I've created a monster, and I need help killing it.

My setup:

I'm using Rails for some scientific work. I run it on my desktop
machine, draco, and have starling and memcache on both draco and
another machine, lovelace.

Both machines remote mount my home directory, so both have access to
the Rails sources.

I run the Rails server on draco, my machine.

The worker is setup to call a binary located in Rails.root's bin
folder. Unfortunately, this binary malfunctions on lovelace, so I
changed config/workling.yml so it only points to draco. I also
restarted starling and memcached a bunch of times on draco and killed
them on lovelace.

But when I ps ax on lovelace or draco, I get tons and tons of workling
processes:
24199 ?        S      0:21 workling
24215 ?        S      0:20 workling
24231 ?        S      0:21 workling
24243 ?        S      0:20 workling
24247 ?        S      0:20 workling
24251 ?        S      0:20 workling
24254 ?        S      0:01 workling
24258 ?        S      0:21 workling

killall -9 workling doesn't work on these.
$ killall -9 workling
workling: no process killed

'script/workling_client stop' does nothing.

I can kill each PID individually, but that takes forever, and they
just re-spawn eventually.

The biggest problem is that when I tell the worker to run on draco, it
runs on lovelace instead. Even though I changed the config, these
worklings seem to be spawning from some process using the old config.
It's maddening.

I've Googled endlessly. I've read all the docs I can find. I have no
idea how to control which machine runs the workers, or how to stop
them. I can't shut lovelace down because other people are using it. I
can't reboot draco because it's in another state and there's no
guarantee it'll come back up.

It's more like a golem than a worker. How do I stop it? Or control it?

Thanks!

--

You received this message because you are subscribed to the Google Groups "Ruby 
on Rails: Talk" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/rubyonrails-talk?hl=en.


Reply via email to