Hi Dikang, Two things, 1) You don't need to use two ports with .8, we don't use openid... if you need more workers, use unicorn... it's your friend and no special sauce required to use it (except it won't work with the chef-server script...
) What I personally did is create an upstart service to handle this, I attached my files, so ideally if you used this... first turn off chef-server or whatever sitting on port 4000... put chef-start.sh in /usr/bin put unicorn.rb in /usr/lib/ruby/gems/1.8/gems/chef-server-0.8.0/ put chef-server.conf in /etc/init then type start chef-server if you want to see what's going on stdout/stderr this configuration drops those in /tmp/unicorn.stdout.log and /tmp/unicorn.stderr.log and sets the workers to 4... adjust as you see fit, but .8 is insanely responsive with 2 workers... 4-8 workers with monit/bluepill and I can't see how you can fail. 2) -c2 failing should be fixed by http://tickets.opscode.com/browse/CHEF-737 are you running a current master? Scott |
Attachment:
chef-start.sh
Description: Binary data
Attachment:
chef-server.conf
Description: Binary data
# What ports/sockets to listen on, and what options for them. listen 4000, :tcp_nodelay => true, :backlog => 1024 # What the timeout for killing busy workers is, in seconds timeout 3600 # Whether the app should be pre-loaded preload_app false # How many worker processes worker_processes 4 stderr_path "/tmp/unicorn.stderr.log" stdout_path "/tmp/unicorn.stdout.log"
On Jan 25, 2010, at 10:39 PM, Dikang Gu wrote: Is this a bug for chef 0.8 on ubuntu 9.10? |
Archive powered by MHonArc 2.6.16.