thanks Tim

Timothy Bish wrote:
> 
> On Fri, 2009-10-30 at 11:35 -0700, nmittal wrote:
>> ok will change the configuration then.. just to make sure..
>> transport.keepAliveResponseRequired=true or false?
>> 
>> also I am assuming this would need to be done for all connectors
>> (including
>> stomp) ..right?
> 
> It should be set to true, doing so will cause the CMS client to bounce a
> command back which keeps the channel active.  This doesn't apply to
> Stomp as it has no KeepAlive command.
> 
> Regards
> Tim.
> 
>> 
>> thanks
>> Nishant
>> 
>> 
>> Timothy Bish wrote:
>> > 
>> > On Fri, 2009-10-30 at 11:22 -0700, nmittal wrote:
>> >> Tim, thanks!
>> >> Yes I am using 5.3.0 and that change seems to have fixed this issue.
>> >> 
>> >> however, I am now worried about migrating to the new version of Broker
>> >> and
>> >> CPP as I have a ton of existing code that would need changing. are
>> there
>> >> other changes like these that would need to be made in my code?
>> >> 
>> >> thanks again
>> >> Nishant
>> > 
>> > There aren't any that I'm aware of, there are of course slight changes
>> > in the CMS API since 2.x but nothing that should cause you problems.
>> > This issue is a result of a change in the way KeepAlive messages are
>> > sent from the broker in 5.3.0 and will be resolved in the next release
>> > of CMS, for now you just need to either add that option to your
>> client's
>> > URI or change the configuration file for your broker to have the
>> > following openwire connector config:
>> > 
>> > <transportConnectors>
>> >    <transportConnector name="openwire"
>> > uri="tcp://0.0.0.0:61616?transport.keepAliveResponseRequired=true"/>
>> > </transportConnectors>
>> > 
>> >> 
>> >> 
>> >> Timothy Bish wrote:
>> >> > 
>> >> > On Fri, 2009-10-30 at 10:00 -0700, nmittal wrote:
>> >> >> I ported my code from CPP 2.2.5 to CPP 3.0.1 and after a successful
>> >> build
>> >> >> tried to run it. I can see the Queues and Topics created by the
>> code
>> >> but
>> >> >> after some 10 secs or so.. i get the following messages...
>> >> >> 
>> >> >> Exception: DataInputStream::readLong - Reached EOF
>> >> >>         FILE: ./decaf/io/DataInputStream.h, LINE: 378
>> >> >>         FILE: decaf/io/DataInputStream.cpp, LINE: 162
>> >> >>         FILE: activemq/wireformat/openwire/OpenWireFormat.cpp,
>> LINE:
>> >> 275
>> >> >>         FILE: activemq/transport/IOTransport.cpp, LINE: 235
>> >> >> 
>> >> >> Exception: DataInputStream::readLong - Reached EOF
>> >> >>         FILE: ./decaf/io/DataInputStream.h, LINE: 378
>> >> >>         FILE: decaf/io/DataInputStream.cpp, LINE: 162
>> >> >>         FILE: activemq/wireformat/openwire/OpenWireFormat.cpp,
>> LINE:
>> >> 275
>> >> >>         FILE: activemq/transport/IOTransport.cpp, LINE: 235
>> >> >> 
>> >> >> Exception: DataInputStream::readLong - Reached EOF
>> >> >>         FILE: ./decaf/io/DataInputStream.h, LINE: 378
>> >> >>         FILE: decaf/io/DataInputStream.cpp, LINE: 162
>> >> >>         FILE: activemq/wireformat/openwire/OpenWireFormat.cpp,
>> LINE:
>> >> 275
>> >> >>         FILE: activemq/transport/IOTransport.cpp, LINE: 235
>> >> >> 
>> >> >> 
>> >> >> Please note that this happens even when there are no messages sent
>> or
>> >> >> received. This same code worked fine with CMS 2.2.5
>> >> >> 
>> >> >> thanks
>> >> >> Nishant
>> >> > 
>> >> > What version of the broker are you using?  If its 5.3.0 you can try
>> >> > adding wireFormat.MaxInactivityDuration=-1 to you connection URI and
>> >> see
>> >> > if that resolves the issue.
>> >> > 
>> >> > 
>> >> > -- 
>> >> > Tim Bish
>> >> > http://fusesource.com
>> >> > http://timbish.blogspot.com/
>> >> > 
>> >> > 
>> >> > 
>> >> > 
>> >> > 
>> >> 
>> > -- 
>> > Tim Bish
>> > http://fusesource.com
>> > http://timbish.blogspot.com/
>> > 
>> > 
>> > 
>> > 
>> > 
>> 
> -- 
> Tim Bish
> http://fusesource.com
> http://timbish.blogspot.com/
> 
> 
> 
> 
> 

-- 
View this message in context: 
http://old.nabble.com/CMS-3.0.1%3A-Get-exception-after-sometime-%28approx-10secs%29-tp26133585p26136101.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Reply via email to