12 Replies Latest reply on May 20, 2010 1:06 AM by funa516

    Mod_cluster can't read status in non-multicast environment

      Hi,

       

      I implemented mod_cluster, but frontend httpd cannot read STATUS command from backend jboss.web.

       

      I checked Mod_cluster status page by web browser,
      the status is following.

       

      ===================================================================
      mod_cluster/1.1.x
      Auto Refresh show DUMP output show INFO output

       

      Node 127.0.0.1:8009:jboss.web (ajp://127.0.0.1:8009):
      Enable Contexts Disable Contexts

       

      Balancer: mycluster,Domain: ,Flushpackets: Off,Flushwait: 10000,Ping: 10000000,Smax: 26,Ttl:
      60000000,Status: NOTOK,Elected: 0,Read: 0,Transferred: 0,Connected: 0,Load: -1
      ===================================================================

       

      "Status: NOTOK". And on backend jboss server, executing run.sh -c all, it outputs following error.

       

      ========================
      04:16:26,188 ERROR [org.jboss.modcluster.mcmp.impl.DefaultMCMPHandler] Error [MEM: MEM: Can't read node: {4}] sending command STATUS to proxy /xxx.xxx.xxx.xxx:6666, configuration will be reset
      ========================

       

      Httpd proxy port is 6666, and proxyList is xxx.xxx.xxx.xxx:6666 .
      How can I repair the httpd and jboss server ?

       

       

       

      Apache httpd.conf is,
      ========= begin httpd.conf ========================
      <Location /mod_cluster-manager>
      SetHandler mod_cluster-manager
      Order deny,allow
      Deny from all
      Allow from all
      </Location>

       

      Listen xxx.xxx.xxx.xxx:6666
      <VirtualHost xxx.xxx.xxx.xxx:6666>
         <Directory />
            Order deny,allow
            Deny from all
            Allow from all
         </Directory>

       

      KeepAliveTimeout 60
      MaxKeepAliveRequests 0

       

      ManagerBalancerName mycluster
      ServerAdvertise off
      AdvertiseFrequency 5
      CreateBalancers 0

       

      </VirtualHost>

       

      ========= end httpd.conf ========================

       

       

       

      JBOSS mod_cluster-jboss-beans.xml file is
      $JBOSS_HOME/server/all/deploy/mod_cluster.sar/META-INF/mod_cluster-jboss-beans.xml
      ========= mod_cluster-jboss-beans.xml ========================
          <!-- Comma separated list of address:port listing the httpd servers
               where mod_cluster is running. -->
         <!-- <property name="proxyList">${jboss.mod_cluster.proxyList:}</property> -->
          <property name="proxyList">xxx.xxx.xxx.xxx:6666</property>

       

          <!-- URL prefix to send with commands to mod_cluster.  Default is no prefix. -->
          <!--property name="proxyURL"></property-->
      ========= mod_cluster-jboss-beans.xml ========================

       

      Thanks and regards,
      Kenji

        • 1. Re: Mod_cluster can't read status in non-multicast environment
          jfclere
          04:16:26,188 ERROR [org.jboss.modcluster.mcmp.impl.DefaultMCMPHandler] Error [MEM: MEM: Can't read node: {4}] sending command STATUS to proxy /xxx.xxx.xxx.xxx:6666, configuration will be reset

          It seems you get the node more than once in the mod_cluser table {4} you have 4 nodes?

           

          Try to remove the

          ServerAdvertise off
          AdvertiseFrequency 5

          From the httpd.conf file.

          • 2. Re: Mod_cluster can't read status in non-multicast environment

            Thank you.

             

            >> 04:16:26,188 ERROR [org.jboss.modcluster.mcmp.impl.DefaultMCMPHandler] Error [MEM: MEM: Can't read node: {4}] sending command STATUS

            >> to proxy /xxx.xxx.xxx.xxx:6666, configuration will be reset

             

            > It seems you get the node more than once in the mod_cluser table {4} you have 4 nodes?

             

            No, I have only a Mod_cluster httpd server and a JBossWeb-Tomcat server.

            I don't know why mod_cluster table has 4 nodes.

             

            > Try to remove the

            > ServerAdvertise off
            > AdvertiseFrequency 5

            > From the httpd.conf file.

             

            I removed these, and retried.

            However, I got same error messages.

             

            23:55:38,335 ERROR [org.jboss.modcluster.mcmp.impl.DefaultMCMPHandler] Error [MEM: MEM: Can't read node: {4}] sending command

            STATUS to proxy /xx.xxx.xxx.xxx:6666, configuration will be reset

             

            Mod_Cluster version is,

              - LB: mod_cluster-1.1.0.CR1

              - AP: jboss-6.0.0.20100216-M2

             

            And I checked these in Amazon EC2, I opened almost all network ports, but it cannot get status.

            Any ideas ?

            • 3. Re: Mod_cluster can't read status in non-multicast environment
              jfclere

              Set httpd loglevel to debug and post the log (error_log) to the forum.

              • 4. Re: Mod_cluster can't read status in non-multicast environment

                Thank you.

                 

                I set loglevel debug, and  watched error_log.

                I noticed following error.

                  [Tue May 11 03:39:54 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to 127.0.0.1:8009 (127.0.0.1) failed

                 

                ajp couldn't connect 127.0.0.1.

                 

                I checket netstat, 8009 port isn't opened.

                 

                On the mod_cluster server, I just installed jboss httpd server.

                Is it neccessary to install jboss tomcat into the mod_cluster (LoadBalancer) server ?

                 

                I checked following page, and configured it.

                On the same server, I didn't install jboss AS.

                 

                ------------ error_log -------------------------

                [Tue May 11 03:34:30 2010] [info] Init: Generating temporary DH parameters (512/
                1024 bits)
                [Tue May 11 03:34:30 2010] [info] Init: Initializing (virtual) servers for SSL
                [Tue May 11 03:34:30 2010] [info] mod_ssl/2.2.15 compiled against Server: Apache
                /2.2.15, Library: OpenSSL/0.9.8k
                [Tue May 11 03:34:30 2010] [warn] No JkLogFile defined in httpd.conf. Using defa
                ult /opt/jboss/httpd/httpd/logs/mod_jk.log
                [Tue May 11 03:34:30 2010] [warn] No JkShmFile defined in httpd.conf. Using defa
                ult /opt/jboss/httpd/httpd/logs/jk-runtime-status
                [Tue May 11 03:34:30 2010] [notice] Advertise initialized for process 1287
                [Tue May 11 03:34:30 2010] [debug] proxy_util.c(1818): proxy: grabbed scoreboard
                slot 0 in child 1290 for worker proxy:reverse
                [Tue May 11 03:34:30 2010] [debug] proxy_util.c(1914): proxy: initialized worker
                0 in child 1290 for (*) min=0 max=25 smax=25
                [Tue May 11 03:34:30 2010] [debug] proxy_util.c(1818): proxy: grabbed scoreboard
                slot 0 in child 1293 for worker proxy:reverse
                [Tue May 11 03:34:30 2010] [debug] proxy_util.c(1837): proxy: worker proxy:rever
                se already initialized
                [Tue May 11 03:34:30 2010] [debug] proxy_util.c(1914): proxy: initialized worker
                0 in child 1293 for (*) min=0 max=25 smax=25
                [Tue May 11 03:34:30 2010] [debug] proxy_util.c(1818): proxy: grabbed scoreboard
                slot 0 in child 1294 for worker proxy:reverse OK
                [Tue May 11 03:38:55 2010] [debug] mod_proxy_cluster.c(609): update_workers_node
                starting
                [Tue May 11 03:38:55 2010] [debug] mod_proxy_cluster.c(411): add_balancer_node:
                Create balancer balancer://mycluster
                [Tue May 11 03:38:55 2010] [debug] mod_proxy_cluster.c(228): Created: worker for
                ajp://127.0.0.1:8009
                [Tue May 11 03:38:55 2010] [debug] mod_proxy_cluster.c(329): proxy: initialized
                worker 1 in child 1290 for (127.0.0.1) min=0 max=25 smax=25
                [Tue May 11 03:38:55 2010] [debug] mod_proxy_cluster.c(387): Created: worker for
                ajp://127.0.0.1:8009 1 (status): 1
                [Tue May 11 03:38:55 2010] [debug] mod_proxy_cluster.c(624): update_workers_node
                done
                [Tue May 11 03:38:55 2010] [debug] proxy_util.c(2011): proxy: ajp: has acquired
                connection for (127.0.0.1)
                [Tue May 11 03:38:55 2010] [debug] proxy_util.c(2067): proxy: connecting ajp://1
                27.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:38:55 2010] [debug] proxy_util.c(2193): proxy: connected ajp://12
                7.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:38:55 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket
                created to connect to 127.0.0.1
                [Tue May 11 03:38:55 2010] [error] (111)Connection refused: proxy: ajp: attempt
                to connect to 127.0.0.1:8009 (127.0.0.1) failed
                [Tue May 11 03:38:56 2010] [debug] mod_proxy_cluster.c(609): update_workers_node
                starting
                [Tue May 11 03:38:56 2010] [debug] mod_proxy_cluster.c(411): add_balancer_node:
                Create balancer balancer://mycluster
                [Tue May 11 03:38:56 2010] [debug] mod_proxy_cluster.c(228): Created: worker for
                ajp://127.0.0.1:8009
                [Tue May 11 03:38:56 2010] [debug] mod_proxy_cluster.c(329): proxy: initialized
                worker 1 in child 1388 for (127.0.0.1) min=0 max=25 smax=25
                [Tue May 11 03:38:56 2010] [debug] mod_proxy_cluster.c(387): Created: worker for
                ajp://127.0.0.1:8009 1 (status): 129
                [Tue May 11 03:38:56 2010] [debug] mod_proxy_cluster.c(624): update_workers_node
                done
                [Tue May 11 03:38:56 2010] [debug] mod_proxy_cluster.c(609): update_workers_node
                starting
                [Tue May 11 03:38:56 2010] [debug] mod_proxy_cluster.c(624): update_workers_node
                done
                [Tue May 11 03:39:01 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the
                worker for (127.0.0.1)
                [Tue May 11 03:39:01 2010] [error] proxy: ajp: disabled connection for (127.0.0.
                1)
                [Tue May 11 03:39:04 2010] [debug] mod_manager.c(1578): manager_trans STATUS (/)
                [Tue May 11 03:39:04 2010] [debug] mod_manager.c(2115): manager_handler STATUS (
                7.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:39:14 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket
                created to connect to 127.0.0.1
                [Tue May 11 03:39:14 2010] [error] (111)Connection refused: proxy: ajp: attempt
                to connect to 127.0.0.1:8009 (127.0.0.1) failed
                [Tue May 11 03:39:14 2010] [error] ap_proxy_connect_backend disabling worker for
                (127.0.0.1)
                [Tue May 11 03:39:14 2010] [debug] mod_proxy_cluster.c(795): proxy_cluster_try_p
                ingpong: can't connect to backend
                [Tue May 11 03:39:14 2010] [debug] proxy_util.c(2029): proxy: ajp: has released
                connection for (127.0.0.1)
                [Tue May 11 03:39:14 2010] [debug] mod_proxy_cluster.c(1466): proxy_cluster_isup
                : pingpong failed
                [Tue May 11 03:39:14 2010] [debug] mod_manager.c(2158): manager_handler STATUS
                OK
                [Tue May 11 03:39:16 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the
                worker for (127.0.0.1)
                [Tue May 11 03:39:16 2010] [error] proxy: ajp: disabled connection for (127.0.0.
                1)
                [Tue May 11 03:39:21 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the
                worker for (127.0.0.1)
                [Tue May 11 03:39:21 2010] [error] proxy: ajp: disabled connection for (127.0.0.
                1)
                [Tue May 11 03:39:24 2010] [debug] mod_manager.c(1578): manager_trans STATUS (/)
                [Tue May 11 03:39:24 2010] [debug] mod_manager.c(2115): manager_handler STATUS (
                [Tue May 11 03:39:34 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket created to connect
                to 127.0.0.1
                [Tue May 11 03:39:34 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to
                127.0.0.1:8009 (127.0.0.1) failed
                [Tue May 11 03:39:34 2010] [error] ap_proxy_connect_backend disabling worker for (127.0.0.1)
                [Tue May 11 03:39:34 2010] [debug] mod_proxy_cluster.c(795): proxy_cluster_try_pingpong: can't connect to backend
                [Tue May 11 03:39:34 2010] [debug] proxy_util.c(2029): proxy: ajp: has released connection for (127.0.0.1)
                [Tue May 11 03:39:34 2010] [debug] mod_proxy_cluster.c(1466): proxy_cluster_isup: pingpong failed
                [Tue May 11 03:39:34 2010] [debug] mod_manager.c(2158): manager_handler STATUS  OK
                [Tue May 11 03:39:36 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:39:36 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:39:41 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:39:41 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:39:44 2010] [debug] mod_manager.c(1578): manager_trans STATUS (/)
                [Tue May 11 03:39:44 2010] [debug] mod_manager.c(2115): manager_handler STATUS (/) processing: "JVMRoute=127.0.0.1%3A8009%3Ajboss.web&Load=100"
                [Tue May 11 03:39:44 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:39:44 2010] [debug] proxy_util.c(1955): proxy: ajp: worker for (127.0.0.1) has been marked for retry
                [Tue May 11 03:39:44 2010] [debug] proxy_util.c(2011): proxy: ajp: has acquired connection for (127.0.0.1)
                [Tue May 11 03:39:44 2010] [debug] proxy_util.c(2067): proxy: connecting ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:39:44 2010] [debug] proxy_util.c(2193): proxy: connected ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:39:44 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket created to connect to 127.0.0.1
                [Tue May 11 03:39:44 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to 127.0.0.1:8009 (127.0.0.1) failed
                [Tue May 11 03:39:44 2010] [error] ap_proxy_connect_backend disabling worker for (127.0.0.1)
                [Tue May 11 03:39:44 2010] [debug] mod_proxy_cluster.c(795): proxy_cluster_try_pingpong: can't connect to backend
                [Tue May 11 03:39:44 2010] [debug] proxy_util.c(2029): proxy: ajp: has released connection for (127.0.0.1)
                [Tue May 11 03:39:44 2010] [debug] mod_proxy_cluster.c(1466): proxy_cluster_isup: pingpong failed
                [Tue May 11 03:39:44 2010] [debug] mod_manager.c(2158): manager_handler STATUS  OK
                [Tue May 11 03:39:46 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:39:46 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:39:51 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:39:51 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:39:54 2010] [debug] mod_manager.c(1578): manager_trans STATUS (/)
                [Tue May 11 03:39:54 2010] [debug] mod_manager.c(2115): manager_handler STATUS (/) processing: "JVMRoute=127.0.0.1%3A8009%3Ajboss.web&Load=100"
                [Tue May 11 03:39:54 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:39:54 2010] [debug] proxy_util.c(1955): proxy: ajp: worker for (127.0.0.1) has been marked for retry
                [Tue May 11 03:39:54 2010] [debug] proxy_util.c(2011): proxy: ajp: has acquired connection for (127.0.0.1)
                [Tue May 11 03:39:54 2010] [debug] proxy_util.c(2067): proxy: connecting ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:39:54 2010] [debug] proxy_util.c(2193): proxy: connected ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:39:54 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket created to connect to 127.0.0.1
                [Tue May 11 03:39:54 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to 127.0.0.1:8009 (127.0.0.1) failed
                [Tue May 11 03:39:54 2010] [error] ap_proxy_connect_backend disabling worker for (127.0.0.1)
                [Tue May 11 03:39:54 2010] [debug] mod_proxy_cluster.c(795): proxy_cluster_try_pingpong: can't connect to backend
                [Tue May 11 03:39:54 2010] [debug] proxy_util.c(2029): proxy: ajp: has released connection for (127.0.0.1)
                [Tue May 11 03:39:54 2010] [debug] mod_proxy_cluster.c(1466): proxy_cluster_isup: pingpong failed
                [Tue May 11 03:39:54 2010] [debug] mod_manager.c(2158): manager_handler STATUS  OK
                [Tue May 11 03:39:56 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:39:56 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:40:01 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:01 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:40:04 2010] [debug] mod_manager.c(1578): manager_trans STATUS (/)
                [Tue May 11 03:40:04 2010] [debug] mod_manager.c(2115): manager_handler STATUS (/) processing: "JVMRoute=127.0.0.1%3A8009%3Ajboss.web&Load=100"
                [Tue May 11 03:40:04 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:04 2010] [debug] proxy_util.c(1955): proxy: ajp: worker for (127.0.0.1) has been marked for retry
                [Tue May 11 03:40:04 2010] [debug] proxy_util.c(2011): proxy: ajp: has acquired connection for (127.0.0.1)
                [Tue May 11 03:40:04 2010] [debug] proxy_util.c(2067): proxy: connecting ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:04 2010] [debug] proxy_util.c(2193): proxy: connected ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:04 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket created to connect to 127.0.0.1
                [Tue May 11 03:40:04 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to 127.0.0.1:8009 (127.0.0.1) failed
                [Tue May 11 03:40:04 2010] [error] ap_proxy_connect_backend disabling worker for (127.0.0.1)
                [Tue May 11 03:40:04 2010] [debug] mod_proxy_cluster.c(795): proxy_cluster_try_pingpong: can't connect to backend
                [Tue May 11 03:40:04 2010] [debug] proxy_util.c(2029): proxy: ajp: has released connection for (127.0.0.1)
                [Tue May 11 03:40:04 2010] [debug] mod_proxy_cluster.c(1466): proxy_cluster_isup: pingpong failed
                [Tue May 11 03:40:04 2010] [debug] mod_manager.c(2158): manager_handler STATUS  OK
                [Tue May 11 03:40:06 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:06 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:40:11 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:11 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:40:14 2010] [debug] mod_manager.c(1578): manager_trans STATUS (/)
                [Tue May 11 03:40:14 2010] [debug] mod_manager.c(2115): manager_handler STATUS (/) processing: "JVMRoute=127.0.0.1%3A8009%3Ajboss.web&Load=100"
                [Tue May 11 03:40:14 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:14 2010] [debug] proxy_util.c(1955): proxy: ajp: worker for (127.0.0.1) has been marked for retry
                [Tue May 11 03:40:14 2010] [debug] proxy_util.c(2011): proxy: ajp: has acquired connection for (127.0.0.1)
                [Tue May 11 03:40:14 2010] [debug] proxy_util.c(2067): proxy: connecting ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:14 2010] [debug] proxy_util.c(2193): proxy: connected ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:14 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket created to connect to 127.0.0.1
                [Tue May 11 03:40:14 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to 127.0.0.1:8009 (127.0.0.1) failed
                [Tue May 11 03:40:14 2010] [error] ap_proxy_connect_backend disabling worker for (127.0.0.1)
                [Tue May 11 03:40:14 2010] [debug] mod_proxy_cluster.c(795): proxy_cluster_try_pingpong: can't connect to backend
                [Tue May 11 03:40:14 2010] [debug] proxy_util.c(2029): proxy: ajp: has released connection for (127.0.0.1)
                [Tue May 11 03:40:14 2010] [debug] mod_proxy_cluster.c(1466): proxy_cluster_isup: pingpong failed
                [Tue May 11 03:40:14 2010] [debug] mod_manager.c(2158): manager_handler STATUS  OK
                [Tue May 11 03:40:16 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:16 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:40:21 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:21 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:40:24 2010] [debug] mod_manager.c(1578): manager_trans STATUS (/)
                [Tue May 11 03:40:24 2010] [debug] mod_manager.c(2115): manager_handler STATUS (/) processing: "JVMRoute=127.0.0.1%3A8009%3Ajboss.web&Load=100"
                [Tue May 11 03:40:24 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:24 2010] [debug] proxy_util.c(1955): proxy: ajp: worker for (127.0.0.1) has been marked for retry
                [Tue May 11 03:40:24 2010] [debug] proxy_util.c(2011): proxy: ajp: has acquired connection for (127.0.0.1)
                [Tue May 11 03:40:24 2010] [debug] proxy_util.c(2067): proxy: connecting ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:24 2010] [debug] proxy_util.c(2193): proxy: connected ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:24 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket created to connect to 127.0.0.1
                [Tue May 11 03:40:24 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to 127.0.0.1:8009 (127.0.0.1) failed
                [Tue May 11 03:40:24 2010] [error] ap_proxy_connect_backend disabling worker for (127.0.0.1)
                [Tue May 11 03:40:24 2010] [debug] mod_proxy_cluster.c(795): proxy_cluster_try_pingpong: can't connect to backend
                [Tue May 11 03:40:24 2010] [debug] proxy_util.c(2029): proxy: ajp: has released connection for (127.0.0.1)
                [Tue May 11 03:40:24 2010] [debug] mod_proxy_cluster.c(1466): proxy_cluster_isup: pingpong failed
                [Tue May 11 03:40:24 2010] [debug] mod_manager.c(2158): manager_handler STATUS  OK
                [Tue May 11 03:40:26 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:26 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:40:31 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:31 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:40:34 2010] [debug] mod_manager.c(1578): manager_trans STATUS (/)
                [Tue May 11 03:40:34 2010] [debug] mod_manager.c(2115): manager_handler STATUS (/) processing: "JVMRoute=127.0.0.1%3A8009%3Ajboss.web&Load=100"
                [Tue May 11 03:40:34 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:34 2010] [debug] proxy_util.c(1955): proxy: ajp: worker for (127.0.0.1) has been marked for retry
                [Tue May 11 03:40:34 2010] [debug] proxy_util.c(2011): proxy: ajp: has acquired connection for (127.0.0.1)
                [Tue May 11 03:40:34 2010] [debug] proxy_util.c(2067): proxy: connecting ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:34 2010] [debug] proxy_util.c(2193): proxy: connected ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:34 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket created to connect to 127.0.0.1
                [Tue May 11 03:40:34 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to 127.0.0.1:8009 (127.0.0.1) failed
                [Tue May 11 03:40:34 2010] [error] ap_proxy_connect_backend disabling worker for (127.0.0.1)
                [Tue May 11 03:40:34 2010] [debug] mod_proxy_cluster.c(795): proxy_cluster_try_pingpong: can't connect to backend
                [Tue May 11 03:40:34 2010] [debug] proxy_util.c(2029): proxy: ajp: has released connection for (127.0.0.1)
                [Tue May 11 03:40:34 2010] [debug] mod_proxy_cluster.c(1466): proxy_cluster_isup: pingpong failed
                [Tue May 11 03:40:34 2010] [debug] mod_manager.c(2158): manager_handler STATUS  OK
                [Tue May 11 03:40:36 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:36 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)

                [Tue May 11 03:40:41 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:41 2010] [error] proxy: ajp: disabled connection for (127.0.0.1)
                [Tue May 11 03:40:44 2010] [debug] mod_manager.c(1578): manager_trans STATUS (/)
                [Tue May 11 03:40:44 2010] [debug] mod_manager.c(2115): manager_handler STATUS (/) processing: "JVMRoute=127.0.0.1%3A8009%3Ajboss.web&Load=100"
                [Tue May 11 03:40:44 2010] [debug] proxy_util.c(1949): proxy: ajp: retrying the worker for (127.0.0.1)
                [Tue May 11 03:40:44 2010] [debug] proxy_util.c(1955): proxy: ajp: worker for (127.0.0.1) has been marked for retry
                [Tue May 11 03:40:44 2010] [debug] proxy_util.c(2011): proxy: ajp: has acquired connection for (127.0.0.1)
                [Tue May 11 03:40:44 2010] [debug] proxy_util.c(2067): proxy: connecting ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:44 2010] [debug] proxy_util.c(2193): proxy: connected ajp://127.0.0.1:8009/ to 127.0.0.1:8009
                [Tue May 11 03:40:44 2010] [debug] proxy_util.c(2444): proxy: ajp: fam 2 socket created to connect to 127.0.0.1
                [Tue May 11 03:40:44 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to 127.0.0.1:8009 (127.0.0.1) failed

                • 5. Re: Mod_cluster can't read status in non-multicast environment
                  jfclere

                  [Tue May 11 03:39:54 2010] [error] (111)Connection refused: proxy: ajp: attempt to connect to 127.0.0.1:8009 (127.0.0.1) failed

                   

                  ajp couldn't connect 127.0.0.1.

                   

                  I checket netstat, 8009 port isn't opened.

                  It seems jboss is running on another box and the AJP connector use the address localhost that can't work.

                  • 6. Re: Mod_cluster can't read status in non-multicast environment

                    > It seems jboss is running on another box and the AJP connector use the address localhost that can't work.

                     

                    Thanks.

                    But I still don't understand the meaning.

                    How can I make it ?

                     

                    By the way,

                    I configured httpd and Java side in a box. ( in one node)

                    After that, I checked mod_cluster-manager page, I got following messages.

                     

                    --------------------------

                    mod_cluster/1.1.x

                    Auto Refresh show DUMP output show INFO output

                    Node 127.0.0.1:8009:jboss.web (ajp://127.0.0.1:8009):

                    Enable Contexts Disable Contexts
                    Balancer: mycluster,Domain: ,Flushpackets: Off,Flushwait: 10000,Ping: 10000000,Smax: 26,Ttl: 60000000,Status: OK,Elected: 0,Read: 0,Transferred: 0,Connected: 0,Load: 100

                    -----------------------------

                     

                    Like this, the status became OK.

                    But this is in a box, and I'd like to configure httpd and ap which are separated.

                     

                    Thanks and regards,

                    Kenji

                    • 7. Re: Mod_cluster can't read status in non-multicast environment
                      brian.stansberry

                      When you start the JBoss instances, use the -b switch to cause them to bind their sockets to a different address than localhost/127.0.0.1. If you don't use the -b switch, by default all sockets are bound to localhost. The httpd server running on a different machine obviously can't communicate with such a socket.

                      1 of 1 people found this helpful
                      • 8. Re: Mod_cluster can't read status in non-multicast environment

                        Thank you.

                         

                        > When you start the JBoss instances, use the -b switch to cause them to bind their sockets to a different address than localhost/127.0.0.1. If you don't use the -b switch, by default all sockets are bound to localhost. The httpd server running on a different machine obviously can't communicate with such a socket.

                         

                        I run following cmd on backend java-side server.

                        run.sh -c all -b xxx.xxx.xxx.xxx

                         

                        And I got following information on mod-manager page.

                        --------

                        mod_cluster/1.1.x

                        Auto Refresh

                        show DUMP output

                        show INFO output

                        Node xxx.xxx.xxx.xxx:8009:jboss.web (ajp://xxx.xxx.xxx.xxx:8009):

                        Enable Contexts

                        Disable Contexts

                        Balancer: mycluster,Domain: ,Flushpackets: Off,Flushwait: 10000,Ping: 10000000,Smax: 26,Ttl: 60000000,Status: OK,Elected: 0,Read: 0,Transferred: 0,Connected: 0,Load: 100

                        --------

                         

                        It seems that it could get the status.

                         

                         

                        However, I tried to access http://<mod_cluster ip>/ ,

                        mod_cluster doesn't foward it to backend jbossweb.

                        I just saw "It works!" apache httpd page.

                         

                        I'd like mod_cluster to foward it.

                        Is it necessary to add ProxyPass into httpd.conf ?

                        Otherwise, how can I config it ?

                         

                        httpd.conf is following

                        ---------- /opt/jboss/httpd/httpd/conf/httpd.conf --------

                        <Location /mod_cluster-manager>
                        SetHandler mod_cluster-manager
                        Order deny,allow

                        Allow from all
                        </Location>

                         

                        # ProxyPass /balancer-manager !

                         

                        <VirtualHost xxx.xxx.xxx.xxx:6666>
                           <Directory />
                              Order deny,allow
                              Deny from all
                              Allow from xx.xxx.xxx.
                              Allow from xx.xxx.xxx.
                           </Directory>

                         

                        KeepAliveTimeout 60
                        MaxKeepAliveRequests 0

                         

                        ManagerBalancerName mycluster
                        # ServerAdvertise off
                        # AdvertiseFrequency 2
                        CreateBalancers 0

                         

                        </VirtualHost>

                        -----------------------------------------------------------------

                         

                        Thanks and regard,

                        Kenji

                        • 9. Re: Mod_cluster can't read status in non-multicast environment
                          jfclere

                          Check excludedContexts. It seems you don't have any context deployed or all are in the excludedContexts list.

                          1 of 1 people found this helpful
                          • 10. Re: Mod_cluster can't read status in non-multicast environment

                            > Check #excludedContexts. It seems you don't have any context deployed or all are in the excludedContexts list.

                             

                            Thanks.

                            I add "all" in mod_cluster-jboss-beans.xml, like this.

                            ------

                            <property name="excludedContexts">${jboss.mod_cluster.excludedContexts:ROOT,admin-console,invoker,jbossws,jmx-console,juddi,web-console,all}</property>

                            ------

                             

                            And I can access jbosswd and juddi pages .etc.

                            But how can I access them from mod_cluster to them.

                             

                            For example,

                              - JBOSS WS page url : http://<JBOSS AP IP>:8080/jbossws  ( Access succeeded )

                              - httpd mod_cluster url : ???  ( http://<httpd mod_cluster IP>/jbossws ?? )

                             

                            I'd like to know the url.

                            And if I mistook some configurations, please let me know.

                             

                            Thanks and regards,

                            Kenji

                            • 11. Re: Mod_cluster can't read status in non-multicast environment
                              jfclere

                              try:

                              <property name="excludedContexts">${jboss.mod_cluster.excludedContexts:ROOT,admin-console,invoker,jmx-console,juddi,web-console}</property>

                              • 12. Re: Mod_cluster can't read status in non-multicast environment

                                > try:

                                > <property name="excludedContexts">${jboss.mod_cluster.excludedContexts:ROOT,admin-console,invoker,jmx-console,juddi,web-console}</property>

                                 

                                I tried this, and I succeeded to access from mod_cluster httpd to jbossws.

                                I got following message.

                                 

                                  23:50:25,059 INFO  [org.jboss.web.tomcat.service.deployers.TomcatDeployment] deploy, ctxPath=/jbossws

                                 

                                But strange to say, after run.sh -c all -b <ip address>, and Jboss AS started,

                                I waited for deployment completed, but 5min, 10min, 15min, I could not get above message.

                                 

                                And finally, deployment was executied at a moment's notice after I access http://<jboss ap ip>:8080/jbossws directly by using Web Browser.

                                I don't know why deployment has not started until I access Jboss AP directly.

                                 

                                Anyway, thank you for your help!

                                Kenji