to remove the cluster through power-shell still couldn't happen because of cluster service not running.
Help me please.. thank you.
Regards
Shamil

Hi,
Could you confirm which account when you start the cluster service? The Cluster service is a service that requires a domain user account.
The server cluster Setup program changes the local security policy for this account by granting a set of user rights to the account. Additionally, this account is made a member
of the local Administrators group.
If one or more of these user rights are missing, the Cluster service may stop immediately during startup or later, depending on when the Cluster service requires the particular
user right.
Hope this helps.
We
are trying to better understand customer views on social support experience, so your participation in this
interview project would be greatly appreciated if you have time.
Thanks for helping make community forums a great place.

HttpSession across application in the cluster

I have two question:
          Can I deploy two webapplication in the cluster, one that is clustered and
          the second one no clustered?
          Assuming I can do this, can I keep a HttpSession across these two
          applications?
          Christian Corcino


Sharing a session across webapp's is prohibited by the servlet spec. You
          could store it in a db and have the other web app access it.
          Christian Corcino wrote:
          > I have two question:
          > Can I deploy two webapplication in the cluster, one that is clustered and
          > the second one no clustered?
          > Assuming I can do this, can I keep a HttpSession across these two
          > applications?
          >
          > Christian Corcino
          Rajesh Mirchandani
          Developer Relations Engineer
          BEA Support

How to create a cluster, whose elements are all scattered across the front panel?

Hi everyone!
I've been designing a front panel, which (I think :-)) feels pretty well when the override switches are scattered all over the screen. Anyway I'd like to process their states in a cluster. I tried pulling them in a cluster in the block diagram, which didn't do anything (actually they hid underneath the cluster). I tried bundling them, but the only way I could do so was after creating a hidden cluster on the front panel and filling it with the same number of boolean controls as I've got override switches.
Anybody got any suggestions?
Kindest regards,
Severin

Severin,
You should be able to bundle the elements without creating a dummy cluster. You cannot use Bundle by Name without a control (or indicator), however.
One possibility, if you need a cluster, is to make a cluster which is bigger than your front panel. Make the background transparent or the same color as the panel. The problem with this is how you handle other controls or indicators which are not part of the cluster. You could put them behnid the transparent cluster, but then you cannot operate the controls and panel updates are slowed considerably by overlapping front panel objects. You may be able to slide them in front of the cluster without embedding them into the cluster by placing them on the panel outside the cluster and using the keyboard arrows to move them. This works for some things; I have not tried it with clusters. You still have overlapping objects.
Lynn

How do I turn off Skype highlighting, so it doesn't automatically do it when I put a phone number on our website through Plone? The code gets repeated so it says "Skype highlighting" all across the page, and it's a pain to have to delete them all.

See question above. Our website hosts think it's something to do with Firefox since it doesn't happen for them. When I go into a web page to update information, if it has a phone number, after I make an update, the words "Skype highlighting" extend all across the page. It's frustrating.

Do you have the Skype extension for Firefox installed?
That probably opens happens for users who have the Skype extension installed.
Try posting at the Web Development / Standards Evangelism forum at MozillaZine. The helpers over there are more knowledgeable about web page development issues with Firefox. <br />
http://forums.mozillazine.org/viewforum.php?f=25 <br />
You'll need to register and login to be able to post in that forum.

Sync files across the Managed Server.

We have exploded war deployed in the Weblogic Cluster and in this there is a feature which would create a zip file in the exploded folder and this zip file will be downloaded through the URL. Now the problem in clustered environment is when the application write the zip file which will be available in only one of the managed server and when we try to download the zip file through the LoadBalancer URL it may not hit the same Managed Server in which the zip file is available.
Is there a way in weblogic cluster to sync the files across all the managed server so that we can access the file through loadbalancer.

Hi Suresh
1. Your issue has nothing to do with the cluster setup and the way you deploy the application (as EAR or Exploded). Basically in your application, you generate a file and automatically that gets stored in that machines location. Now when you access it from a load balancer, the request may go to other machine and naturally it will not find the file there. There is nothing in the cluster setup or something that can copy the files and replicate acorss the machines. Infact that will be way too complex. I will couple of solutions.
2. Create a shared folder like public folder. Say in Windows, create a folder with some folder structure and map it as say X: Drive. now this X Drive can be accessed from any other machine like Machine1, Machine2 (after setting the sharing properties). First try directly from Windows Explorer itself. Now within your application, when you generate and store the file, refer this X: Drive with full path. You can make it more flexible by using a .properties file that has key value pair. Key is like FILE_LOCATION and value can be like X:\a\b\c. Now using simple Java Properties class API to load the contents from this file. You can add all the constants, check for if .properties file exist like that. Once you have all the properties loaded, then refer this key, to get its value and use that to store the file generated. Use the same approach, while retrieving the file also. Use the key to get the location and then look for the file you want to download. This should work, when the request goes to any managed server in the cluster. I used it. I created a utility class like MyAppPropertiesLoader, with some static methods, that reads the file only once, if it is null, and then get values using simle methods like getString(String aKey), getLong(String aKey) etc. The huge BENEFIT of this is, once you deploy the application say in Unix, natually, the shared location has to be one of the unix boxes. Then just mount this location on both the managed servers (mount is similar to mapping a drive in Windows OS). Also the location of the files can be changed anytime in the properties file. Say in QC, the location may be X:\a\b\c\qc. But for UAT on Unix it can be like /x/y/z/uat/. Now this .properties file can exist directly under the Domain root folder so you can access it directly without any path in your java code. You can also use System properties like user.home etc. Make sure to copy this .properties file on both the domains. The original master domain. And the psueduo domain that is generated when you run the unpack command if you have managed server in another machine.
3. This is much simpler but kind of not recommended. Simply, create a simple table with few header columns and one BLOB/CLOB column. Store the file in this database table and retrieve from it. This means you will have JDBC code, and the code to create clob to store and generate file from clob for retrieval.
My preference is the first option given above. Getting tools to copy files may be complex. Its not like syncing the data across but its the files that needs to be copied across the machines.
Thanks
Ravi Jegga

Code synchronization across the cluster

Similar Messages

Maybe you are looking for