OpenAMQ

Error message from client: 06:20:31: E: could not connect to localhost

A packet trace shows version mismatch and client seems to misinterpret server response

…/clients/amq_client without any args was used in the test
The rabbitmq release was 1.0.0-alpha with a std install of both OTP and rabbitmq
Client and server were run on the same host
The host runs Fedora Core 6

From an internal loopback packet capture tool:

Client > Server (127.0.0.1:50739 > 127.0.0.1:5672)
41 4d 51 50 01 01 00 09 | AMQP….

Client < Server (127.0.0.1:50739 < 127.0.0.1:5672)
41 4d 51 50 01 01 08 00 | AMQP….

Client > Server (127.0.0.1:50739 > 127.0.0.1:5672)
01 00 00 00 00 00 38 00 0a 00 32 01 f5 2d 50 65 | ……8…2..-Pe
65 72 20 73 65 6e 74 20 6f 76 65 72 73 69 7a 65 | er sent oversize
64 20 66 72 61 6d 65 20 28 31 33 34 32 32 34 33 | d frame (1342243
30 38 30 20 3e 20 34 30 39 36 29 00 00 00 00 ce | 080 > 4096)…..

The rabbitmq server then sends a TCP reset.

Am I doing some simple mistake here?
Or is there a way to get an OpenAMQ client to negotiate a 0.8 connection?

Thx,
Rolf