On 2014-12-09 19:01, Mark Eggers wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1On 12/9/2014 9:48 AM, Daniel Mikusa wrote:On Tue, Dec 9, 2014 at 9:16 AM, Théo Chamley <theo...@mley.fr> wrote:On 2014-12-08 21:22, Ameer Mawia wrote:Hi Theo, Since you are using static membership and NOT dynamic(multi-cast) which auto-detect members, my current understanding says that you will have to add entry of all the members of your cluster in each of nodes server.xml. Example: My cluster has two nodes. So to configure these I had add these two members entries in both node's server.xml(since running on the same machine, they have varying port with same ip): <Interceptor className="org.apache.catalina.tribes.group.interceptors. StaticMembershipInterceptor"> <Member className="org.apache.catalina.tribes.membership.StaticMember" port="4000" host="192.168.1.6" uniqueId="{10,1,0,100,0,0,0,0,0,0,0,0,0,0,0,0}"/> <Member className="org.apache.catalina.tribes.membership.StaticMember" port="4001" host="192.168.1.6" uniqueId="{10,1,0,101,0,0,0,0,0,0,0,0,0,0,0,0}"/> Regards, Ameer Mawia On Mon, Dec 8, 2014 at 8:26 PM, Théo Chamley <theo...@mley.fr> wrote: Hello,I am trying to setup a simple Tomcat cluster with static membership. I can't use multicast because I am on a virtualization environment that does not allow it. Debian 7 Tomcat 8.0.14 Oracle JVM 1.8.0_25 Both Tomcat are ok on their own, but I can't seem to make the clustering work: the sessions are not replicated from one to another. Following the official documentation, I wrote this configuration : <Engine name="Catalina" defaultHost="localhost"> <Cluster className="org.apache.catalina.ha.tcp. SimpleTcpCluster"> <Manager className="org.apache.catalina.ha.session. BackupManager" expireSessionsOnShutdown="false" notifyListenersOnReplication="true" mapSendOptions="6"/> <Channel className="org.apache.catalina.tribes.group. GroupChannel"> <Receiver className="org.apache. catalina.tribes.transport.nio.NioReceiver" address="0.0.0.0" port="4110" selectorTimeout="100" maxThreads="6"/> <Interceptor className="org.apache.catalina.tribes.group. interceptors.StaticMembershipInterceptor"> <Member className="org.apache. catalina.tribes.membership.StaticMember" port="4110" host="my.server.1" domain="staging-cluster" uniqueId="{1,2,3,4,5,6,7,8,9, 10,11,12,13,14,15,0}"/> </Interceptor> <Sender className="org.apache.catalina.tribes.transport. ReplicationTransmitter"> <Transport className="org.apache. catalina.tribes.transport.nio.PooledParallelSender"/> </Sender> <Interceptor className="org.apache.catalina.tribes.group. interceptors.TcpFailureDetector"/> <Interceptor className="org.apache.catalina.tribes.group. interceptors.MessageDispatch15Interceptor"/> <Interceptor className="org.apache.catalina.tribes.group. interceptors.ThroughputInterceptor"/> </Channel> <Valve className="org.apache.catalina.ha.tcp. ReplicationValve" filter=".*\.gif|.*\.js|.*\. jpeg|.*\.jpg|.*\.png|.*\.css|. *\.txt"/> <ClusterListener className="org.apache.catalina.ha.session. ClusterSessionListener"/> </Cluster> [...] </Engine> Note: I changed the host and uniqId on the StaticMembershipInterceptor on the other Tomcat. This is not a network problem as I can telnet into the 4110 port from one server to another. Also, by running a tcpdump, I can't see any trafic between the two servers. The Tomcats seem to be doing something, because I have the following in my catalina.out: ****** 08-Dec-2014 15:38:15.309 INFO [main] org.apache.catalina.ha.tcp. SimpleTcpCluster.startInternal Cluster is about to start 08-Dec-2014 15:38:15.312 INFO [main] org.apache.catalina.tribes. transport.ReceiverBase.bind Receiver Server Socket bound to:/0.0.0.0:4110 08-Dec-2014 15:38:15.328 INFO [Thread-5] org.apache.catalina.ha.tcp. SimpleTcpCluster.memberAdded Replication member added:org.apache.catalina.tribes.membership. StaticMember[t cp://my.server.1:4110,my.server.1,4110, alive=0, securePort=-1, UDP Port=-1, id={1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 }, payload={}, command={}, domain={115 116 97 103 105 110 103 45 99 ...(15)}, ] 08-Dec-2014 15:38:15.330 INFO [main] org.apache.catalina.tribes. membership.McastServiceImpl.setupSocket Setting cluster mcast soTimeout to 500 08-Dec-2014 15:38:15.332 INFO [main] org.apache.catalina.tribes. membership.McastServiceImpl.waitForMembers Sleeping for 1000 milliseconds to establish cluster membership, sta rt level:4 08-Dec-2014 15:38:16.155 INFO [Membership-MemberAdded.] org.apache.catalina.ha.tcp.SimpleTcpCluster.memberAdded Replication member added:org.apache.catalina.tribes.membership .MemberImpl[tcp://{0, 0, 0, 0}:4110,{0, 0, 0, 0},4110, alive=1277686, securePort=-1, UDP Port=-1, id={-22 -45 110 -29 21 -22 75 95 -103 86 95 -119 15 48 -17 -27 }, payload={} , command={}, domain={}, ] 08-Dec-2014 15:38:16.259 INFO [Tribes-Task-Receiver-1] org.apache.catalina.tribes.io.BufferPool.getBufferPool Created a buffer pool with max size:104857600 bytes of type: org.apache.catalina.tribes.io. BufferPool15Impl 08-Dec-2014 15:38:16.332 INFO [main] org.apache.catalina.tribes. membership.McastServiceImpl.waitForMembers Done sleeping, membership established, start level:4 08-Dec-2014 15:38:16.335 INFO [main] org.apache.catalina.tribes. membership.McastServiceImpl.waitForMembers Sleeping for 1000 milliseconds to establish cluster membership, start level:8 08-Dec-2014 15:38:17.335 INFO [main] org.apache.catalina.tribes. membership.McastServiceImpl.waitForMembers Done sleeping, membership established, start level:8 ****** Could someone, please, help me finding what I am doing wrong? Thanks, Théo C. ---------------------------------------------------------------------To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.orgFor additional commands, e-mail: users-h...@tomcat.apache.orgHello, Thank you for your answer. Indeed, I need both nodes in the Interceptor, but something else is wrong. I missed it the first time, but the official documentation mentions that the TcpFailureDetector must be above the StaticMembershipInterceptor. I suspect that it is also true of the other Interceptors, but I'm not sure. Here is my conf right now. I made some progress: I now have trafic between the two servers on the port 4110. Its regularity makes my think it's a heartbeat. However, the two Tomcat still do not share any sessions. <Engine name="Catalina" defaultHost="localhost"> <Cluster className="org.apache.catalina.ha.tcp.SimpleTcpCluster" channelStartOptions="3"> <Manager className="org.apache.catalina.ha.session. DeltaManager" expireSessionsOnShutdown="false" notifyListenersOnReplication="true"/> <Channel className="org.apache.catalina.tribes.group. GroupChannel"> <Receiver className="org.apache. catalina.tribes.transport.nio.NioReceiver" address="0.0.0.0" port="4110" selectorTimeout="100" maxThreads="6"/> <Sender className="org.apache.catalina.tribes.transport. ReplicationTransmitter"> <Transport className="org.apache. catalina.tribes.transport.nio.PooledParallelSender"/> </Sender> <Interceptor className="org.apache.catalina.tribes.group. interceptors.TcpPingInterceptor"/> <Interceptor className="org.apache.catalina.tribes.group. interceptors.TcpFailureDetector"/> <Interceptor className="org.apache.catalina.tribes.group. interceptors.MessageDispatch15Interceptor"/> <Interceptor className="org.apache.catalina.tribes.group. interceptors.StaticMembershipInterceptor"> <Member className="org.apache. catalina.tribes.membership.StaticMember" port="4110" host="my.server.1" domain="staging-cluster" uniqueId="{1,2,3,4,5,6,7,8,9, 10,11,12,13,14,15,0}"/> <Member className="org.apache. catalina.tribes.membership.StaticMember" port="4110" host="my.server.2" domain="staging-cluster" uniqueId="{0,1,2,3,4,5,6,7,8, 9,10,11,12,13,14,15}"/> </Interceptor> </Channel> <Valve className="org.apache.catalina.ha.tcp.ReplicationValve" filter=".*\.gif|.*\.js|.*\.jpeg|.*\.jpg|.*\.png|.*\.css|. *\.txt"/> <Valve className="org.apache.catalina.ha.session. JvmRouteBinderValve"/> <ClusterListener className="org.apache.catalina.ha.session. ClusterSessionListener"/> </Cluster> [...] </Engine> Is there some way for me to debug what is happening?Have you turned on debug level logging? Maybe try packages like "org.apache.catalina.ha" and "org.apache.catalina.tribes". From what I remember, it spits out a lot of information. Dan- From my Tomcat 7 cluster experiment (check the documentation to see ifthings have changed in Tomcat 8): # logging.properties # all one line # note 5cluster.org.apache.juli.FileHandler at the end # handlers = 1catalina.org.apache.juli.FileHandler, 2localhost.org.apache.juli.FileHandler, 3manager.org.apache.juli.FileHandler, 4host-manager.org.apache.juli.FileHandler, java.util.logging.ConsoleHandler,5cluster.org.apache.juli.FileHandler # defining the general log information # 5cluster.org.apache.juli.FileHandler.level = FINER 5cluster.org.apache.juli.FileHandler.directory = ${catalina.base}/logs 5cluster.org.apache.juli.FileHandler.prefix = cluster. # cluster logs - have removed farm deployer log configuration # beware the line wrap # org.apache.catalina.tribes.MESSAGES.level = FINE org.apache.catalina.tribes.MESSAGES.handlers = 5cluster.org.apache.juli.FileHandler org.apache.catalina.tribes.level = FINEorg.apache.catalina.tribes.handlers = 5cluster.org.apache.juli.FileHandlerorg.apache.catalina.ha.level = FINE org.apache.catalina.ha.handlers = 5cluster.org.apache.juli.FileHander . . . just my two cents /mde/ -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBAgAGBQJUhzkVAAoJEEFGbsYNeTwtaeUIAJvtpaO//El/lz108vpgGkDZ g8DCIeNyhbekeo/2YI5lhTXTWsybHfwNOQEKgKhAGm76NyklJQT6GwMU6m2EmWJQ U+lfTGce7lX0sBMTKRBL/0VX9DV86f+p+1x0mH05gJRXQy/PfyCp/gwlE6OwLGoH 1Ut+qp6NJ7Bte3Z4SOt7SvqZrTCz4dGRH9+tKA5IwHfynOrJDS62p4L2+DNa2xwG x03V5vdCpaMq1ZpYps+wdBFQgWOfqJwOR7MO3mnP/l9ZSjWy3LGoIXbeY2OEZbKl 0CSBzJZYDmnzywna4YUcy5dnxl7fZJR7yDKKsWoyNiPfGOnkVaMcmQl0FlafBeQ= =rh+S -----END PGP SIGNATURE----- --- This email is free from viruses and malware because avast! Antivirus protection is active. http://www.avast.com --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Hello, Thank you all for your help.Mark, things haven't changed on Tomcat8 for the logging properties, your configuration did work.
In the end, I had two problems: * My client did not have a <distributable/> element in his web.xml * More importantly, the NioReceiver was not well configured.The "address" parameter as defined in the documentation (http://tomcat.apache.org/tomcat-8.0-doc/config/cluster-receiver.html) was more that what I thought. It is not only the binding address for the receiver, it's also the address that will be advertised to the other nodes of the cluster. I configured it to "0.0.0.0", therefore the other node tried to connect to "0.0.0.0:<port>". You can see the problem...
Replacing this address by a routable one solved my problem.Even if it is quite logical, I think the documentation could be clearer on this point.
Thank you again, Théo --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
