10 Ответы Последний ответ: 25.08.2002 4:14, автор: 3004

    Clustering with wl6.1 - Problems

    3004
      Hi,
                
                After reading a bit about clustering with weblogic 6.1 (and thanks to
                this list), I have done the following:
                
                1. Configure machines - two boxes (Solaris and Linux).
                2. Configure servers - weblogic 6.1 running on both at port 7001.
                3. Administration server is Win'XP. Here is the snippet of config.xml
                on the Administration Server:
                
                <Server Cluster="MyCluster" ListenAddress="192.168.1.239"
                Machine="dummy239" Name="wls239">
                <Log Name="wls239"/>
                <SSL Name="wls239"/>
                <ServerDebug Name="wls239"/>
                <KernelDebug Name="wls239"/>
                <ServerStart Name="wls239"/>
                <WebServer Name="wls239"/>
                </Server>
                <Server Cluster="MyCluster" ListenAddress="192.168.1.131"
                Machine="dummy131" Name="wls131">
                <Log Name="wls131"/>
                <SSL Name="wls131"/>
                <ServerDebug Name="wls131"/>
                <KernelDebug Name="wls131"/>
                <ServerStart Name="wls131"
                OutputFile="C:\bea\wlserver6.1\.\config\NodeManagerClientLogs\wls131\startserver_1029504698175.log"/>
                <WebServer Name="wls131"/>
                </Server>
                
                Problems:
                1. I can't figure out how I set the "OutputFile" parameter for the
                server "wls131".
                2. I have NodeManager started on 131 listening on port 5555. But when
                I try to start server "wls131" from the Administration Server, I get
                the following error:
                <Aug 16, 2002 6:56:58 AM PDT> <Error> <NodeManager> <Could not start
                server 'wls131' via Node Manager - reason: '[SecureCommandInvoker:
                Could not create a socket to the NodeManager running on host
                '192.168.1.131:5555' to execute command 'online null', reason:
                Connection refused: connect. Ensure that the NodeManager on host
                '192.168.1.131' is configured to listen on port '5555' and that it is
                actively listening]'
                
                
                Any help will be greatly appreciated.
                
                TIA,
                
        • 1. Re: Clustering with wl6.1 - Problems
          3004

                    
                    Manavendra Gupta wrote:
                    
                    > Hi,
                    >
                    > After reading a bit about clustering with weblogic 6.1 (and thanks to
                    > this list), I have done the following:
                    >
                    > 1. Configure machines - two boxes (Solaris and Linux).
                    > 2. Configure servers - weblogic 6.1 running on both at port 7001.
                    > 3. Administration server is Win'XP. Here is the snippet of config.xml
                    > on the Administration Server:
                    >
                    > <Server Cluster="MyCluster" ListenAddress="192.168.1.239"
                    > Machine="dummy239" Name="wls239">
                    > <Log Name="wls239"/>
                    > <SSL Name="wls239"/>
                    > <ServerDebug Name="wls239"/>
                    > <KernelDebug Name="wls239"/>
                    > <ServerStart Name="wls239"/>
                    > <WebServer Name="wls239"/>
                    > </Server>
                    > <Server Cluster="MyCluster" ListenAddress="192.168.1.131"
                    > Machine="dummy131" Name="wls131">
                    > <Log Name="wls131"/>
                    > <SSL Name="wls131"/>
                    > <ServerDebug Name="wls131"/>
                    > <KernelDebug Name="wls131"/>
                    > <ServerStart Name="wls131"
                    > OutputFile="C:\bea\wlserver6.1\.\config\NodeManagerClientLogs\wls131\startserver_1029504698175.log"/>
                    > <WebServer Name="wls131"/>
                    > </Server>
                    >
                    > Problems:
                    > 1. I can't figure out how I set the "OutputFile" parameter for the
                    > server "wls131".
                    
                    There is no provision to configure output files as of now. It's just
                    stdout and stderr. The usual stuff is written to weblogic.log file
                    anyway. May be you can manually edit the config.xml but i wouldn't
                    recommend that.
                    
                    
                    > 2. I have NodeManager started on 131 listening on port 5555. But when
                    > I try to start server "wls131" from the Administration Server, I get
                    > the following error:
                    > <Aug 16, 2002 6:56:58 AM PDT> <Error> <NodeManager> <Could not start
                    > server 'wls131' via Node Manager - reason: '[SecureCommandInvoker:
                    > Could not create a socket to the NodeManager running on host
                    > '192.168.1.131:5555' to execute command 'online null', reason:
                    > Connection refused: connect. Ensure that the NodeManager on host
                    > '192.168.1.131' is configured to listen on port '5555' and that it is
                    > actively listening]'
                    >
                    
                    
                    If the nodemanager is not on the same host as admin server, you have to
                    give host name of NodeManager in Machine >> NodeManager tab. Otherwise
                    there is no way an admin server can contact a remote machine.
                    
                    --
                    Kumar
                    
                    
                    >
                    > Any help will be greatly appreciated.
                    >
                    > TIA,
                    >
                    
                    
          • 2. Re: Clustering with wl6.1 - Problems
            3004
            The output file is generated with a random name each time you make a
                      request. Its best to ignore the setting but check the contents of the log
                      file.
                      
                      As for the actual communication problem, start with your nodemanager script
                      wtih explicit settings for the Listen Address for the Nodemanager (run java
                      weblogic.nodemanager.NodeManager -help to get full usage)
                      like -Dweblogic.nodemanager.ListenAddress=192.... and ListenPort and also
                      make sure you have all the security certs working correctly. If you are
                      using demo certs, set on teh admin server security/ssl config to
                      "xxxIgnoreHostNameVerification" enabled. Check whether you are able to start
                      now.
                      
                      Also, ensure that the admin server is running on a specified Listen address
                      and add that address to teh nodemanager.hosts file as a trusted host.
                      
                      -Sabha
                      
                      "Manavendra Gupta" <manav@zycus.com> wrote in message
                      news:f6fec444.0208160608.d64df3a@posting.google.com...
                      > Hi,
                      >
                      > After reading a bit about clustering with weblogic 6.1 (and thanks to
                      > this list), I have done the following:
                      >
                      > 1. Configure machines - two boxes (Solaris and Linux).
                      > 2. Configure servers - weblogic 6.1 running on both at port 7001.
                      > 3. Administration server is Win'XP. Here is the snippet of config.xml
                      > on the Administration Server:
                      >
                      > <Server Cluster="MyCluster" ListenAddress="192.168.1.239"
                      > Machine="dummy239" Name="wls239">
                      > <Log Name="wls239"/>
                      > <SSL Name="wls239"/>
                      > <ServerDebug Name="wls239"/>
                      > <KernelDebug Name="wls239"/>
                      > <ServerStart Name="wls239"/>
                      > <WebServer Name="wls239"/>
                      > </Server>
                      > <Server Cluster="MyCluster" ListenAddress="192.168.1.131"
                      > Machine="dummy131" Name="wls131">
                      > <Log Name="wls131"/>
                      > <SSL Name="wls131"/>
                      > <ServerDebug Name="wls131"/>
                      > <KernelDebug Name="wls131"/>
                      > <ServerStart Name="wls131"
                      >
                      OutputFile="C:\bea\wlserver6.1\.\config\NodeManagerClientLogs\wls131\startse
                      rver_1029504698175.log"/>
                      > <WebServer Name="wls131"/>
                      > </Server>
                      >
                      > Problems:
                      > 1. I can't figure out how I set the "OutputFile" parameter for the
                      > server "wls131".
                      > 2. I have NodeManager started on 131 listening on port 5555. But when
                      > I try to start server "wls131" from the Administration Server, I get
                      > the following error:
                      > <Aug 16, 2002 6:56:58 AM PDT> <Error> <NodeManager> <Could not start
                      > server 'wls131' via Node Manager - reason: '[SecureCommandInvoker:
                      > Could not create a socket to the NodeManager running on host
                      > '192.168.1.131:5555' to execute command 'online null', reason:
                      > Connection refused: connect. Ensure that the NodeManager on host
                      > '192.168.1.131' is configured to listen on port '5555' and that it is
                      > actively listening]'
                      >
                      >
                      > Any help will be greatly appreciated.
                      >
                      > TIA,
                      
                      
                      
            • 3. Re: Clustering with wl6.1 - Problems
              3004
              sabha is right. I didn't notice that the log file is created everytime
                        you restart the MS via NM.
                        
                        --
                        Kumar
                        
                        Sabha wrote:
                        
                        > The output file is generated with a random name each time you make a
                        > request. Its best to ignore the setting but check the contents of the log
                        > file.
                        >
                        > As for the actual communication problem, start with your nodemanager script
                        > wtih explicit settings for the Listen Address for the Nodemanager (run java
                        > weblogic.nodemanager.NodeManager -help to get full usage)
                        > like -Dweblogic.nodemanager.ListenAddress=192.... and ListenPort and also
                        > make sure you have all the security certs working correctly. If you are
                        > using demo certs, set on teh admin server security/ssl config to
                        > "xxxIgnoreHostNameVerification" enabled. Check whether you are able to start
                        > now.
                        >
                        > Also, ensure that the admin server is running on a specified Listen address
                        > and add that address to teh nodemanager.hosts file as a trusted host.
                        >
                        > -Sabha
                        >
                        > "Manavendra Gupta" <manav@zycus.com> wrote in message
                        > news:f6fec444.0208160608.d64df3a@posting.google.com...
                        >
                        >>Hi,
                        >>
                        >>After reading a bit about clustering with weblogic 6.1 (and thanks to
                        >>this list), I have done the following:
                        >>
                        >>1. Configure machines - two boxes (Solaris and Linux).
                        >>2. Configure servers - weblogic 6.1 running on both at port 7001.
                        >>3. Administration server is Win'XP. Here is the snippet of config.xml
                        >>on the Administration Server:
                        >>
                        >> <Server Cluster="MyCluster" ListenAddress="192.168.1.239"
                        >> Machine="dummy239" Name="wls239">
                        >> <Log Name="wls239"/>
                        >> <SSL Name="wls239"/>
                        >> <ServerDebug Name="wls239"/>
                        >> <KernelDebug Name="wls239"/>
                        >> <ServerStart Name="wls239"/>
                        >> <WebServer Name="wls239"/>
                        >> </Server>
                        >> <Server Cluster="MyCluster" ListenAddress="192.168.1.131"
                        >> Machine="dummy131" Name="wls131">
                        >> <Log Name="wls131"/>
                        >> <SSL Name="wls131"/>
                        >> <ServerDebug Name="wls131"/>
                        >> <KernelDebug Name="wls131"/>
                        >> <ServerStart Name="wls131"
                        >>
                        >>
                        > OutputFile="C:\bea\wlserver6.1\.\config\NodeManagerClientLogs\wls131\startse
                        > rver_1029504698175.log"/>
                        >
                        >> <WebServer Name="wls131"/>
                        >> </Server>
                        >>
                        >>Problems:
                        >>1. I can't figure out how I set the "OutputFile" parameter for the
                        >>server "wls131".
                        >>2. I have NodeManager started on 131 listening on port 5555. But when
                        >>I try to start server "wls131" from the Administration Server, I get
                        >>the following error:
                        >><Aug 16, 2002 6:56:58 AM PDT> <Error> <NodeManager> <Could not start
                        >>server 'wls131' via Node Manager - reason: '[SecureCommandInvoker:
                        >>Could not create a socket to the NodeManager running on host
                        >>'192.168.1.131:5555' to execute command 'online null', reason:
                        >>Connection refused: connect. Ensure that the NodeManager on host
                        >>'192.168.1.131' is configured to listen on port '5555' and that it is
                        >>actively listening]'
                        >>
                        >>
                        >>Any help will be greatly appreciated.
                        >>
                        >>TIA,
                        >>
                        >
                        >
                        
                        
              • 4. Re: Clustering with wl6.1 - Problems
                3004
                I added the following to the start script of the NodeManager on 131:
                          -Dweblogic.nodemanager.listenAddress=192.168.1.131
                          -Dweblogic.nodemanager.listenPort=5555
                          
                          #netstat -an | grep more
                          tcp 0 0 192.168.1.131:5555 0.0.0.0:* LISTEN
                          
                          When I now try to start this managed server from the admin server, I
                          get the following message:
                          
                          
                          
                          Connected to dummy135:7001 Active Domain: domain Aug 16, 2002
                          11:14:47 PM PDT
                          
                          
                          
                          <Aug 17, 2002 11:47:51 AM IST> <Info> <NodeManager@192.168.1.131:5555>
                          <BaseProcessControl: saving process id of Weblogic Managed server
                          'wls131', pid: 17906>
                          Starting WebLogic Server ....
                          Connecting to http://localhost:7001...
                          ***************************************************************************
                          The WebLogic Server did not start up properly.
                          Exception raised:
                          java.net.ConnectException: Tried all: '1' addresses, but could not
                          connect over HTTP to server: 'localhost', port: '7001'
                          at weblogic.net.http.HttpClient.openServer(HttpClient.java:211)
                          at weblogic.net.http.HttpClient.openServer(HttpClient.java:263)
                          at weblogic.net.http.HttpClient.<init>(HttpClient.java:121)
                          at weblogic.net.http.HttpClient.New(HttpClient.java:156)
                          at weblogic.net.http.HttpURLConnection.connect(HttpURLConnection.java:111)
                          at weblogic.net.http.HttpURLConnection.getInputStream(HttpURLConnection.java:283)
                          at weblogic.net.http.HttpURLConnection.getInternalResponseCode(HttpURLConnection.java:661)
                          at weblogic.net.http.HttpURLConnection.getResponseCode(HttpURLConnection.java:646)
                          at weblogic.management.Admin.getBootstrapLocalServer(Admin.java:1073)
                          at weblogic.management.Admin.initialize(Admin.java:340)
                          at weblogic.t3.srvr.T3Srvr.initialize(T3Srvr.java:359)
                          at weblogic.t3.srvr.T3Srvr.run(T3Srvr.java:206)
                          at weblogic.Server.main(Server.java:35)
                          --------------- nested within: ------------------
                          weblogic.management.configuration.ConfigurationException: connecting
                          to http://localhost:7001/wl_management_internal2/Bootstrap - with
                          nested exception:
                          [java.net.ConnectException: Tried all: '1' addresses, but could not
                          connect over HTTP to server: 'localhost', port: '7001']
                          at weblogic.management.Admin.getBootstrapLocalServer(Admin.java:1164)
                          at weblogic.management.Admin.initialize(Admin.java:340)
                          at weblogic.t3.srvr.T3Srvr.initialize(T3Srvr.java:359)
                          at weblogic.t3.srvr.T3Srvr.run(T3Srvr.java:206)
                          at weblogic.Server.main(Server.java:35)
                          Reason: Fatal initialization exception
                          ***************************************************************************
                          
                • 5. Re: Clustering with wl6.1 - Problems
                  3004
                  I have made some progress:
                            
                            1. The environment settings on 131 were missing. I executed setEnv.sh
                            to setup the required environment variables.
                            2. nodemanager.hosts (on 131) had the following entries earlier:
                                 # more nodemanager.hosts
                                 127.0.0.1
                                 localhost
                                 192.168.1.135
                                 #
                            I changed it to:
                                 #more nodemanager.hosts
                                 192.168.1.135
                            3. The Administration Server (135) did not have any listen Address
                            defined (since it was working without it), but since one of the errors
                            thrown by NodeManager on 131 was - "could not connect to
                            localhost:70001 via HTTP", I changed the listen Address to
                            192.168.1.135 (instead of the null).
                            4. I deleted all the logs (NodeManagerInternal logs on 131) and all
                            log files on NodeManagerClientLogs on 135.
                            5. Restarted Admin Server. Restarted NodeManager on 131.
                            NodeManagerInternalLogs on 131 has:
                                 [root@]# more NodeManagerInternal_1029567030003
                                 <Aug 17, 2002 12:20:30 PM IST> <Info> <NodeManager> <Setting
                            listenAddress to '1
                                 92.168.1.131'>
                                 <Aug 17, 2002 12:20:30 PM IST> <Info> <NodeManager> <Setting
                            listenPort to '5555
                                 '>
                                 <Aug 17, 2002 12:20:30 PM IST> <Info> <NodeManager> <Setting WebLogic
                            home to '/
                                 home/weblogic/bea/wlserver6.1'>
                                 <Aug 17, 2002 12:20:30 PM IST> <Info> <NodeManager> <Setting java
                            home to '/home
                                 /weblogic/jdk1.3.1_03'>
                                 <Aug 17, 2002 12:20:33 PM IST> <Info>
                            <NodeManager@192.168.1.131:5555> <SecureSo
                                 cketListener: Enabled Ciphers >
                                 <Aug 17, 2002 12:20:33 PM IST> <Info>
                            <NodeManager@192.168.1.131:5555> <TLS_RSA_
                                 EXPORT_WITH_RC4_40_MD5>
                                 <Aug 17, 2002 12:20:33 PM IST> <Info>
                            <NodeManager@192.168.1.131:5555> <SecureSo
                                 cketListener: listening on 192.168.1.131:5555>
                            
                                 And the wls131 logs contain:
                                 [root@dummy131 wls131]# more config
                                 #Saved configuration for wls131
                                 #Sat Aug 17 12:24:42 IST 2002
                                 processId=18437
                                 savedLogsDirectory=/home/weblogic/bea/wlserver6.1/NodeManagerLogs
                                 classpath=NULL
                                 nodemanager.debugEnabled=false
                                 TimeStamp=1029567282621
                                 command=online
                                 java.security.policy=NULL
                                 bea.home=NULL
                                 weblogic.Domain=domain
                                 serverStartArgs=NULL
                                 weblogic.management.server=192.168.1.135\:7001
                                 RootDirectory=NULL
                                 nodemanager.sslEnabled=true
                                 weblogic.Name=wls131
                                 
                                 The error generated for the client (131) was:
                                 [root@dummy131 wls131]# more wls131_error.log
                                 ***************************************************************************
                                 The WebLogic Server did not start up properly.
                                 Exception raised:
                                 java.lang.ClassCastException:
                            weblogic.security.acl.DefaultUserInfoImpl
                                 <<no stack trace available>>
                                 --------------- nested within: ------------------
                                 weblogic.management.configuration.ConfigurationException:
                            weblogic.security.acl.
                                 DefaultUserInfoImpl - with nested exception:
                                 [java.lang.ClassCastException:
                            weblogic.security.acl.DefaultUserInfoImpl]
                                 at weblogic.management.Admin.initializeRemoteAdminHome(Admin.java:1042)
                                 at weblogic.management.Admin.start(Admin.java:381)
                                 at weblogic.t3.srvr.T3Srvr.initialize(T3Srvr.java:373)
                                 at weblogic.t3.srvr.T3Srvr.run(T3Srvr.java:206)
                                 at weblogic.Server.main(Server.java:35)
                                 Reason: Fatal initialization exception
                                 ***************************************************************************
                                 
                                 and the output on the admin server (135) is:
                                 <Aug 17, 2002 12:24:42 PM IST> <Info>
                            <NodeManager@192.168.1.131:5555> <BaseProcessControl: saving process
                            id of Weblogic Managed server 'wls131', pid: 18437>
                                 Starting WebLogic Server ....
                                 Connecting to http://192.168.1.135:7001...
                                 <Aug 17, 2002 12:24:50 PM IST> <Emergency> <Configuration Management>
                            <Errors detected attempting to connect to admin server at
                            192.168.1.135:7001 during initialization of managed server (
                            192.168.1.131:7001 ). The reported error was: <
                            weblogic.security.acl.DefaultUserInfoImpl > This condition generally
                            results when the managed and admin servers are using the same listen
                            address and port.>
                                 <Aug 17, 2002 12:24:50 PM IST> <Emergency> <Server> <Unable to
                            initialize the server: 'Fatal initialization exception
                                 Throwable: weblogic.management.configuration.ConfigurationException:
                            weblogic.security.acl.DefaultUserInfoImpl - with nested exception:
                                 [java.lang.ClassCastException:
                            weblogic.security.acl.DefaultUserInfoImpl]
                                 java.lang.ClassCastException:
                            weblogic.security.acl.DefaultUserInfoImpl
                                 <<no stack trace available>>
                                 --------------- nested within: ------------------
                                 weblogic.management.configuration.ConfigurationException:
                            weblogic.security.acl.DefaultUserInfoImpl - with nested exception:
                                 [java.lang.ClassCastException:
                            weblogic.security.acl.DefaultUserInfoImpl]
                                 at weblogic.management.Admin.initializeRemoteAdminHome(Admin.java:1042)
                                 at weblogic.management.Admin.start(Admin.java:381)
                                 at weblogic.t3.srvr.T3Srvr.initialize(T3Srvr.java:373)
                                 at weblogic.t3.srvr.T3Srvr.run(T3Srvr.java:206)
                                 at weblogic.Server.main(Server.java:35)
                                 '>
                                 ***************************************************************************
                                 The WebLogic Server did not start up properly.
                                 Exception raised:
                                 java.lang.ClassCastException:
                            weblogic.security.acl.DefaultUserInfoImpl
                                 <<no stack trace available>>
                                 --------------- nested within: ------------------
                                 weblogic.management.configuration.ConfigurationException:
                            weblogic.security.acl.DefaultUserInfoImpl - with nested exception:
                                 [java.lang.ClassCastException:
                            weblogic.security.acl.DefaultUserInfoImpl]
                                 at weblogic.management.Admin.initializeRemoteAdminHome(Admin.java:1042)
                                 at weblogic.management.Admin.start(Admin.java:381)
                                 at weblogic.t3.srvr.T3Srvr.initialize(T3Srvr.java:373)
                                 at weblogic.t3.srvr.T3Srvr.run(T3Srvr.java:206)
                                 at weblogic.Server.main(Server.java:35)
                                 Reason: Fatal initialization exception
                                 ***************************************************************************
                                 
                            6. Now from the client (131) error, I thought it was something to do
                            with security. So I tried to start weblogic manually (connected as the
                            same user). Curiosly enough, it does start (it threw some errors for
                            some EJBs, but I got the final message):
                                 <Aug 17, 2002 12:30:39 PM IST> <Notice> <WebLogicServer>
                            <ListenThread listening on port 7001>
                                 <Aug 17, 2002 12:30:39 PM IST> <Notice> <WebLogicServer>
                            <SSLListenThread listening on port 7002>
                                 <Aug 17, 2002 12:30:40 PM IST> <Notice> <WebLogicServer> <Started
                            WebLogic Admin Server "myserver" for domain "mydomain" running in
                            Production Mode>
                            7. As you can see the domain on the client (131) is "mydomain". But
                            shouldn't the Admin server be 192.168.1.135, since this is what I have
                            configured for the NodeManager? Or is it that the error occurs because
                            the Admin server Node Manager is configured to work with is 135, while
                            in the default scripts the admin server is itself? I'm confused :-)
                            
                            
                            Help, anyone?
                            
                  • 6. Re: Clustering with wl6.1 - Problems
                    3004
                    6. Now from the client (131) error, I thought it was something to do
                              > with security. So I tried to start weblogic manually (connected as the
                              > same user). Curiosly enough, it does start (it threw some errors for
                              > some EJBs, but I got the final message):
                              > <Aug 17, 2002 12:30:39 PM IST> <Notice> <WebLogicServer>
                              > <ListenThread listening on port 7001>
                              > <Aug 17, 2002 12:30:39 PM IST> <Notice> <WebLogicServer>
                              > <SSLListenThread listening on port 7002>
                              > <Aug 17, 2002 12:30:40 PM IST> <Notice> <WebLogicServer> <Started
                              > WebLogic Admin Server "myserver" for domain "mydomain" running in
                              > Production Mode>
                              
                              Looks like you started the 131 server as a standalone admin server instead
                              of a managed server?
                              
                              Just to simplify the problem space a bit... if you start the 135 admin server, then
                              start the 131 server manually as a managed (not admin) server (and without a
                              nodemanager), does the managed server correctly connect to the 135 admin
                              server and then starts up?
                              
                              regards,
                              -Ade
                              
                              
                              
                              
                    • 7. Re: Clustering with wl6.1 - Problems
                      3004
                      Make sure you are at the ssame service pack and patch levels for the admin and the
                                managed servers.
                                
                                Ade Barkah wrote:
                                
                                > > 6. Now from the client (131) error, I thought it was something to do
                                > > with security. So I tried to start weblogic manually (connected as the
                                > > same user). Curiosly enough, it does start (it threw some errors for
                                > > some EJBs, but I got the final message):
                                > > <Aug 17, 2002 12:30:39 PM IST> <Notice> <WebLogicServer>
                                > > <ListenThread listening on port 7001>
                                > > <Aug 17, 2002 12:30:39 PM IST> <Notice> <WebLogicServer>
                                > > <SSLListenThread listening on port 7002>
                                > > <Aug 17, 2002 12:30:40 PM IST> <Notice> <WebLogicServer> <Started
                                > > WebLogic Admin Server "myserver" for domain "mydomain" running in
                                > > Production Mode>
                                >
                                > Looks like you started the 131 server as a standalone admin server instead
                                > of a managed server?
                                >
                                > Just to simplify the problem space a bit... if you start the 135 admin server, then
                                > start the 131 server manually as a managed (not admin) server (and without a
                                > nodemanager), does the managed server correctly connect to the 135 admin
                                > server and then starts up?
                                >
                                > regards,
                                > -Ade
                                
                                --
                                Rajesh Mirchandani
                                Developer Relations Engineer
                                BEA Support
                                
                                
                                
                      • 8. Re: Clustering with wl6.1 - Problems
                        3004
                        Hi,
                                  
                                  Thanks a lot for your responses. However I'm stuck completing this
                                  stupid document that I am required to submit today (stupid
                                  management!).
                                  
                                  I will post an update tomorrow.
                                  
                                  Thanks once again :-)
                                  
                        • 9. Re: Clustering with wl6.1 - Problems
                          3004
                          Finally I exorcised the demons of the administrative staff! Now that
                                    they have their documents, I have some time to myself :-)
                                    
                                    In the meantime, the machines that i was using were given away to
                                    someone else, so i had to scrounge another one. Here is what i did:
                                    
                                    Administration Server: 192.168.1.135 (XP)
                                    Managed Server:192.168.1.239 (SunOS)
                                    
                                    1. Configure a new Unix machine (dummy239).
                                    - NodeManager listen address 192.168.1.135
                                    2. Configure a new Server (dummy239).
                                    - Selected dummy239 as the machine for the Server.
                                    - Listen Address: 192.168.1.239
                                    3. Create a new cluster (wlcluster)
                                    - Cluster Address: 192.168.1.239
                                    - Multicast Address: 237.0.0.1
                                    - Servers selected: dummy239
                                    4. started weblogic on dummy239 (Admin server is already running) by:
                                    $java -ms64m -mx64m -classpath $CLASSPATH-Dweblogic.Domain=mydomain
                                    -Dweblogic.Name=clusterServer1-
                                    Dweblogic.management.server=192.168.0.101:7001
                                    -Dweblogic.management.username=system
                                    -Dweblogic.management.password=password weblogic.Server
                                    
                                    Above gives me a nicely formatted exception that simply says "$$$$
                                    License Exception $$$$" repeated twice.
                                    
                                    I went back, check the license.bea file - looked file to me, other
                                    than the encrypted signature etc. Compared it with the one on Admin
                                    server, they looked the same to me.
                                    Started dummy239 as a standalone-Admin server, and it works!
                                    
                                    Finally, I went back and used the "startManagedWeblogic.sh" script:
                                    ./startManagedWebLogic.sh dummy239 192.168.1.135:7001
                                    
                                    and voila! this is what it givs me:
                                    Connecting to http://192.168.1.135:7001...
                                    <Aug 23, 2002 4:07:45 PM GMT-05:00> <Notice> <WebLogicServer>
                                    <Starting WebLogic Managed Server "dummy239" for domain "domain">
                                    Starting Cluster Service ....
                                    <Aug 23, 2002 4:07:55 PM GMT-05:00> <Notice> <WebLogicServer>
                                    <ListenThread listening on port 7001, ip address 192.168.1.239>
                                    <Aug 23, 2002 4:07:56 PM GMT-05:00> <Notice> <Cluster> <Listening for
                                    multicast messages (cluster wlcluster) on port 7001 at address
                                    237.0.0.1>
                                    <Aug 23, 2002 4:07:56 PM GMT-05:00> <Notice> <WebLogicServer> <Started
                                    WebLogic Managed Server "dummy239" for domain "domain" running in
                                    Development Mode>
                                    
                                    So, now I wonder:
                                    1. What was I doing wrong in that command line startup?
                                    2. Do I need to specify the NodeManager listen address (on the Managed
                                    Node)?
                                    3. Is the NodeManager involved in any way, for managed servers with a
                                    cluster? Wouldn't the admin server manage all the nodes within the
                                    cluster? Am I missing something?
                                    
                                    Now that my theoretical cluster is working (with just one machine), I
                                    can start looking at capacity planning, what services to cluster and
                                    what to farm, and related stuff....any pointers?
                                    
                                    Thanks to everyone - at least I've got started.
                                    
                                    
                          • 10. Clustering with wl6.1 - Questions
                            3004
                            Hi,
                                      
                                      My setup is:
                                      Admin: 192.168.1.135:7001 (XP). Cluster address:192.168.1.239, 192.168.1.71
                                      managed server 1: 192.168.1.239:7001 (SunOS)
                                      managed server 1: 192.168.1.71:7001 ('98)
                                      
                                      I have a sample SLSB deployed on the server. The deployment desriptors are
                                      configured as:
                                      <stateless-session-descriptor>
                                      <stateless-clustering>
                                      
                                      <stateless-bean-is-clusterable>true</stateless-bean-is-clusterable>
                                      
                                      <stateless-bean-load-algorithm>random</stateless-bean-load-algorithm>
                                      
                                      <stateless-bean-methods-are-idempotent>true</stateless-bean-methods-are-idem
                                      potent>
                                      </stateless-clustering>
                                      </stateless-session-descriptor>
                                      
                                      A test client acess the cluster by specifying the JNDI URL
                                      "t3://192.168.1.71, 192.168.1.239:7001" and works properly.
                                      
                                      1. If I kill one of the managed servers, the requests are load balanced to
                                      the other server. But the behaviour is erratic - after one server dies, it
                                      actually executes faster!?! And no, its independed of whichever managed
                                      server I kill.
                                      2. How do I monitor how many instances were created on each managed server?
                                      Does it the WL console show this somewhere?
                                      3. I have not done anything special to setup the JNDI tree? I'm a li'l hazy
                                      on this (am reading documents on the same). Any pointers? I'm still trying
                                      to grasp the need (I understand the bit about synchronizing the JNDI tree in
                                      the cluster to keep all servers aware of the EJBs deployed on them, and on
                                      others on the cluster, but i'm sure there's more to it that I don't know
                                      about).
                                      4. I find when the managed servers join the cluster, any errors/info
                                      messages are not stored in their respective logs. Is this the right
                                      behaviour? Are these logs stored on the admin server instead? (Is this what
                                      the log file wl-domain.log used for?)
                                      5. How do i monitor the deployed ejbs through the console? Does it tell me
                                      the number of creations/destructions, method calls, etc? How to determine if
                                      the pool is adequate or requires to be increased?
                                      6. What is a realm? How is it helpful?
                                      7. What is a replication group? How does it help me in clustering?
                                      
                                      The more I read, the more I keep getting confused. I don't expect the gurus
                                      to answers all questions, but any help will be appreciated.
                                      
                                      TIA,
                                      Manav.