RMI firewall problems

We are having problems sending RMI calls across the network when going through a VPN connection.
I suspect that the VPNs have embedded firewalls that are blocking the RMI calls.
(Note that If these same calls go through the local network instead of the VPN, or if run with client on server on the same machine then there are no problems at all)
When an RMI call is attempted, the following exception is thrown:
public class NoRouteToHostException
extends SocketException
Exception description: Signals that an error occurred while attempting to connect a socket to a remote address and port.
Typically, the remote host cannot be reached because of an intervening firewall, or if an intermediate router is down.
The VM making the RMI call is a deployed server running on weblogic on a HP Unix server.
JVM version 1.5_016 windows
The VM receiving the call is a thick client code running on a windows PC.
JVM version 1.5_014 unix
We are using the spring RMI libraries to provide the RMI functionality.
I also suspect the blocking is happening at the PC end and not the server,
because whether the call is blocked or not depends on the PC rather than the server environment.
I have found a couple of sources on the web below which I am going through, however it may take a while to try and test things and we need to fix this problem asap, so if anyone has experience or expertise in this area and can advise it would
be really, really helpful.
http://java.sun.com/j2se/1.4.2/docs/guide/rmi/faq.html
http://java.sun.com/developer/onlineTraining/rmi/RMI.html
Many thanks for your help
Paul

See item A.1 in the Javadoc/Guide to Features/Remote Method Invocation/RMI FAQ. You need to find an IP address for the server that you can ping from the client. Then you need to set java.rmi.server.hostname to that address at the server before it exports any remote objects.
If you can't find such an IP address you are hosed. But that'w not an RMI issue, it's a network connectivity issue.

Similar Messages

  • Anyone using Multi plexing to solve RMI firewall problem?

    Hello,
    I read in the RMI protocol spec about Rmi's multiplexing protocol (http://java.sun.com/j2se/1.3/docs/guide/rmi/spec/rmi-protocol7.html)
    which states:
    The purpose of multiplexing is to provide a model where two endpoints can each open multiple full duplex connections to the other endpoint in an environment where only one of the endpoints is able to open such a bidirectional connection using some other facility.
    However i still have not found how to do this using the RMI api provided within Java. I have read numerous posts saying that you CANNOT using the multiplexing protocol, I have read posts which say that you should not allow the client to create a server socket which will force the client to use the multiplexing protocol over an existing socket.
    My main concern is the product i am developing will be used across firewalls, where we are able to create connections from our server to our client, but the firewall disallows the opposite, and our IT department WILL NOT open access on known ports. Thusly our only option is to use the multiplexing protocol if it does in fact exist/work.
    If anyone has any experience using the multiplexing protocol in a firewall environment please post your experience here.
    Thanks,
    Dan

    Hi Dan,
    I have had a lot of success using proxies to work around firewalled clients. I host a free software project at java.net dedicated to spontaneously linking Virtual Machines using RMI.
    https://cajo.dev.java.net
    It uses two classes, ClientProxy & ItemProxy, to implement the protocol.
    More detailed information about this can be found here.
    Best wishes,
    John

  • RMI/Firewall Problem. New approach.

    Hi,
    Please, have a look at our new Socket Factory that provides much
    faster tunneling through firewalls and also allows
    “callback”.
    http://ca.geocities.com/rmi_doves/index.html
    Regards,

    Hi Dan,
    I have had a lot of success using proxies to work around firewalled clients. I host a free software project at java.net dedicated to spontaneously linking Virtual Machines using RMI.
    https://cajo.dev.java.net
    It uses two classes, ClientProxy & ItemProxy, to implement the protocol.
    More detailed information about this can be found here.
    Best wishes,
    John

  • FWM 01009 (null):firewall problem?

    Hi Guys,
    I have a problem with a DMZ installation and the firewall settings
    Here's the situation.
    On our internal network we have installed the servercomponents of BOXI Edge 3.1, which seems to be running fine.
    We have a seperate Webserver with Tomcat 5.5 installed on it in the DMZ zone. Via Wdeploy I have deployed the war files on it (according to the wdeploy deploy document).
    Whenever I want to logon to the CMC on my webserver machine, my login screen is appearing fine, but after I hit logon, I receive the error message FWM 010009 (null), a communication failure has occured. According to our firewall team, all ports are open in both directions. They have executed a trace and see that the BOXI Edge server is sending back packages, but that in that package should be an end communication order.
    I still have (an older) VM with Edge 3.1 behind the firewalls (thus internally), from which I can access my newly installed machine in a glance...
    As an extra test we have installed a complet Edge version on the Webserver as well. Communication on the webserver is OK, but as soon as we try to connect to the server behind the firewall, it goes wrong and the error appears.
    So here comes my question, is there anyone who can give me a clue where it goes wrong?
    Is it correct to say that it is a firewall problem?
    If all ports are open on the firewall, communication shouldn't be a problem, isn't it?
    Any help is appreciated.
    Regards,
    Jorn

    Hi,
    we have appointed a static port 4982 to the request port.
    According to our firewall guys, no communications is coming from the request port.  Even before the request port can answer, the port 6400 is sending a FIN package (termination package) to the application resulting in the error message.
    They say that all ports are open and that it's not a firewall problem...  However when I communicate within my LAN environment this is not a problem...
    Can please someone help me out?
    Regards,
    Jorn
    Edited by: Jornvdd on Sep 29, 2010 12:08 PM

  • I have just installed Lion OS and Face Time encounters server problems on sign up. I have sought the firewall problem without success and even temporarily turned off firewall with no success.

    I have just installed Lion OS and Face Time encounters server problems on sign up. I have sought to rectify the firewall problem without success and even temporarily turned off firewall with no success. Any ideas?

    Some folks have discovered that changing their DNS service fixes FaceTime connection issues.
    The ideal way is to configure your modem/router with DNS service, but often settings in System Preferences/Network/Advanced/DNS on your Mac will override the router settings. Try either of these;
    OpenDNS
    208.67.222.222, 208.67.220.220
    Google Public DNS
    8.8.8.8, 8.8.4.4

  • RMI Marshalling Problem:  weblogic.Admin PING

    WebLogic 5.1.0 with service pack 8 has been installed on a HPUX server. I'm
    trying to ping the server from an NT box, but I'm getting some RMI
    marshalling problems.
    Here is the command that I run on NT:
    C:\weblogic\jre1_2\jre\bin\java -classpath
    c:weblogic/lib/weblogic510sp.jar;c:/weblogic/classes;c:/weblogic/lib/weblogi
    caux.jar weblogic.Admin t3://HPServer:7001 PING > MarshallingProblem.txt
    Here is what is in MarshallingProblem.txt (modified server name):
    Failed to connect to t3://HPServer:7001 due to:
    [weblogic.rmi.UnexpectedException: Marshalling:
    - with nested exception:
    [weblogic.rjvm.PeerGoneException:
    - with nested exception:
    [weblogic.utils.AssertionError: ***** ASSERTION FAILED *****[ Exception
    creating response stream ] - with nested exception:
    [java.io.InvalidClassException:
    weblogic.security.acl.internal.AuthenticatedUser; Local class not
    compatible: stream classdesc serialVersionUID=6699361079932480379 local
    class serialVersionUID=2825328378974757378]]]]
    I previously had similar problems pinging the server from the server itself
    until I included the servicepack in the classpath.
    Anyone have any idea what going on in this situation?
    Cameron Taggart

    Cameron
    Can you test with sp8 installed on your NT machine too ? And also make
    sure you set the classpath with the sp8 jar files on NT before you run
    weblogic.Admin PING
    Raj Alagumalai
    Cameron Taggart wrote:
    WebLogic 5.1.0 with service pack 8 has been installed on a HPUX server. I'm
    trying to ping the server from an NT box, but I'm getting some RMI
    marshalling problems.
    Here is the command that I run on NT:
    C:\weblogic\jre1_2\jre\bin\java -classpath
    c:weblogic/lib/weblogic510sp.jar;c:/weblogic/classes;c:/weblogic/lib/weblogi
    caux.jar weblogic.Admin t3://HPServer:7001 PING > MarshallingProblem.txt
    Here is what is in MarshallingProblem.txt (modified server name):
    Failed to connect to t3://HPServer:7001 due to:
    [weblogic.rmi.UnexpectedException: Marshalling:
    - with nested exception:
    [weblogic.rjvm.PeerGoneException:
    - with nested exception:
    [weblogic.utils.AssertionError: ***** ASSERTION FAILED *****[ Exception
    creating response stream ] - with nested exception:
    [java.io.InvalidClassException:
    weblogic.security.acl.internal.AuthenticatedUser; Local class not
    compatible: stream classdesc serialVersionUID=6699361079932480379 local
    class serialVersionUID=2825328378974757378]]]]
    I previously had similar problems pinging the server from the server itself
    until I included the servicepack in the classpath.
    Anyone have any idea what going on in this situation?
    Cameron Taggart

  • Home Sharing- alleged firewall problem (-3259)

    Hi,
    I've had this problem since I first tried to share any content between libraries with iTunes 9. In fact, I noticed the same problem when I was trying to stream content to my Apple TV a couple of weeks ago.
    So I have an iMac, MacBook Pro and Airport Extreme (802.11n) Base Station. Both Macs have Snow Leopard installed with iTunes 9. None have any third-party firewall software installed nor do they have the firewall turned on in System Preferences.
    Here's the problem: when I try to share content from library to library, it doesn't happen. Specifically, I can initialise a connection between both computers so the other computer's library shows up in the sidebar in iTunes, but that's as far as it gets. When I click on said library, there's a 2 minute wait before the following message is displayed.
    "The shared library "Liam's Music" is not responding. (-3259) Check that any firewall software running on either the shared computer or this computer has been set to allow communication on port 3689."
    I've tried absolutely everything (including a direct Ethernet connection between both Macs- same error message). Most recently, I even verified port 3689 is open using the Network Activity app. If anybody could help me here, I would really appreciate it. I have so much content I want to share to my family's iPods, etc.
    Liam.

    I am having the same problem. Except I am sharing the library from a D-Link NAS. It worked perfectly before upgrading to iTunes 9. It's not a network or firewall problem. Nothing has changed there. The only change is iTunes 9. Any help is appreciated.
    T.

  • Problem of RMI Firewall

    Hi Friends,
    I have a problem when runing RMI currently. My java application has RMI server run on a machine called 'amach' and has rmi client run on PC. The application worked well before change of server machine 'amach' (IP: 393.188.0.12)
    But it doesn't work after server machine be as two IP address (inside machine IP: 299.177.0.14 and outside call this machine use IP: 166.155.0.12 ).
    RMI server is running normally.
    It doesn�t work use either of both IP addresses
    It shows following message when we run RMI client in IP: 166.155.0.12
    java.rmi.ConnectIOException: Exception creating connection to: 166.155.0.12; nested exception is:
         java.net.SocketException: Network is unreachable: connect)
    It shows following message when we run RMI client in IP: 299.177.0.14
    java.rmi.ConnectException: Connection refused to host: 166.155.0.12; nested exception is:
         java.net.ConnectException: Connection timed out: connect
    at sun.rmi.transport.tcp.TCPEndpoint.newSocket(TCPEndpoint.java:601)
         at sun.rmi.transport.tcp.TCPChannel.createConnection(TCPChannel.java:198)
         at sun.rmi.transport.tcp.TCPChannel.newConnection(TCPChannel.java:184)
         at sun.rmi.server.UnicastRef.invoke(UnicastRef.java:110)
    Can you help me to fix this problem?
    Many thanks.
    Hong

    RMI makes the very dubious assumption that a multi-homed server will have at least one IP address that's accessible to everybody. In your case that would be 166.155.0.12 if I've read your posting right. You should set -Djava.rmi.server.hostname=166.155.0.12 at the server JVM and your RMI clients should do Naming.looking("rmi://166.155.0.12/..."). However it all depends on the validity of the assumption above. If that isn't true you either need to talk to your netadmin or reconsider your configuration.

  • RMI activation problem under LINUX

    When I try to run the activation example from the RMI tutorial, I get the following exception:
    Exception in thread "main" java.rmi.activation.ActivationException: ActivationSystem not running; nested exception is:
    java.rmi.NotBoundException: java.rmi.activation.ActivationSystem
    at java.rmi.activation.ActivationGroup.getSystem(ActivationGroup.java:453)
    at examples.activation.Setup.main(Setup.java:68)
    Caused by: java.rmi.NotBoundException: java.rmi.activation.ActivationSystem
    at sun.rmi.registry.RegistryImpl.lookup(RegistryImpl.java:106)
    at sun.rmi.registry.RegistryImpl_Skel.dispatch(Unknown Source)
    at sun.rmi.server.UnicastServerRef.oldDispatch(UnicastServerRef.java:342)
    at sun.rmi.server.UnicastServerRef.dispatch(UnicastServerRef.java:207)
    at sun.rmi.transport.Transport$1.run(Transport.java:148)
    at java.security.AccessController.doPrivileged(Native Method)
    at sun.rmi.transport.Transport.serviceCall(Transport.java:144)
    at sun.rmi.transport.tcp.TCPTransport.handleMessages(TCPTransport.java:460)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run(TCPTransport.java:701)
    at java.lang.Thread.run(Thread.java:536)
    at sun.rmi.transport.StreamRemoteCall.exceptionReceivedFromServer(StreamRemoteCall.java:247)
    at sun.rmi.transport.StreamRemoteCall.executeCall(StreamRemoteCall.java:223)
    at sun.rmi.server.UnicastRef.invoke(UnicastRef.java:350)
    at sun.rmi.registry.RegistryImpl_Stub.lookup(Unknown Source)
    at java.rmi.Naming.lookup(Naming.java:84)
    at java.rmi.activation.ActivationGroup.getSystem(ActivationGroup.java:449)
    ... 1 more
    This is with J2SE 1.4.1 on RedHat 7.3 running a vanilla 2.4.18 kernel. I'm running everything out of shell scripts to keep from getting bitten by typos.
    Here's the punchline: The identical code works just fine on RedHat 6.2 (kernel 2.2.14 or so) - same 1.4.1.
    Netstat tells me there is something listening on the appropriate ports, and output from strace suggests that there is some traffic back and forth to the server. There is no firewall running - I unloaded ipchains in the interest of paranoia. (There is firewalling running on the working system, go figure...)
    Help? I can post more details if that would help resolve the problem.

    I wrote
    When I try to run the activation example from the RMI
    tutorial, I get the following exception:<snip>
    The key was this:
    The identical code works just
    fine on RedHat 6.2 (kernel 2.2.14 or so) - same 1.4.1."Identical" really was - I'd used a tar archive to make a complete copy of the code running on the RH 6.2 system. The copy included the file log/Logfile.1, written by rmid (or rmiregistry). This contains lots of IP addresses; I suspect it's more than a log. Since the original machine and the clone are separated by a firewall, attempts by the clone to contact the original failed.
    Deleting the "log" directory made the problem go away.

  • RMI | Firewall  issue revisited

    Hi all,
    I have searched through these forums for a suitable answer to my problem, but have not gotten any.
    My RMI client-server app was working perfectly well on a Windows 2000 Advanced Server (with its firewall restrictions), until my company decided to buy another firewall appliance. My RMI server used to run on the Advanced server machine with a static IP and I ran my server with the following command :
    java -Djava.security.policy=server.polocy -Djava.rmi.server.hostname=209.XXX.XXX.XXX Server
    (209.XXX.XXX.XXX being the static IP used by the advanced server)
    Now, with the addition of the firewall appliance and after having opened up 2 ports for RMI communication and forwarding them to the Host machine, my application had stopped functioning across the internet.
    The host machine has an internal IP address of say 129.128.***.***
    When an external client tries to establish contact with the my RMI server using 209.XXX.XXX.XXX (the external IP), I get a
    "Connection refused: Unable to connect with 129.128.***.***"
    This makes me believe that even though the client is able to initially establish connection with the RMI server, the RMI server sends back a reply to the client, asking it to find the server at the INTERNAL ip address.
    Is there a way to avoid this ? Where am I going wrong ? How can I resolve this issue ?

    The error is happening because the client is finding and connecting to the RMI registry OK, it is getting the Remote reference OK but when it tries to send a message to the Remote reference it is sending to your internal IP address and hence getting the unable to connect error.
    The IP address in the Remote reference returned by the RMI registry is the one that the server object is bound to. No amount of port forwarding will change that.
    You have two options in my experience:
    1. You can try to force your server to believe it is 209.x.x.x when you bind the Remote instance. You can do this on Linux, but I have no idea how to do it on W2K.
    2. The server.rmi.hostname works for fully qualified domain names, not IP addresses. If you only have one RMI server running, or if you use different ports for different RMI servers, you should be fine by setting the hostname value to your fully qualified domain name and then use NAT to forward to the Internal IP address. If you have more than one server running on the same port number, you'll need to register different FQDN's for each. Then all you need to do is update your internal DNS to point to the internal IP address and the external clients will use their existing DNS in the normal manner.

  • Rmi firewall

    I'm having trouble connecting to my RMI server outside the firewall.
    I've read about HTTP tunneling but It's currently not an option since
    I can't access the web server being used on port 80.
    The concrete problem is this:
    I'm trying to connect from IP address 111.111.111.111 inside one firewall to IP address 222.222.222.222(external) 333.333.333.333 (internal) inside another firewall.
    I'm able to establish a connection to 222.222.222.222 (even though it takes quite a while to establish it), but once I try to execute a method on the remote object I get:
    java.rmi.ConnectException: Connection refused to host: 333.333.333.333;
    Can anybody help me with this?

    I've tried doing that but it doesn't seem to work.
    How do I know if I'm using callbacks and how can I be sure I don't use them?
    Here's my SocketFactory code:
    ----server----
    java.rmi.server.RMISocketFactory.setSocketFactory(new DemoSocketFactory());
    ----server----
    ----DemoSocketFactory---
    public class DemoSocketFactory extends java.rmi.server.RMISocketFactory
    public DemoSocketFactory() {
    public java.net.ServerSocket createServerSocket(int param) throws java.io.IOException
         if (param==0)
         param=1199;
         System.out.println("Creating server socket on port: "+param);
         return new java.net.ServerSocket(param);
    public java.net.Socket createSocket(java.lang.String str, int param) throws java.io.IOException
         System.out.println("Creating client socket to: "+str+":"+param);
         return new java.net.Socket(str, param);
    ----DemoSocketFactory----
    Thanks for the help!

  • RMI-NAT Problem with random ports

    Hi,
    I had a RMI server in a machine with the private IP 10.XX.XX.XX and a firewall with the public IP 196.XX.XX.XX
    I am using the property: java.rmi.server.hostname to 196.XX.XX.XX and a small test application. The server is behind the firewall/NAT machine with a limited set of open ports (including the non standard 8001 port that I am using instead 1099)
    1. The server start perfectly
    2. The client can connect to the server and execute the method Naming.list() successfully
    3. At the moment that the client attempt to perform the bind, the client hang up for almost 7 minutes and finally throws and exception due time-outs.
    According a "strace" and "netstat" commands under Linux and also the flag: java.rmi.server.logCalls we obtain the next data:
    1. The server open the port 8001 (PERFECT)
    2. But also open a random port and the client -at the bind moment- use that port (in the range 34000-35000). And of course our firewall block all the ports except 8001.
    So, is that the expect behaviour? If that is true then RMI is not functional under NAT/Firewall.
    Am I missing something?
    Thanks!!
    Gerardo

    Hello,
    Let me see if I got it right:
    1. You're starting the registry (programatically) in port 8001.
    2. You're creating a server implementation object.
    3. Next you bind/rebind your server to this registry.
    4. Your client hangs at Naming.lookup().
    If this is the problem then that random port you mentioned is the server's attributed port (a ServerSocket that is hanging on accept) at the time of creation. You can avoid this by instead of using the super() call in you serverimpl constructor using the UnicastRemoteObject constructor that specifies a port (check the interface). That way you can configure your firewall to allow traffic through that port too.
    However, what puzzles me is the fact that it hangs on step 4 instead of in a subsequent remote call. step 4 should be using port 8001. Can you confirm that it's hanging in Naming.lookup ?
    (two other sources of data you may find usefull are the Firewall logs and a tcpdump analisys of the traffic between the firewall and the server machine)
    Nuno

  • ASA 5510 FireWall Problem

    Hi All
    After some advise and direction
    Our ASA firewall using ASA version 8.4 has recently started presenting us with a problem to one external website
    called http://partners.highnet.com/login/  ip address 62.233.82.181.
    Our firewall is letting everything on our inside Trusted site 192.168.254.0/24 out through our outside interface on x.x.x.x
    to any website and brings back the details
    However when we try to reach http://partners.highnet.com/login/ we recently started receiving (Internet Explorer cannot display the webpage)
    on checking the ASA under Home TAB       -       Firewall Dashboard    -    and then under     -      Top 10 protected Servers under SYN attack we are receiving the below error.
    Rank        Server IP-Port           Interface     Average          Current                    Total                           Source IP (Last Attack Time)
    5
               62.233.82.181:80
          INSIDE
                0
                     0
                            8
                              192.168.254.130 (1 mins ago)
    I have tried rebooting the ASA firewall (Still did not resolve).
    I have also  disabled basic threat detection and threat detection statistics and then re-enabled after a period of time under > configuration > Firewall > threat detection  (Still did not resolve).
    Have created a number of access list both from the inside to outside and outside to inside allowing TCP just to the specific IP address 62.233.82.181 (Still did not resolve).
    Tried editing Global Policy for Http configuration > connection settings TCP and UDP connections and also Embryonic connections (Still did not resolve).
    Also tried using the shun command on the ASA to clear connection and statistics and (Still did not resolve).
    So you see there is nothing else I can think of doing, so that is why I have asked you for some pointers maybe someone has come across this sort of issue before.
    If you can help or advise it is much appreciated.

    Hi,
    Are you sending logs from your ASA to any Syslog server from which you could pull all the connection logs for that destination IP address?
    On the ASA you can naturally use "packet-tracer" also to simulate one such packet coming from your LAN towards this WAN IP address (of the server) and confirm that all rules are correct.
    packet-tracer input INSIDE tcp 192.168.254.130 12345 62.233.82.181 80
    You could maybe also try to generate TCP SYNs directly from the ASA
    ping tcp 62.233.82.181 80
    And see if the server replies
    - Jouni

  • Finally a solution to all your firewall problems

    I finally have answer to all of you who are seeking this
    question.
    When the Flash Player encounters the connect() method, it
    attempts to connect to your FCS following a very specific pattern.
    It first attmpts to connect over port 1935, then port 433, then
    port 80 and then it automatically attempts to tunnel through by
    sending rtmp data over http also called RTMPT it turns out the
    automatic sequence of attempts is fine for many situations however
    you may find some users can't make a connection (well , the
    NetConnection can't) because their firewall blocks data from
    traveling through port 1935. If those users can connect to any
    website surely they can use port 80 .And even if their setup blocks
    rtmp surely they can use RTMPT over port 80 because its really just
    plain http ...you might think the final attempt by flash player
    RTMPT on port 80 to be failsafe however the problem is that if you
    are running your webserver on the same machine as FMS then they
    can't both use port 80 at the same time....a perfectly legitimate
    solution is to use to computers one for the webserver and one for
    FMS ......two computers probably means you need to specify the
    domain in your rtmp (rtmp://mydomain.com/video).
    To force the flash player to try to connect through a
    specific port other than the defaults as specified above.....write
    rtmp as ....my_nc.connect("rtmp::8500/mydomain.com/video);
    FlashMX2004 by Phillip Kerman

    changing my post back to orginal ...........open port 433 and
    1935 if that doesn't work try one of the other solutions.
    I finally have answer to all of you who are seeking this
    question.
    When the Flash Player encounters the connect() method, it
    attempts to connect to your FCS following a very specific pattern.
    It first attmpts to connect over port 1935, then port 433, then
    port 80 and then it automatically attempts to tunnel through by
    sending rtmp data over http also called RTMPT it turns out the
    automatic sequence of attempts is fine for many situations however
    you may find some users can't make a connection (well , the
    NetConnection can't) because their firewall blocks data from
    traveling through port 1935. If those users can connect to any
    website surely they can use port 80 .And even if their setup blocks
    rtmp surely they can use RTMPT over port 80 because its really just
    plain http ...you might think the final attempt by flash player
    RTMPT on port 80 to be failsafe however the problem is that if you
    are running your webserver on the same machine as FMS then they
    can't both use port 80 at the same time....a perfectly legitimate
    solution is to use to computers one for the webserver and one for
    FMS ......two computers probably means you need to specify the
    domain in your rtmp (rtmp://mydomain.com/video).
    To force the flash player to try to connect through a
    specific port other than the defaults as specified above.....write
    rtmp as ....my_nc.connect("rtmp::8500/mydomain.com/video);
    FlashMX2004 by Phillip Kerman

  • Java.rmi.ServerException Problem

    The exception that has occured is java.rmi.ServerException: RemoteException occurred in server thread; nested exception is:
    java.rmi.UnmarshalException: error unmarshalling arguments; nested exception is:
    java.lang.ClassNotFoundException: Process00Impl_Stub
    I have the interface, Client, Server and the Implementation java files
    in the same machine.
    11/29/2002 08:52p 282 Process00.class
    11/29/2002 08:25p 221 Process00.java
    11/29/2002 07:40p 215 Process00.java.bak
    11/29/2002 08:56p 1,008 Process00Client.java
    11/29/2002 08:55p 1,010 Process00Client.java.bak
    11/29/2002 08:57p 451 Process00Impl.class
    11/29/2002 08:57p 936 Process00Impl.java
    11/29/2002 08:52p 934 Process00Impl.java.bak
    11/29/2002 08:57p 2,153 Process00Impl_Skel.class
    11/29/2002 08:57p 3,938 Process00Impl_Stub.class
    11/29/2002 08:56p 995 Process00Server.class
    11/29/2002 08:57p 645 Process00Server.java
    11/29/2002 08:56p 643 Process00Server.java.bak
    I also have the _Stub.class. I do not know why I get this error. If any
    one have an idea, kindly share with me.
    Thanks.

    Hi,
    I had the same problem. In my machine, path was having jre1.1.7 before my jdk 1.4. So it was picking up from jre1.1.7.
    Java was happy when I moved my jdk1.4 path to the beginning of the path variable.
    Hope this helps you.

Maybe you are looking for