[rabbitmq-discuss] Clustering not working for some connections

Thu Oct 21 17:08:39 BST 2010

  I just noticed that on my tests the producers are also getting
blocked (we are using a java client and basicPublish()).

  Trying to list_consumers also becomes unresponsive at this point.

  Any help would be really appreciated.

  Thank you,

--
Ivan Sanchez

On Oct 21, 3:55 pm, Ivan Sanchez <s4nc... at gmail.com> wrote:
>   Hi all,
>
>   We are trying to run a cluster of 2 rabbitmq machines on Amazon EC2
> and although it runs fine for a little while, at some stage it stops
> working only for messages where producer and consumer are connected to
> different nodes. At this point, "rabbitmqctl list_connections" becomes
> completely unresponsive, as well as trying to restart the servers. The
> only option is kill -9 all erlang process and start them again.
>
>   rabbitmqctl status shows:
>
> Status of node rabbit at rabbit1 ...
> [{running_applications,
>      [{rabbit_management,"RabbitMQ Management Console","2.1.1"},
>       {webmachine,"webmachine","1.7.0"},
>       {amqp_client,"RabbitMQ AMQP Client","2.1.1"},
>       {rabbit,"RabbitMQ","2.1.0"},
>       {os_mon,"CPO  CXC 138 46","2.2.5"},
>       {sasl,"SASL  CXC 138 11","2.1.9"},
>       {rabbit_mochiweb,"RabbitMQ Mochiweb Embedding","2.1.1"},
>       {mochiweb,"MochiMedia Web Server","1.3"},
>       {crypto,"CRYPTO version 1","1.6.4"},
>       {inets,"INETS  CXC 138 49","5.3"},
>       {mnesia,"MNESIA  CXC 138 12","4.4.13"},
>       {stdlib,"ERTS  CXC 138 10","1.16.5"},
>       {kernel,"ERTS  CXC 138 10","2.13.5"}]},
>  {nodes,[{disc,[rabbit at rabbit1,rabbit at rabbit2]}]},
>  {running_nodes,[rabbit at rabbit2,rabbit at rabbit1]}]
> ...done.
>
> Status of node rabbit at rabbit2 ...
> [{running_applications,
>      [{rabbit_management,"RabbitMQ Management Console","2.1.1"},
>       {webmachine,"webmachine","1.7.0"},
>       {amqp_client,"RabbitMQ AMQP Client","2.1.1"},
>       {rabbit,"RabbitMQ","2.1.0"},
>       {os_mon,"CPO  CXC 138 46","2.2.5"},
>       {sasl,"SASL  CXC 138 11","2.1.9"},
>       {rabbit_mochiweb,"RabbitMQ Mochiweb Embedding","2.1.1"},
>       {mochiweb,"MochiMedia Web Server","1.3"},
>       {crypto,"CRYPTO version 1","1.6.4"},
>       {inets,"INETS  CXC 138 49","5.3"},
>       {mnesia,"MNESIA  CXC 138 12","4.4.13"},
>       {stdlib,"ERTS  CXC 138 10","1.16.5"},
>       {kernel,"ERTS  CXC 138 10","2.13.5"}]},
>  {nodes,[{disc,[rabbit at rabbit1,rabbit at rabbit2]}]},
>  {running_nodes,[rabbit at rabbit1,rabbit at rabbit2]}]
> ...done.
>
> On the logs of rabbit2, the only error I see some of these:
>
> =ERROR REPORT==== 21-Oct-2010::14:40:47 ===
> exception on TCP connection <0.19069.0> from 88.211.55.18:13580
> {bad_header,<<"<policy-">>}
>
>   Other information:
>   - The hostnames (rabbit1, rabbit2) are defined in /etc/hosts on both
> machines using their private IP, and consumers access them through a
> DNS round-robin to their public IP
>   - Both machines use NODENAME=rabbit@<host> on /etc/rabbitmq/
> rabbitmq.conf
>   - Cluster is defined in /etc/rabbitmq/rabbitmq.config using
> {cluster_nodes, ['rabbit at rabbit1','rabbit at rabbit2']}
>   - We are using RabbitMQ 2.1.0 and Erlang R13B04 (erts-5.7.5)
> [source] [64-bit] [smp:2:2] [rq:2] [async-threads:0] [hipe] [kernel-
> poll:false]
>
>   Any ideas of what can be wrong?
>
> --
> Ivan Sanchez
>
> _______________________________________________
> rabbitmq-discuss mailing list
> rabbitmq-disc... at lists.rabbitmq.comhttps://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss