I eventually see lots of error messages relating to Consumer tags must be unique in the error logs.
I think that means I'm subscribing to something I already subscribed to. In that situation, does my client get an exception or not? Thanks, Adam On Tue, Mar 24, 2009 at 3:44 PM, Adam Chase <[email protected]> wrote: > Clustering was happening for sure. > > I was running some tests. > > Here is the part from the log before the errors happened. > > Adam > > 2009-mar-24 09:11:04 notice qpid/broker/Broker.cpp:308:virtual void > qpid::broker::Broker::run(): Broker running > 2009-mar-24 09:11:04 info qpid/cluster/Cluster.cpp:521:void > qpid::cluster::Cluster::memberUpdate(qpid::sys::ScopedLock<qpid::sys::Mutex>&): > 12.11.168.192:23403(READY) member update: 12.11.168.192:2 > 3403(member) > 2009-mar-24 09:11:04 notice qpid/cluster/Cluster.cpp:328:void > qpid::cluster::Cluster::configChange(const qpid::cluster::MemberId&, > const std::string&, qpid::sys::ScopedLock<qpid::sys::Mutex>&): 12. > 11.168.192:23403(READY) first in cluster > 2009-mar-24 09:11:07 info qpid/cluster/Cluster.cpp:354:void > qpid::cluster::Cluster::makeOffer(const qpid::cluster::MemberId&, > qpid::sys::ScopedLock<qpid::sys::Mutex>&): 12.11.168.192:23403(OFFER) > s > end update-offer to 11.11.168.192:24956 > 2009-mar-24 09:11:07 info qpid/cluster/Cluster.cpp:429:void > qpid::cluster::Cluster::updateStart(const qpid::cluster::MemberId&, > const qpid::Url&, qpid::sys::ScopedLock<qpid::sys::Mutex>&): 12.11.16 > 8.192:23403(UPDATER) stall for update to 11.11.168.192:24956 at > amqp:tcp:192.168.11.11:5672,tcp:10.32.89.14:5672 > 2009-mar-24 09:11:07 info qpid/client/ConnectionImpl.cpp:130:void > qpid::client::ConnectionImpl::open(): Connecting to > tcp:192.168.11.11:5672 > 2009-mar-24 09:11:07 info qpid/cluster/Cluster.cpp:467:void > qpid::cluster::Cluster::updateOutDone(qpid::sys::ScopedLock<qpid::sys::Mutex>&): > 12.11.168.192:23403(UPDATER) sent update > 2009-mar-24 09:11:07 info qpid/cluster/Cluster.cpp:521:void > qpid::cluster::Cluster::memberUpdate(qpid::sys::ScopedLock<qpid::sys::Mutex>&): > 12.11.168.192:23403(READY) member update: 11.11.168.192:2 > 4956(member) 12.11.168.192:23403(member) > 2009-mar-24 09:11:15 error > qpid/amqp_0_10/SessionHandler.cpp:100:virtual void > qpid::amqp_0_10::SessionHandler::handleIn(qpid::framing::AMQFrame&): > Channel exception: not-attached: receiving Frame[B > Ebe; channel=2; {ExchangeBindBody: queue=AVAILABLE_SLOTS; > exchange=amq.direct; binding-key=AVAILABLE_SLOTS; arguments={}; }]: > channel 2 is not attached (qpid/amqp_0_10/SessionHandler.cpp:79) > > > On Tue, Mar 24, 2009 at 1:06 PM, Gordon Sim <[email protected]> wrote: >> Adam Chase wrote: >>> >>> What does this indicate? >> >> At the most basic level, that a frame was received on a channel to which no >> session is believed to be attached. >> >> E.g. a command was sent after the session was closed (possibly due to an >> exception). >> >> Can you give any more context on what was happening at the time? Was it a >> standalone broker? >> >>> Was this when my client lost a connection? >>> >>> I'm trying to understand something that occurred on my system that is >>> likely a bug of mine. >> >> Are there any other errors in the log? >> >> --------------------------------------------------------------------- >> Apache Qpid - AMQP Messaging Implementation >> Project: http://qpid.apache.org >> Use/Interact: mailto:[email protected] >> >> > --------------------------------------------------------------------- Apache Qpid - AMQP Messaging Implementation Project: http://qpid.apache.org Use/Interact: mailto:[email protected]
