L.S.,
Could you try to set the logging level to DEBUG to see if this gives you
more information about what's going on? Any other
errors/warnings/suspicious messages in the log to help us tackle this
one? Did you change anything in the flow configuration (I see this
exchange is using JCA flow)?
Concerning your previous message: not sure if that is the cause of your
problems, but since you are extending/using PojoSupport, so you probably
need to do an additional super.init() to properly initialize it.
Gert
BM wrote:
I still haven't managed to get this to work, despite trying to use the
lightweight ftp component instead of servicemix-ftp, and packaging it up
into the lightweight container.
I thought that I'd try a 2-way communication between 2 pojos while I hoped
somebody might have some ideas. Interestingly I'm getting the following
exception when pojo 1 sends to pojo 2 :
17:27:09,562 | ERROR |
pool-flow.jca.org.apache.servicemix.jca.{http://servicemix.apache.org/nftest}nfreceiver:nfreceiver-thread-2
| ActiveMQEndpointWorker | vemq.ra.ActiveMQEndpointWorker 245 | Endpoint
connection to JMS broker failed: Could not connect to broker URL:
tcp://localhost:61616. Reason: java.net.ConnectException: Connection
refused: connect
I've seen a post elsewhere suggesting that JMS problems could cause the
original not-initialised error.
I've disabled my firewall, and because I'm not on a network I've removed the
discoveryUri="multicast://default" from the following line in activemq.xml
:-
<transportConnector uri="tcp://localhost:61616"
discoveryUri="multicast://default"/>
as suggested in the comments.
Using a socket tracer I see 15 sockets have been made to port 61616, but
they're all in TIME_WAIT closing state after I try my pojo-pojo test. If I
don't run the test there are 15 connected sockets to port 61616, so there
isn't a problem with the port already in use etc.
I'm running java 1.5, servicemix 3.1.2 on a Windows XP SP2 box.
I'm at my wits end now - any ideas would really be appreciated.
Cheers,
BM
