jakesgordon / rack-rabbit Goto Github PK

Am I right thinking that RR::Client#request

does not close channel after it received response?

So the temporary queue won't be auto-deleted, as the connection (channel) is still open.

RackRabbit::Subscriber#subscribe uses deprecated bunnys option :ack`.

This causes bunny to output:

[DEPRECATION] `:ack` is deprecated.  Please use `:manual_ack` instead.
[DEPRECATION] `:ack` is deprecated.  Please use `:manual_ack` instead.

Pull request will follow.

ruby 2.2.1

Playing with rack-rabbit.

Tried to send request:

$DEBUG = true

client = RR.new(host: '172.28.128.17', port: 5672, adapter: :bunny)
client.request('my.topic', '/ping', '', { method: :GET })

Got exception:

Exception `IO::EAGAINWaitReadable' at /usr/local/rvm/gems/ruby-2.2.1/gems/bunny-1.7.0/lib/bunny/cruby/socket.rb:45 - Resource temporarily unavailable - read would block

Is it something I should worry about?

Got this in server logs:

[DEPRECATION] `:ack` is deprecated.  Please use `:manual_ack` instead.

I see that there's not much activity on the project. It's really interesting and I'm just curious if there's anyone using it in production?

Given rack app where rack-rabbit is used only as a client library RR.

When I call RR.get it tries to connect to default localhost RabbitMQ.

Is there any shortcut to pass connection configuration to RR without instantiation of RR?

With instantiation it looks like:

client = RR.new(host: '172.28.128.17', port: 5672, adapter: :bunny)
client.get('my.service', '/ping', {})

The goal is to establish connection to RabbitMQ after process fork and reuse already established connection on every incoming HTTP request.