[ https://issues.jboss.org/browse/ISPN-1465?page=com.atlassian.jira.plugin.... ] Paul Ferraro reassigned ISPN-1465: ---------------------------------- Assignee: Paul Ferraro (was: Galder Zamarreño) -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/ISPN-1465?page=com.atlassian.jira.plugin.... ] Galder Zamarreño commented on ISPN-1465: ---------------------------------------- It's very weird this hasn't got caught in our testsuite, which is why I'm slightly worried about rushing a fix for this in BETA2. I'm probably gonna release BETA2 tomorrow morning CEST time anyway. We're likely to do a BETA3 next Monday anyway with other missing bits in BETA2. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/ISPN-1465?page=com.atlassian.jira.plugin.... ] Paul Ferraro commented on ISPN-1465: ------------------------------------ Perhaps the offset is usually 0? Also the above fix didn't actually fix the IOException in my tests. So, I'm still looking into this as well. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/ISPN-1465?page=com.atlassian.jira.plugin.... ] Paul Ferraro updated ISPN-1465: ------------------------------- Summary: CommandAwareRpcDispatcher message handling doesn't handle buffer offset correctly (was: Unsupported protocol version when handling RPC messages) Assignee: Paul Ferraro (was: Galder Zamarreño) Priority: Minor (was: Blocker) Description: Looking at the CommandAwareRpcDispatcher.handle(Message) (line 139): cmd = (ReplicableCommand) req_marshaller.objectFromBuffer(req.getBuffer(), req.getOffset(), req.getLength()); Message.getBuffer() already returns a copy of the message buffer that takes into account the offset and length. So, it would seem the above code will return a subset of the desired buffer if the offset > 0. Either this code should operate on the raw buffer (i.e. req.getRawBuffer()), or pass 0 as the offset to objectFromBuffer(...). was: When testing the latest BETA1 as well as the latest SNAPSHOT (in anticipation of BETA2), I see the following exceptions in the AS7 testsuite: WARN: Problems unmarshalling remote command from byte buffer java.io.IOException: Unsupported protocol version 172 at org.jboss.marshalling.river.RiverUnmarshaller.start(RiverUnmarshaller.java:1184) at org.infinispan.marshall.jboss.AbstractJBossMarshaller.startObjectInput(AbstractJBossMarshaller.java:113) at org.infinispan.marshall.VersionAwareMarshaller.startObjectInput(VersionAwareMarshaller.java:156) at org.infinispan.marshall.VersionAwareMarshaller.objectFromByteBuffer(VersionAwareMarshaller.java:112) at org.infinispan.marshall.AbstractDelegatingMarshaller.objectFromByteBuffer(AbstractDelegatingMarshaller.java:80) at org.infinispan.remoting.transport.jgroups.MarshallerAdapter.objectFromBuffer(MarshallerAdapter.java:50) at org.infinispan.remoting.transport.jgroups.CommandAwareRpcDispatcher.handle(CommandAwareRpcDispatcher.java:139) at org.jgroups.blocks.RequestCorrelator.handleRequest(RequestCorrelator.java:446) at org.jgroups.blocks.RequestCorrelator.receiveMessage(RequestCorrelator.java:353) at org.jgroups.blocks.RequestCorrelator.receive(RequestCorrelator.java:229) at org.jgroups.blocks.MessageDispatcher$ProtocolAdapter.up(MessageDispatcher.java:556) at org.jgroups.JChannel.up(JChannel.java:718) at org.jgroups.stack.ProtocolStack.up(ProtocolStack.java:1026) at org.jgroups.protocols.pbcast.FLUSH.up(FLUSH.java:481) at org.jgroups.protocols.pbcast.STATE_TRANSFER.up(STATE_TRANSFER.java:178) at org.jgroups.protocols.FRAG2.up(FRAG2.java:181) at org.jgroups.protocols.FlowControl.up(FlowControl.java:400) at org.jgroups.protocols.FlowControl.up(FlowControl.java:418) at org.jgroups.protocols.pbcast.GMS.up(GMS.java:865) at org.jgroups.protocols.pbcast.STABLE.up(STABLE.java:244) at org.jgroups.protocols.UNICAST.up(UNICAST.java:297) at org.jgroups.protocols.pbcast.NAKACK.handleMessage(NAKACK.java:719) at org.jgroups.protocols.pbcast.NAKACK.up(NAKACK.java:547) at org.jgroups.protocols.BARRIER.up(BARRIER.java:126) at org.jgroups.protocols.FD_ALL.up(FD_ALL.java:171) at org.jgroups.protocols.FD_SOCK.up(FD_SOCK.java:273) at org.jgroups.protocols.MERGE2.up(MERGE2.java:208) at org.jgroups.protocols.Discovery.up(Discovery.java:335) at org.jgroups.protocols.TP.passMessageUp(TP.java:1091) at org.jgroups.protocols.TP$IncomingPacket.handleMyMessage(TP.java:1646) at org.jgroups.protocols.TP$IncomingPacket.run(TP.java:1623) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1110) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:603) at java.lang.Thread.run(Thread.java:679) Looking at the CommandAwareRpcDispatcher.handle(Message) (line 139): cmd = (ReplicableCommand) req_marshaller.objectFromBuffer(req.getBuffer(), req.getOffset(), req.getLength()); However, Message.getBuffer() already returns a copy of the message buffer that takes into account the offset and length. So, it would seem the above code will return a subset of the desired buffer if the offset > 0. Either this code should operate on the raw buffer (i.e. req.getRawBuffer()), or pass 0 as the offset to objectFromBuffer(...). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/ISPN-1465?page=com.atlassian.jira.plugin.... ] Paul Ferraro updated ISPN-1465: ------------------------------- Assignee: Galder Zamarreño (was: Paul Ferraro) Git Pull Request: https://github.com/infinispan/infinispan/pull/585 Pull request created. Reassigning back to Galder for review. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/ISPN-1465?page=com.atlassian.jira.plugin.... ] Sanne Grinovero reassigned ISPN-1465: ------------------------------------- Assignee: Paul Ferraro (was: Galder Zamarreño) -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/ISPN-1465?page=com.atlassian.jira.plugin.... ] Sanne Grinovero commented on ISPN-1465: --------------------------------------- sorry for "stealing" the pull request Galder; I didn't know it was meant to you, but I already checked it and merged. Paul if you create a pull request to be managed by a specific person, please add a comment in the pull or we might jump on it ;-) Anyway it looked like it made a lot of sense; actually I have less failures than usual (luck? will check). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira