Sounds like you need to enable multicasting. This would be a VM/hypervisor configuration issue.
Filip > -----Original Message----- > From: Madhav Bhargava [mailto:unmarsh...@gmail.com] > Sent: Friday, June 29, 2012 10:04 AM > To: users@tomcat.apache.org > Subject: Re: Multicast fails when mcastBindAddress is explicitly set > > Hi All, > > Ok we got resolution for the below exception. The problem was that both > IPV4 and IPv6 addresses were enabled for the multihome machine. We > switched > to IPv6 addresses and the issue was no longer there. However there is > still > one issue: > > With machines on different hypervisors the multicast traffic seems to be > blocked. VM's on different Hypervisors are not able to get presence or > any > other message from each other. So neither the discovery works nor inter > node communication because there is no knowledge of the other VMs > > Best Regard, > Madhav > > On Fri, Jun 29, 2012 at 5:58 PM, Madhav Bhargava > <unmarsh...@gmail.com>wrote: > > > Hi All, > > > > We are using Apache Tribes 7.0.2. We use it for node discovery and p2p > > communication. > > We are currently running into a problem where the discovery fails on > > multihomed machines (multiple IP's). We were not sure to which IP the > > multicast bind address was getting bound to, so we thought of > explicitly > > binding the interface via "mcastBindAddress" property. However when we > set > > this property then we get the following exception: > > > > Exception occured: java.io.IOException: Invalid argument; No faulty > > members identified.org.apache.catalina.tribes.ChannelException: > > java.io.IOException: Invalid argument; No faulty members identified. > > > > at > > > org.apache.catalina.tribes.group.ChannelCoordinator.internalStart(Channe > lCoordinator.java:178) > > > > at > > > org.apache.catalina.tribes.group.ChannelCoordinator.start(ChannelCoordin > ator.java:99) > > > > at > > > org.apache.catalina.tribes.group.ChannelInterceptorBase.start(ChannelInt > erceptorBase.java:162) > > > > at > > > org.apache.catalina.tribes.group.interceptors.MessageDispatchInterceptor > .start(MessageDispatchInterceptor.java:153) > > > > at > > > org.apache.catalina.tribes.group.ChannelInterceptorBase.start(ChannelInt > erceptorBase.java:162) > > > > at > > > org.apache.catalina.tribes.group.GroupChannel.start(GroupChannel.java:41 > 9) > > > > at > > > com.sap.it.gizmos.diag.TribesConfigurator.run(TribesConfigurator.java:10 > 9) > > > > at > > > java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471) > > > > at > > java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334) > > > > at java.util.concurrent.FutureTask.run(FutureTask.java:166) > > > > at > > > java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.jav > a:1110) > > > > at > > > java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.ja > va:603) > > > > at java.lang.Thread.run(Thread.java:782) > > > > Caused by: java.io.IOException: Invalid argument > > > > at java.net.PlainDatagramSocketImpl.send(Native Method) > > > > at java.net.DatagramSocket.send(DatagramSocket.java:675) > > > > at > > > org.apache.catalina.tribes.membership.McastServiceImpl.send(McastService > Impl.java:503) > > > > at > > > org.apache.catalina.tribes.membership.McastServiceImpl.send(McastService > Impl.java:480) > > > > at > > > org.apache.catalina.tribes.membership.McastServiceImpl.start(McastServic > eImpl.java:269) > > > > at > > > org.apache.catalina.tribes.membership.McastService.start(McastService.ja > va:386) > > > > at > > > org.apache.catalina.tribes.group.ChannelCoordinator.internalStart(Channe > lCoordinator.java:167) > > > > ... 12 more > > > > So we wrote a simple test program (attached) which fails on multi-home > > machines. We also wrote another test program where we just used simple > > java.net.MulticastSocket, set the multicast interface (using > setInterface) > > to one of the interfaces and tried to send a Datagram packet and it > was > > able to send. > > > > So now we wonder: > > > > 1. How do you explicitly set the multicast interface on the group > channel > > in apache tribes? > > 2. I assume that tcpListenHost is the IP address that gets advertised > when > > it joins the group and mcastBindAddress is the interface used to send > out > > messages over a multicast socket. Is my assumption right? > > > > Any help/pointers would be greatly appreciated. > > > > Best Regards, > > Madhav > > > > > > -- > > When I tell the truth, it is not for the sake of convincing those who > do > > not know it, but for the sake of defending those that do > > > > > > -- > When I tell the truth, it is not for the sake of convincing those who do > not know it, but for the sake of defending those that do --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
