I believe that those exceptions occur when a client does not disconnect from the broker gracefully.
Verify that your client is making the appropriate close call on the connection. Do you have STOMP based clients that are frequently connecting and disconnecting? Thanks, Paul On Wed, Oct 23, 2013 at 1:56 AM, Rahn Nicholas, Bedag <nicholas.r...@bedag.ch> wrote: > Hi. Hoping that someone might have some insight into what we're seeing in our > activemq.log file. > > We're running ActiveMQ 5.8 and in the activemq.log file we see lots of > messages like this: > > 2013-10-11 09:03:57,212 | WARN | Transport Connection to: > tcp://10.251.128.28:42015 failed: java.io.EOFException | > org.apache.activemq.broker.TransportConnection.Transport | ActiveMQ > Transport: tcp:///10.251.128.28:42015@61616 > > The IP addresses are from our client applications, of which there are at > least 30. The messages come frequently, roughly 4-5 per minute. Often 2 from > the same IP will come within a couple milliseconds of each other. > > Our applications are running fine (sending and receiving) so I'm wondering if > these messages are actually important or if they could be ignored? Why are > they logged at the WARN level? > > I've also been trying to track down the cause of these messages. One reason > I've found is that if there is a connection open when the JVM exits, one of > these log entries will be written. But that does not explain the vast > majority of them because our applications don't exit that frequently. > > Can anyone tell me if these messages point to something that might be amiss > with our setup/config? We're using the openwire transport connector with the > JDBC persistence adapter. I can provide the full config file if that would > be helpful. > > Thanks in advance for any insight. > Nick > >
