RE: Different apps clustering on the same Tomcat?
Thanks, Chuck. Caldarale, Charles R chuck.caldar...@unisys.com wrote .. From: John Tangney [mailto:jo...@jdtangney.com] Is it possible to get two apps in the same Tomcat instance to exchange session info, in a cluster? If so, how? Not sure what you're asking. Apologies if I was not clear. We have two machines, each with a Tomcat, forming a cluster. An app named gh runs on each cluster node. We want to add a second app to one of the Tomcats (call it gh2 it's just a clone of gh) that synchronizes sessions with gh on all of the nodes. Multiple apps in one Tomcat cannot share a session, since that's prohibited by the servlet spec. I think you've answered my question. So gh2 cannot synchronize sessions with gh on the same cluster node or not. You can create a bean from a class in a shared library to hold data needed by both apps. Ah, this sounds promising! Where would this shared library live so that both apps can talk to it? Just for background... In an ideal world, we'd just buy another server and add it to the (existing) cluster. But we are looking for ways to get the benefit of another cluster node without new hardware. Thanks! --johnt - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Different apps clustering on the same Tomcat?
On Jan 20, 2010, at 10:57 AM, Caldarale, Charles R wrote: From: John Tangney [mailto:jo...@jdtangney.com] Subject: RE: Different apps clustering on the same Tomcat? We have two machines, each with a Tomcat, forming a cluster. An app named gh runs on each cluster node. We want to add a second app to one of the Tomcats (call it gh2 ? it's just a clone of gh) that synchronizes sessions with gh on all of the nodes. Like Chris, I still don't really understand what you're trying to achieve. We use a cluster to do seamless deploys. We use Nginx on the front end to direct traffic to one or other of the nodes in the cluster. To deploy, we bring up the new version of the app in a one node, tell Nginx to send traffic to that node, then bring down the other node. Because sessions are propagated, users never notice that one node has been exchanged for another. We need to add two more nodes that run on Windows (we have platform- specific code, and Nginx sends windows traffic to one of the windows nodes based on URL) but we only have one Windows box. We thought that perhaps we could just run a second instance of the app on the Windows node. From what you said previously, this does not appear to be possible. So we'll be looking new hardware. Thanks for all your help! --johnt - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Different apps clustering on the same Tomcat?
On Jan 20, 2010, at 7:38 PM, Caldarale, Charles R wrote: You can run multiple instances of Tomcat on the same box, and each could have a copy of the app deployed, and each Tomcat could be a separate node in the cluster. That's exactly what we do on Linux, but it fails on Windows due to a bug I have already logged. --johnt - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Different apps clustering on the same Tomcat?
Is it possible to get two apps in the same Tomcat instance to exchange session info, in a cluster? If so, how? Thanks! --johnt - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Running two tomcat servers on same system?
Dean, On Jan 7, 2010, at 1:47 PM, Dean Chester wrote: I am wondering if it would be possible to have two tomcat servers running on the same system Yeah, we do this all the time. All you gave to do is to make sure that all the ports listed in your server.xml are unique. Also is it possible to share shutdown ports of tomcat servers? I am not sure. It's easy enough to give it a try. --johnt - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Error running two tomcats in same cluster on same Windows box
Konstantin, Thanks for your in-depth analysis! I am still on vacation so I have had not had a chance to follow all your references. Here is some log output: INFO: Initializing Coyote HTTP/1.1 on http-8180 Dec 22, 2009 2:17:07 PM org.apache.catalina.startup.Catalina load INFO: Initialization processed in 332 ms Dec 22, 2009 2:17:07 PM org.apache.catalina.core.StandardService start INFO: Starting service Catalina Dec 22, 2009 2:17:07 PM org.apache.catalina.core.StandardEngine start INFO: Starting Servlet Engine: Apache Tomcat/6.0.18 Dec 22, 2009 2:17:07 PM org.apache.catalina.ha.tcp.SimpleTcpCluster start INFO: Cluster is about to start Dec 22, 2009 2:17:07 PM org.apache.catalina.tribes.transport.ReceiverBase bind INFO: Receiver Server Socket bound to:/10.5.1.102:4000 Dec 22, 2009 2:17:07 PM org.apache.catalina.tribes.membership.McastServiceImpl setupSocket INFO: Attempting to bind the multicast socket to /224.0.0.0:45564 Dec 22, 2009 2:17:07 PM org.apache.catalina.tribes.membership.McastServiceImpl setupSocket INFO: Setting multihome multicast interface to:/10.5.1.102 Dec 22, 2009 2:17:07 PM org.apache.catalina.ha.tcp.SimpleTcpCluster start SEVERE: Unable to start cluster. org.apache.catalina.tribes.ChannelException: java.net.SocketException: An operation was attempted on something that is not a socket; No faulty members identified. ... Caused by: java.net.SocketException: An operation was attempted on something that is not a socket On Dec 26, 2009, at 1:34 AM, Konstantin Kolinko wrote: 2) John Tangney, do you see the following one or two INFO messages in Tomcat log: log.info(Attempting to bind the multicast socket to +address +:+port); log.info(Binding to multicast address, failed. Binding to port only.); INFO: Attempting to bind the multicast socket to /224.0.0.0:45564 3) It is possible to create unbound MulticastSocket, by passing a null instead of SocketAddress to its constructor. Maybe the socket.setInterface(mcastBindAddress); call will succeed if we create an unbound socket with new MulticastSocket(null), make setInterface() call, and call bind() later. When I get back to work I will be able to follow up on your research and perhaps try #3 – unless someone else tries it first. If that works, we'll either need to run a custom Tomcat or wait for the patch (assuming it's accepted) is released. Thanks again for your help! ---johnt - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Error running two tomcats in same cluster on same Windows box
We are still experiencing this problem. To summarize: 1) We are using 6.0.18. 2) A single Tomcat works fine and joins the cluster. 3) Two Tomcats on the SAME MACHINE (different ports - duh) but multicasting to DIFFERENT multicast IPs each join their respective clusters. (yes, two clusters.) 4) Two Tomcats on the SAME machine multicasting to the SAME multicast IP and we get An operation was attempted on something that is not a socket when we launch the second tomcat. 5) The exact setup works fine on RHEL and Mac OS X This is NOT the 6.0.20 regression (https://issues.apache.org/bugzilla/show_bug.cgi?id=47308 ) – we're using 6.0.18. This is NOT an issue with Windows getting confused by the bind attribute – it works fine in cases 2 3 above. Could it be that windows won't allow two apps to multicast to the same address at the same time? Can anyone help with this issue? --johnt - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Error running two tomcats in same cluster on same Windows box
Thanks for your response, Filip! On Dec 23, 2009, at 6:05 AM, Filip Hanik - Dev Lists wrote: sorry, ignore previous post, Yeah, I had the same reaction to 224.0.0.0 but it works when one Tomcat is running, so I assumed it's OK. it's failing on this operation: socket.setInterface(mcastBindAddress); There is a small chance that Window's doesn't allow multicast sockets to be bound to an IP try removing the bind attribute No, that can't be it, because when I run just one tomcat on that windows machine (with the bind – without it Tomcat will bind to the wrong NIC which is itself a problem) everything works just fine. It's only when I run two tomcats at the same time with the same Membership ... Could it be that Windows (or some layer in between) is holding a socket open? Is that what An operation was attempted on something that is not a socket means? Thanks! --johnt - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Re: Error running two tomcats in same cluster on same Windows box
On Dec 23, 2009, at 12:19 PM, Filip Hanik - Dev Lists wrote: On 12/23/2009 11:42 AM, John Tangney wrote: it's failing on this operation: socket.setInterface(mcastBindAddress); There is a small chance that Window's doesn't allow multicast sockets to be bound to an IP try removing the bind attribute No, that can't be it did you try it? Yes, I did try that. With the bind attribute, a SINGLE tomcat binds to the correct NIC and the cluster works perfectly. It's only when we have TWO tomcats both on the same machine that we have the problem. Without the bind attribute, Tomcat binds to the wrong NIC which prevents the other nodes from getting the multicast packets. I could experiment with another windows box to see if I can run 2 tomcats without the bind attribute. (I can't do that on the machine that's experiencing the prob because it's a production box.) But even if that works, we stuck, because we have to use the 2nd NIC for clustering. Could it be that a socket is not being closed? Thanks! --johnt
Re: Error running two tomcats in same cluster on same Windows box
On Dec 23, 2009, at 2:36 PM, Filip Hanik - Dev Lists wrote: On 12/23/2009 01:46 PM, John Tangney wrote: On Dec 23, 2009, at 12:19 PM, Filip Hanik - Dev Lists wrote: On 12/23/2009 11:42 AM, John Tangney wrote: it's failing on this operation: socket.setInterface(mcastBindAddress); There is a small chance that Window's doesn't allow multicast sockets to be bound to an IP try removing the bind attribute No, that can't be it did you try it? Yes, I did try that. With the bind attribute, a SINGLE tomcat binds to the correct NIC and the cluster works perfectly. It's only when we have TWO tomcats both on the same machine that we have the problem. That, my friend, is a totally different problem. I tried to be complete in the subject line and in the body of my post. My apologies if I was not 100% clear. A regression in in 6.0.20 only https://issues.apache.org/bugzilla/show_bug.cgi?id=47308 We're using 6.0.18. (Yes, I know I should have mentioned that in my original posting.) --johnt - To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
Error running two tomcats in same cluster on same Windows box
We run two tomcats on the same Windows Server 2008 box. They have their server.xml tweaked so that they use unique ports. Here's a diff: $ diff server.xml /cygdrive/c/Program\ Files/Apache\ Software\ Foundation/Tomcat\ 6.0/conf/server.xml 22c22 Server port=8105 shutdown=SHUTDOWN --- Server port=8005 shutdown=SHUTDOWN 67c67 Connector port=8180 protocol=HTTP/1.1 --- Connector port=8080 protocol=HTTP/1.1 69c69 redirectPort=8543 / --- redirectPort=8443 / 88c88 Connector port=8109 protocol=AJP/1.3 redirectPort=8543 / --- Connector port=8009 protocol=AJP/1.3 redirectPort=8443 / 100c100 Engine name=WinBox_Temp defaultHost=localhost jvmRoute=WinBox_temp --- Engine name=WinBox defaultHost=localhost jvmRoute=WinBox If we have one of the tomcats broadcast its multicast packets on 228.0.0.0 while the second broadcasts on 224.0.0.0, everything works fine and they each cluster with other nodes that use those multicast addresses. (Think staging and prod clusters) But when we want two tomcats on the same machine to join the _same_ cluster (224.0.0.0), we get this in the log file: SEVERE: Unable to start cluster. org.apache.catalina.tribes.ChannelException: java.net.SocketException: An operation was attempted on something that is not a socket; No faulty members identified. at org.apache.catalina.tribes.group.ChannelCoordinator.internalStart(ChannelCoordinator.java:169) at org.apache.catalina.tribes.group.ChannelCoordinator.start(ChannelCoordinator.java:97) at org.apache.catalina.tribes.group.ChannelInterceptorBase.start(ChannelInterceptorBase.java:149) at org.apache.catalina.tribes.group.interceptors.MessageDispatchInterceptor.start(MessageDispatchInterceptor.java:147) at org.apache.catalina.tribes.group.ChannelInterceptorBase.start(ChannelInterceptorBase.java:149) at org.apache.catalina.tribes.group.ChannelInterceptorBase.start(ChannelInterceptorBase.java:149) at org.apache.catalina.tribes.group.GroupChannel.start(GroupChannel.java:407) at org.apache.catalina.ha.tcp.SimpleTcpCluster.start(SimpleTcpCluster.java:669) at org.apache.catalina.core.ContainerBase.start(ContainerBase.java:1035) at org.apache.catalina.core.StandardEngine.start(StandardEngine.java:443) at org.apache.catalina.core.StandardService.start(StandardService.java:516) at org.apache.catalina.core.StandardServer.start(StandardServer.java:710) at org.apache.catalina.startup.Catalina.start(Catalina.java:578) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25) at java.lang.reflect.Method.invoke(Method.java:597) at org.apache.catalina.startup.Bootstrap.start(Bootstrap.java:288) at org.apache.catalina.startup.Bootstrap.main(Bootstrap.java:413) Caused by: java.net.SocketException: An operation was attempted on something that is not a socket at java.net.PlainDatagramSocketImpl.socketSetOption(Native Method) at java.net.PlainDatagramSocketImpl.setOption(PlainDatagramSocketImpl.java:299) at java.net.MulticastSocket.setInterface(MulticastSocket.java:420) at org.apache.catalina.tribes.membership.McastServiceImpl.setupSocket(McastServiceImpl.java:206) at org.apache.catalina.tribes.membership.McastServiceImpl.init(McastServiceImpl.java:173) at org.apache.catalina.tribes.membership.McastServiceImpl.init(McastServiceImpl.java:169) at org.apache.catalina.tribes.membership.McastService.start(McastService.java:356) at org.apache.catalina.tribes.group.ChannelCoordinator.internalStart(ChannelCoordinator.java:154) ... 18 more Here's the juicy part of server.xml: Membership className=org.apache.catalina.tribes.membership.McastService address=224.0.0.0 bind=10.5.1.102 port=45564 frequency=500 dropTime=3000/ Naturally, I have googled the exception, but the only hits I have found go on about corrupted winsock registry entries, which seems irrelevant since everything works when the two tomcats use separate multicast addresses. Note that 10.5.1.102 is a statically-assigned IP on this box's second NIC. So... a) Is it possible to run two tomcat nodes in the same cluster on the same windows box? b) If so, what's the magic to keep windows happy? It _appears_ that windows won't let the two tomcats bind to the same multicast address/port. (Same config works fine on Linux.) Thanks! --johnt -- View this message in context: http://old.nabble.com/Error-running-two-tomcats-in-same-cluster-on-same-Windows-box-tp26896489p26896489.html Sent from the
