Distributing connections

Just to save up my time testing the following I'm gonna ask it here because I know you guys know the answer anyway :)
Q: Can I pass on socket via RMI?
Description : My server receives client connection via ServerSocket. I need to distribute that socket directly to other computer in the LAN. What happens if I send the socket(object) to another computer, does the client get disconnected when it notices that server IP has changed? What about if the distributed environment is behind NAT so that the client doesn't know which computer behind the NAT it is currently talking with? I don't know how socket is handled at ip stack level or how NAT reacts to ip shift. Is Java too high level language for this kind of purpose? What do you consider the best way to distribute multiclient connections in a multiserver environment?
Thanks in forward for all the answers :)

Passing objects that implement serializable is possible, no matter what object it is. However, sockets use parts of the native system which some parts will NOT be the same for each machine. When you create a network connection, no matter which protocol used, will translate the IP into a MAC address. MAC addresses are globally unique (well pretty much globally). This means that machine A sends the Socket object to machine B, machine B will not be able to use the socket object sent in any way. This renders the whole operation useless.

Similar Messages

Unable to get distributed destination in cluster

[Cross post with interest.jms]
          Hi,
          I'm working with Weblogic 8.1.2 on Windows. Up until now I've always worked with a single server on which I have configured a JMS server and a number of queues. My clients get references to the queues using the createQueue() method. I now need to test in a clustered environment and my clients are no longer able to create queues; my clients gets a "Invalid destination name" each time.
          I've created a simple cluster with 2 servers and one distributed connection. Each server in the cluster has a uniquely named JMS server. I created a distributed connection called C1_DIST_SECURITY_PERSISTENCE and then created the members automatically for the cluster. My understanding from what I've read is that is the client should no longer need to name the JMS server when calling createQueue, instead the distributed destination name should be sufficient. So, my client calls
          session.createQueue("C1_DIST_SECURITY_PERSISTENCE")
          Everytime it does this it gets the exception: weblogic.jms.common.JMSException: Invalid destination name: C1_DIST_SECURITY_PERSISTENCE
          As an aside I modified my code to try explicitly looking up one of the destination members. In this case a totally different exception occurs:
          eblogic.utils.AssertionError: ***** ASSERTION FAILED *****[ Environment not found on thread ] at weblogic.jndi.internal.NamingNodeReplicaHandler.<init>(NamingNodeReplicaHandler.java:150) at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method) at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:39)
          Any help here greatly appreciated.
          Thanks,
          Aoife

Hi Gsrc,
This is my code in the PR.
public void processRequest(OAPageContext pageContext, OAWebBean webBean)
super.processRequest(pageContext, webBean);
String itemNum=pageContext.getParameter("itemNum");
System.out.println("item nnum "+itemNum);
i am getting a null value in the print stmt.
Thanks.
Message was edited by:
SRK84

Consumers for distributed queues -- writing the JMS API

Hi,
I have Created a cluster MS_01 and MS_02.
JmsServer-MS01 for MS_01 and JmsServer-MS02 for MS_02.
Also have create a distributed queue called DistributedQueue-0 and a distributed connection factory called ConnectionFactory-0.
I have been successfully able to loadbalance sending of messages using the cluster address.
When i try to consume messages the consumers gets pinned to a MS_01. (As expected)
I am looking for ways where in using JMS API i can read messages from both the manages instances.
Appreciate any help.
thanking you all in advance.
Regards
Girish Subbaramu.

Hi,
For consuming messages from a distributed queue, by far the best option is to use a WebLogic MDB, which has internal code designed to work well with WebLogic distributed destinations. When a distributed queue runs in the same cluster as the MDB, the MDB service will ensure that there's an MDB deployment instance that is locally servicing each queue member -- even when the queue members migrate between servers, or when they are added or removed due to configuration changes. When the MDB is communicating with a distributed queue in a remote cluster, an MDB running on a single server will automatically set up consumers for each remote member and will dynamically react to member up/down add/remove events.
If you must use a straight javax.jms client, then I think there are two basic options. The first option is to ensure that that there are more than enough consumers to provide full coverage, and periodically close and re-open the consumers. The periodic close and re-open forces the consumers to load-balance again, and therefore helps handle the case when new destination members become available after the consumers have all attached. The second option is to create consumers directly on each member rather than on the distributed queues logical name (each member has a unique name that's advertised in JNDI).
Hope this helps,
Tom

Help Needed to configure sites on IIS CFMX 6.1 in distributed mode

I have a distributed connection between my IIS 4.0 and CFMX 6.1 STANDALONE server.
I need add additional web sites. The default site root is at c:/wwwroot/inetpub
The new site is at c:/wwwroot/iweb1_psms
The new site is created on IIS, including virtual dir JRunScripts which point to the root\runtime\lib\wsconfig\2 folder.
default site is \1, so new site is \2,
I'm not sure what belongs in the wsconfig.properties file. ?
Can anyone provide a sample of this file using multiple sites on CFMX 6.1 Standalone server?
Thanks,

I found a fix for my wsconfig error and now I can use the wsconfig tool but I still can't get my sites to be recognized.
if I do a command line list:
C:\CFusionMX\runtime\bin>wsconfig -list
Internet Information Server (IIS) :
There are 2 configured sites and 0 unconfigured sites.
Configured Site Number and Site Name:
0 All
1 Default Web Site
If I try to create the connection for a web site named iweb1_psms with an IP 192.168.251.90 (that's a valid IIS web site on the IIS server)
C:\CFusionMX\runtime\bin>wsconfig -server iweb1 -ws iis -site "192.168.251.90" -
coldfusion -v
Created file C:\CFusionMX\runtime\lib\wsconfig\jrunwin32.dll
IIS web site 192.168.251.90 does not exist
I've found out from my network guys that the web server name iweb has the same IP as the default web site. I don't know if that is standard practice or not. My question is how do you folks that run multiple websites with CFMX 6.1 set them up? What is the physical location of the files?
Thanks, Jim

Number of users per room limit = 5000

Hello,
Regarding the 5000 users limit per room in LCCS:
Is it an automatic limitation, or is there a way to over pass this limit?
And why do you apply this limit?
In a previous discussion, i was answered: "ideally if you want to build a really scalable app, you should distribute connections in multiple rooms"
But if i have multiple chat rooms, the user will have to endure the connection time (not so short) every time he changes chat room. When i use a single LCCS room for all my chat discussions, i can limit to only one connection time for the whole App, which is much better user experience.
And if i need a global chat room available for all logged in users?
Moreover, i also use the global room connection between all users to keep some common informations updated live, especially for example to keep the online users, and the online state of chat rooms hosts. How can i do differently without connecting every user at same time?
If i can't get over this 5000 limitation, what solution Adobe has to offer, to get the great LCCS features, whitout limitations? A FMS server? but i'll have to recreate all LCCS services myself?
To be clear: I need a scalable realtime service between all users of my app (not only a restricted chat room at a time). How can i do?
Thanks,
Regards,
Seb

Hello,
Thanks for the detailed answer.
I should read more about those customizable LCCS contracts, but are those contracts still rulled by the "you pay what you use" basis?
And about the servers. You say the FMS servers are not unlimited. But is the LCCS service not based on cloud infrastructure?
I thought cloud computing, was able to create virtually unlimited servers sizes, which would be perfect for the scalability of FMS heavy uses.
Am i wrong about cloud computing?
is LCCS not totally based on cloud instratucture?
can i get over those limitations if i use my own FMS server on Amazon EC2 recently announced? (or another Adobe product?) but in this case, ill have to create my own LCCS implementation, from scratch, which is quite some work to do
I am a bit disappointed from those limitations (even over 5000) because i love the LCCS business model and powerfull fonctionnalities.
It is so great to say: "you pay only when you have success" (because with a good business model of a web project earnings growth along with success), and you can have a success from 2 users to potentially millions without worries about technical background and its price.
Thanks,
Regards,
Seb

Internet Routing a limited-access PC to my iMac AND xbox 360

Hello All,
I have a company work computer (PC) at home which does not have administrative access, however, I've been able to download software like firefox and save it to my desktop only.
Next to this PC, I have my iMac. On the other side of the wall, is my tv room with our xbox 360 which is really a powerful online gaming console.
I have been frustrated in the past in routing an internet connection to be shared with the PC and iMac. And now I'd like to share the internet with the 360 at the same time. I've had to share the internet "cave man" style by unplugging my Comcast Modem, unplug the ethernet connection, reconnect the ethernet from whichever system I want internet on, and then plug in the power.
I have tried many times to make this work but have been unsuccessful as I think it's tricky to make the PC and iMac work together. The IT people at work say the iMac has it's own IP number and I don't know if that's prevent the network from working. I've been trying to research the "how-to" and came across something about a "DHCP" but I don't even know if that's my issue.
I'm reluctant in buying an airport router from Apple because it's not 100% refundable if the packaging is open.
If you can help, please teach me how to do this. A huge thank you in advance.
Thank you!

By default most routers do not use the MAC address for distributing connections, but they can be configured to do so. Modern routers' security settings can be configured to allow access only to machines whose MAC addresses have been entered into the Include list. This creates a "private" network only accessible to those whose MAC addresses have been added to the list, and prevents others from gaining access to the network. Used together with passwords and firewalls they provide effective barriers to prevent unwanted access to the network. Using the MAC address is a good way of securing a wireless network from access by users "outside" of the local network.
The mechanism used to share an internet connection is called NAT (Network Address Translation) that is used to "map" a real IP address (the one received by the modem from the ISP) to multiple devices. The actual assignment of the local IP numbers is usually done by the DHCP server in the router (Dynamic Host Configuration Protocol.) The DHCP server obtains all the IP parameters required by connected devices so they can communicate through the router - the important ones being the Gateway IP, real IP, network mask, and the local IP.
In theory a router makes it possible for multiple device such as a Mac, a PC, an XBox, and a printer to all connect to a local network and to communicate with each other. I don't use any gaming machines like an XBox, but my home LAN is setup using an Airport Extreme Base Station that is connected to my cable modem and to a high speed network switch. I have three computers, two network printers, and a SlingBox currently connected in the network. The latter connects wireless from an Airport Express. In addition to the devices normally connected there are two other laptops that may connect wirelessly when needed. I have a second Airport Express in the house that is used to help extend the network because my house has plastered walls with metal laths that tend to reduce wireless signal strength.

Installing Oracle 10g Forms and Reports in Cluster Mode

Dear All,
My OS version is RHEL 4
In our current setup we have oracle 10g database two node cluster and Application Server 10g to node cluster.
Currently our report server has been configured to run in standalone mode
I have been asked to configure oracle 10g forms and report server in cluster mode.
Please provide me with the documentation link to configure oracle 10g reports server in cluster mode.
Thanks and Regards
Monoj Das

my main goal is load balancing only
Then what I've done in 2007 might be what you need. In that thread there is a link to Web Cache Administrator's Guide : there you'll find everything you need to configure the web cache as a load balancer, particularly chapter 8 Setup and Configuration.
Anyway, here briefly some configurations : we have two machines, say MachineA and MachineB. The DNS resolves MachineA to a virtual address, which is normally assigned to MachineA; if MachineA crashes, it gets assigned to MachineB.
Users URL looks like this : http://MachineA:<port>/forms/frmservlet?config=<some config>
MachineA's web cache distributes connection requests between MachineA and MachineB.
Of course we're now using 11g as well, but that's another thing.

Java Networking Books

I am doing a project in Java Networking,kindly suggest me a good,solid
book on Java Networking.

I think the combination of the worker thread from Example 12.2 and its driver Example 12.3 do not distribute connections evenly among threads. Or, am I wrong?
p.s. Hey Mr Moderator
If you feel like to delete existing forum thread(s), please give prior alert.
We have to take backup.

Get error from SQL AGENT: The SSIS Runtime has failed to enlist the OLE DB connection in a distributed transaction with error 0x8004D024

I am running SQL Agent that executes an SSIS process from sql server1. The SSIS process executes its SQL/tables/sp’s against another sql server2.
I get an error after adding data flow tasks with transaction supported within a sequence with transaction required. The error, “The SSIS Runtime has failed to enlist the OLE DB connection in a distributed transaction with error 0x8004D024 "The transaction
manager has disabled its support for remote/network transactions"
Prior to adding this sequence everything was working from sql agent, and there were other sequences with oledb destinations.
Everything works when running within SSIS Package.
I see this article on similar issue,
https://social.msdn.microsoft.com/Forums/sqlserver/en-US/0bfa2569-8849-4884-8f68-8edf98a9b4fe/problem-to-execute-a-package-and-tasks-with-a-certain-transactionoption-property-i-need-help?forum=sqlintegrationservices
“I had similar issue and solved by setting the following on both the machines. Allow Remote Clients, Allow Remote Administration,
Allow Inbound Clients, Allow Outbound Clients, and TIP are enabled in Component Services/My Computer/Properties/MSDTC/Security Configuration.”
I don’t want to remove transaction required for the “Remove Duplicates from Staging” sequence.
Anyone seen this?
Greg Hanson

DTC was running on remote computer. The problem was it was no longer accepting transactions from remote servers. This was in SSIS so I had to turn to "Transaction Supported" for all Data Flow Transactions.
Greg Hanson

Is HS Generic Connectivity(db2/400 odbc) supports Distributed Transaction?

hi,
Is HS Generic Connectivity(db2/400 odbc) supports Distributed Transaction?
I am getting an error as follows
"ORA-02047 cannot join the distributed transaction in progress" when i am trying a transaction, which is inserting in to db2/400 and updating in Oracle.
if it does not support this, what is the best way
to do such things using generic connectivity.
(eg. by committing each sql statement ?)
my db is oracle 9i 9.2.0.4.0
remote db is db2/400 connected with client access ODBC.
Thanks
BA

Hi,
one of the restrictions of generic connectivity is, that distributed transactions are not allowed.
See manual: 7-6 Oracle9 i Heterogeneous Connectivity Administrator's Guide
- Generic Connectivity agents cannot participate in distributed transactions; they
support single-site transactions only
So you can only commit after each statement or you have to use the GATEWAY

Connection Factory class for Distributed Transaction?

Definition of Managed Datasources goes like this at URL: http://download.oracle.com/docs/cd/B31017_01/web.1013/b28958/datasrc.htm#CHDDADCE
Managed data sources are managed by OC4J. This means that OC4J provides critical system infrastructure such as global transaction management, connection pooling, and error handling.
In the same document, I could also see a section for configuring Connection Pool, that uses 'factory-class' as 'oracle.jdbc.pool.OracleDataSource' as shown below:
<connection-pool name="myConnectionPool">
<connection-factory
factory-class="oracle.jdbc.pool.OracleDataSource"
user="scott"
password="tiger"
This configuration has worked well for my web application where I was using a single instance of Oracle database. In a new application I've to use two separate instance of Oracle databases and there is a need of distributed transaction.
I know, I've to create two separate datasources for the two separate Oracle instances.
My question is: Since now my transaction will be distributed in nature, do I need to use any other factory-class for configuring XA aware Datasources on OC4J? Or whether the same factory-class (oracle.jdbc.pool.OracleDataSource) will work even for distributed transactions?

Here is the link for using Oracle RAC with WLS
http://e-docs.bea.com/wls/docs81/jdbc/oracle_rac.html

How to create Bridge connecting distributed Topic to queue

Hi,
I am trying to create a bridge which connects distributed topic with JMS queue.
Both the topic and queue are in the same domain.
But it is not getting connected. Status is failed.
Below is the configuration -
<messaging-bridge>
<name>DisTopicBridge</name>
<target>Managed_soa_server01</target>
<source-destination>DisTopicSource</source-destination>
<target-destination>QueueSource1</target-destination>
<selector></selector>
<quality-of-service>Exactly-once</quality-of-service>
<qos-degradation-allowed>false</qos-degradation-allowed>
<durability-enabled>true</durability-enabled>
<idle-time-maximum>60</idle-time-maximum>
<async-enabled>true</async-enabled>
<started>true</started>
<preserve-msg-property>false</preserve-msg-property>
</messaging-bridge>
<jms-bridge-destination>
<name>DisTopicSource</name>
<adapter-jndi-name>eis.jms.WLSConnectionFactoryJNDIXA</adapter-jndi-name>
<classpath></classpath>
<connection-factory-jndi-name>jndi/TestConnectionFactory</connection-factory-jndi-name>
<initial-context-factory>weblogic.jndi.WLInitialContextFactory</initial-context-factory>
<connection-url>t3://xxx.xx.xxx.40:9101</connection-url>
<destination-jndi-name>jndi/TestDistributedTopic</destination-jndi-name>
<destination-type>Topic</destination-type>
</jms-bridge-destination>
<jms-bridge-destination>
<name>QueueSource1</name>
<adapter-jndi-name>eis.jms.WLSConnectionFactoryJNDIXA</adapter-jndi-name>
<classpath></classpath>
<connection-factory-jndi-name>jndi/QConnectionFactory</connection-factory-jndi-name>
<initial-context-factory>weblogic.jndi.WLInitialContextFactory</initial-context-factory>
<connection-url>t3://xxx.xx.xxx.40:9101</connection-url>
<destination-jndi-name>jndi/TestQueue2</destination-jndi-name>
<destination-type>Queue</destination-type>
</jms-bridge-destination>
Error in the log:
<Warning> <MessagingBridge> <server-app-01> <Managed_soa_server01> <[ACTIVE] ExecuteThread: '0' for queue: 'weblogic.kernel.Default (self-tuning)'> <<WLS Kernel>> <> <d632395af8efc868:3b2a67dd:13752cfef88:-8000-00000000000d3c66> <1337198463908> <BEA-200026> <Bridge "DisTopicBridge" encountered some problems in one of its adapters or underlying systems. It stopped transferring messages and will try to reconnect to the adapters shortly. (The exception caught was javax.resource.ResourceException: Error setting message listener.)>
<Info> <MessagingBridge> <server-app-01> <Managed_soa_server01> <[ACTIVE] ExecuteThread: '0' for queue: 'weblogic.kernel.Default (self-tuning)'> <<WLS Kernel>> <> <d632395af8efc868:3b2a67dd:13752cfef88:-8000-00000000000d3c67> <1337198463908> <BEA-200020> <Bridge "DisTopicBridge" is stopped.>
<Info> <MessagingBridge> <server-app-01> <Managed_soa_server01> <[ACTIVE] ExecuteThread: '1' for queue: 'weblogic.kernel.Default (self-tuning)'> <<WLS Kernel>> <> <d632395af8efc868:3b2a67dd:13752cfef88:-8000-00000000000d3cb9> <1337198471319> <BEA-200034> <Bridge "DisTopicBridge" is shut down.>
Is there any special way we have to configure the Bridge while using Distributed topic?

Hello,
Please find the more details about the error we are receiving ( BEA-200026 )
http://docs.oracle.com/cd/E13222_01/wls/docs100/messages/Bridge.html
BEA-200026
Warning: Bridge "arg01" encountered some problems in one of its adapters or underlying systems. It stopped transferring messages and will try to reconnect to the adapters shortly. (The exception caught was t.)
Description
This message indicates that errors occurred during the process of transferring messages. The bridge stopped its connections to both adapters and will attempt reconnect soon.
Cause
There was a problem in receiving or sending messages to one of the sides of the bridge.
Action
No action required.
I would also request you to specify the version of weblogic and verify if the adapter is being deployed properly.
Please reveiw the bridge document
http://docs.oracle.com/cd/E11035_01/wls100/bridge/bridgefaq.html
Can you also check the status of the bridge
Messaging –> Bridges –> YOUR_Bridge_NAME –> Monitoring [tab]
Please provide the above required details
-Vishal Iyer

OVD adapter in distributed mode fails to connect to secondary AD

We have 2 AD domain controllers configured on the OVD adapter in distributed (load-balancing mode). The secondary LDAP server fails on a user bind request but the primary is fine. If we remove the primary we get very few, if any, connections. Any known issues with 10g and AD in fail over mode?
Thanks

Hi everyone, I got it finally after some more headbashing! It turns out I can't run networkmanager along with wpa_supplicant at the same time. I don't understand why though.
Before I happened to install networkmanager along with network-manager-applet via pacman.
# pacman -S network-manager-applet
# pacman -S networkmanager
# systemctl enable NetworkManager
# systemctl start NetworkManager
So I just stopped it
# systemctl disable NetworkManager
# systemctl stop NetworkManager
and then uninstalled it
# pacman -R networkmanager
# pacman -R network-manager-applet
Then wpa_supplicant works! Thanks.

He iTunes Store account entered is already being used for an iTunes Connect account that distributes Apps

I am getting the following message when trying to sign up with iTunesConnect for selling our companies eBooks.
"The iTunes Store account entered is already being used for an iTunes Connect account that distributes Apps. To continue with this application, you must enter a different iTunes account."
The email account I am using is not being used for distributing apps. We need to sell ebooks.
We DO have a developer account to sell apps but it is a different email address (confirmed).
How should I go about getting this straitened so we can get into iTunesConnect?

That worked, thank you very much! The Apple support document is quite misleading then.

Distributed transactions and connection pooling

Hi,
We have a problem with the enlistment of connections in distributed transactions.
The following very simple C# program does not work correctly (a simple form with one button and a class hosted in COM+). If you put a breakpoint on "cnOraTx.Close();" you will see that the inserted record is already visible in the database although the transaction has not commited yet.
If you comment out the lines between //START and //END everything works as expected.
Also, if we use "Pooling=false" in the connection string, everything works fine again, but since we are writing a web application, we really need the connection pooling.
We are using ODP.NET 9.2.0.2.102 (I also tried with the ODP.NET 9.2.0.4.0 BETA) and an oracle 8.1.7 database.
Best Regards,
Piet
private void button1_Click(object sender, System.EventArgs e)
//When commenting the following lines the connection is enlisted in the distributed transaction
//START
OracleConnection cnOraTx;
cnOraTx = new Oracle.DataAccess.Client.OracleConnection("Data Source=ECOMP532;User ID=USER;Password=PWD");
cnOraTx.Open();
cnOraTx.Close();
cnOraTx.Dispose();
//END
OpenOraTransConn obj = new OpenOraTransConn();
obj.Connect();
//This class is going to be hosted within COM+
[Transaction(TransactionOption.Required)]
public class OpenOraTransConn: ServicedComponent
private OracleConnection cnOraTx;
private OracleCommand cmd;
internal void Connect()
cnOraTx = new OracleConnection("Data Source=ECOMP532;User ID=USER;Password=PWD");
cnOraTx.Open();
cmd = new OracleCommand();
cmd.Connection = cnOraTx;
cmd.CommandText = "INSERT INTO SYSADM.PERSON (PE_KEY) VALUES(130000)";
cmd.ExecuteNonQuery();
cnOraTx.Close();
cnOraTx.Dispose();
ContextUtil.SetComplete();

Hello:
We are experiencing the exact same problem. INSERT, UPDATE or DELETE statements that should roll back because an exception is being thrown from inside our ServicedComponent-derived class (which is set to TransactionOption.Required and uses AutoComplete) are not -- and instead are being committed to the database.
Were you able to figure out why this is happening, and what can be done to get around it? We too were able to avoid the problem by setting Pooling=False to our connect string, but for a variety of reasons we want to use connection pooling ...!

Distributing connections

Similar Messages

Maybe you are looking for