A couple of things...
* Dont' use starling. in 99% of cases delayed_job is a better
alternative. And much much easier to test/debug/deploy/support
* If you do have to use startling, still don't :) Use Kestrel instead.
Its a drop-in replacement with the same network protocol and works
with workling w/o changing a thing (even "starling_top" script works).
just make sure ports match
* For workling/starling debugging
  * check the port you are running. both by looking at the conf file
and by running 'netstat -plitn'
  * start a console in production mode "./script/console production"
  * in a separate shell run 'starling_top'
  * now manually run some workling job. smth like
"MyWorker.async_do_some_work(blah blah)"
  * run starling_top again and notice the differences.

  * you should see a queue named after your worker and method called.
like queue_my_workers__do_some_work_XXX
  * you should see the QUEUE_total_items number increase.
  * if you see the QUEUE_items number increased and not decreased for
some time it means your worker is not running or not connecting to the
same queue.
  * if you see the _items increase and decrease a moment after, your
queue is working OK (it it works too fast you might only see the
_total_items increase, its OK too :)


On Aug 25, 3:06 am, "[email protected]" <[email protected]> wrote:
> Has anyone has success with Starling & Workling in production
> (ec2onrails)?  I'm trying to deploy it to EC2 and I can't get my
> Workling methods to fire.  I know the biggest gotcha is running
> starling on port 15151 and I'm 99% sure I'm doing this.  I'm even able
> to open up irb on this machine and set/get messages from localhost:
> 151515.
>
> I've been digging to this for 3 days and I think I'm a bit stuck.  I'd
> appreciate any advice or help.
>
> Thanks,
> Jonathan

--~--~---------~--~----~------------~-------~--~----~
SD Ruby mailing list
[email protected]
http://groups.google.com/group/sdruby
-~----------~----~----~----~------~----~------~--~---

Reply via email to