<div dir="ltr"><div><div>So, I was recently experimenting with 3.0.2 and needing to go back to 2.8.7 for some additional experimentation.<br><br></div>However, RabbitMQ 2.8.7 won't start for me any more. I'm using Erlang R15B01<br>
<br>Here's the console spew for the mq1 node, which I'm unable to start:<br><br>--------<br>Activating RabbitMQ plugins ...<br>8 plugins activated:<br>* amqp_client-0.0.0<br>* mochiweb-2.3.1-rmq0.0.0-gitd541e9a<br>
* rabbitmq_federation-0.0.0<br>* rabbitmq_federation_management-0.0.0<br>* rabbitmq_management-0.0.0<br>* rabbitmq_management_agent-0.0.0<br>* rabbitmq_mochiweb-0.0.0<br>* webmachine-1.9.1-rmq0.0.0-git52e62bc<br><br>{"init terminating in do_boot",{function_clause,[{lists,usort,[['rabbit@mq1'|{['rabbit@mq1','rabbit@play','rabbit@play2'],disc}]],[{file,"lists.erl"},{line,992}]},{rabbit_upgrade,maybe_upgrade_mnesia,0,[{file,"src/rabbit_upgrade.erl"},{line,126}]},{rabbit,prepare,0,[{file,"src/rabbit.erl"},{line,288}]},{init,eval_script,8,[]},{init,do_boot,3,[]}]}}<br>
<br>Crash dump was written to: erl_crash.dump<br>init terminating in do_boot ()<br>--------<br><br>I've blown away both the rabbit@mq1 and rabbit@mq1-plugins-expand
directories multiple times.<br>In the past, blowing away these directories has always worked to get me to a "fresh state" where Rabbit would always start. Is
there some additional places that are referenced?<br><br></div><div>Here's the rabbitmq.config file for mq1:<br><br>--------<br>[<br>{rabbit, [{cluster_nodes, {['rabbit@mq1','rabbit@play','rabbit@play2'], disc}}, {disk_free_limit, 104857600}]},<br>
{mnesia, [{debug, trace}]},<br>{rabbitmq_federation, [{exchanges, [[{type, "topic"}, {upstream_set, "skytap_upstreams"}, {virtual_host, "/"}, {exchange, "skytap"}]]}, <br>{upstream_sets, [{"skytap_upstreams", [[{connection, "upstream-us-west"}, {max_hops, 1}, {exchange, "skytap"}],[{connection, "upstream-us-east"}, {max_hops, 1}, {exchange, "skytap"}]]}]}, <br>
{connections, [{"upstream-us-west", [{username, "guest"}, {ha_policy, "all"}, {host, "mq2"}, {heartbeat, 5}, {password, "guest"}, {port, 5672}]}, {"upstream-us-east", [{username, "guest"}, {ha_policy, "all"}, {host, "mq3"}, {heartbeat, 5}, {password, "guest"}, {port, 5672}]}]}, <br>
{local_nodename, "mq_mpietrek"}]}<br>].<br>--------<br><br></div><div>Just for laughs I've taken out the play and play2 nodes from the config file and I get the same problem, so I don't think it's clustering related.<br>
<br>Thanks,<br><br>Matt<br></div><br></div>