qpid-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Carl Trieloff <cctriel...@redhat.com>
Subject Re: Testing qpidd on Fedora 7
Date Thu, 25 Oct 2007 00:04:07 GMT

ok, yes I see the issue.. it is different versions of the wire protocol.

We build on every check-in.  Builds that all the tests pass are posted 
at http://rhm.et.redhat.com/download/

If you pick up the packages with *-0.2-*.rpm the latest of cpp broker, 
cpp client, java client and python client will all work together. I'll 
get this pushed to yum repo. Let us know if you find any issues and so 
we can fix.

Carl.


James Cooper wrote:
> Sure thing.  My cursory testing today wasn't very fruitful, but that's 
> largely because I don't know what I'm doing!
>
> Not sure if this is useful info, but here's what I found:
>
> Server:  qpidd - 0.2-5.fc7.i386  
> Client: qpid-java-1.0-incubator-M1-bin-client
>
> Java code:
>
> public static void main(String argv) throws Exception {
>     // this blocks forever:
>     AMQConnection conn = new AMQConnection(" 
> amqp://guest:guest@client/test?brokerlist='tcp://turbo.bitmechanic.com:5672'");
>     // We never get here:
>     System.out.println("Got connection!");
> }
>
> Output on Java side:
>
> main 2007-10-24 16:12:07,631 INFO [apache.qpid.client.AMQConnection] 
> Connection:amqp://guest:guest@client 
> /test?brokerlist='tcp://turbo.bitmechanic.com:5672'
> main 2007-10-24 16:12:07,728 WARN 
> [qpid.client.transport.TransportConnection ] Using Mina NIO
> main 2007-10-24 16:12:07,744 INFO 
> [qpid.client.transport.SocketTransportConnection] send-buffer-size = 
> 32768
> main 2007-10-24 16:12:07,744 INFO 
> [qpid.client.transport.SocketTransportConnection] recv-buffer-size = 32768
> main 2007-10-24 16:12:07,788 INFO 
> [qpid.client.transport.SocketTransportConnection] Attempting 
> connection to turbo.bitmechanic.com/209.40.199.163:5672 
> <http://turbo.bitmechanic.com/209.40.199.163:5672>
> AnonymousIoService-3 2007-10-24 16:12:07,938 WARN 
> [apache.qpid.framing.AMQDataBlockDecoder] AMQDataBlockDecoder does not 
> handle frame type 9
>
> Output on server side:
>
> [root@turbo log]# qpidd -t
> WARNING: Persistence not enabled, no recovery of queues or messages.
> Listening on port 5672...
> INIT [0x98642c0]
> SENT [0x98642c0]: Frame[channel=0; request(id=1,mark=0): 
> ConnectionStart: versionMajor=8; versionMinor=0; serverProperties={}; 
> mechanisms=PLAIN; locales=en_US]
>
> Not sure if the C++ broker and Java M1 client are designed to interop 
> at this point.  That's sort of what it looks like -- some wire 
> protocol mismatch.
>
> any ideas?  or is this just user error?
>
> -- James
>
> On 10/24/07, *Carl Trieloff* <cctrieloff@redhat.com 
> <mailto:cctrieloff@redhat.com>> wrote:
>
>
>     James,
>
>     I expect we will be pushing an update soon.. If you want to help
>     do some
>     of the shake
>     down testing on the next release, shoot and I will see if I can get an
>     update pushed in
>     the next few days.
>
>     Carl.
>
>
>     James Cooper wrote:
>     > Hi guys,
>     >
>     > Just stumbled on your project today.  Looking for something a
>     bit easier for
>     > my SAs to install than ActiveMQ and wanted interop between
>     Python/Java.
>     > This could be the ticket.
>     >
>     > Looks like the users list is just getting spun up.
>     >
>     > First question: Is it productive for me to start doing some
>     testing against
>     > the qpidd that Fedora 7 ships with?  I just did a naive "yum
>     install qpidd"
>     > and it fires up.  Since you guys are moving pretty quickly I'm
>     not sure if
>     > reports I make about that build will be useful for you.
>     >
>     > Are there plans to release patch releases to the yum repository
>     servers?
>     >
>     > thanks,
>     >
>     > -- James
>     >
>     >
>
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message