Puppet users - Nov 2011 - My ssh tunnels get no love from exec

I''m working on a sweet little module to manage ssh tunnels using
Net::SSH and daemons. Before I try to add a service provider in order
to manage daemons, I figured I''d just exec and am running into a very
strange situation.

First, here is a simple proof of concept involving 3 small files:

(Please forgive the overly simplified and abstract examples.. just
trying to keep this as short as possible)

#/tmp/foo.rb:

loop do
  sleep 5
end
#eof

#/tmp/foo_control.rb:

require ''rubygems'' unless defined?(Gem)
require ''daemons''

Daemon.run(''/tmp/foo.rb'')
#eof

Manual runs of ruby /tmp/foo_control.rb start/stop/restart results in
the expected behavior, the (useless) daemon is managable and well
behaved :)

Taking this a bit further, I can involve Puppet like so:

#foo.pp
exec { test_it
  command => "ruby /tmp/foo_control.rb start",
  creates => ''/tmp/foo.rb.pid'',
  user => ''akosmin'',
}

and then I can start the daemon with a simple

sudo puppet apply foo.pp

OK, everything seems great at this point but if I replace foo.rb''s
useless sleep with the following ssh tunnel, the daemon appears to
start but then exits but only when exec''d from Puppet. Here''s
the
updated /tmp/foo.rb:


require ''rubygems''
require ''net/ssh''

Net::SSH.start(''box1.example.com, ENV[''LOGNAME'']) do
|ssh|
  ssh.forward.local(7777, box2.example.com, 22)
  ssh.loop { true }
end

Running this manually (ruby /tmp/foo_control.rb start) from a shell
spawns the daemon, no problem. I can start it, query the status, or
stop it. I''m pointing this out in order to spare anyone from wondering
if the ssh bits are working. The problem comes when I simply exec it
via Puppet (see above). It seems like the daemon starts but then after
a second, it dies. Again, I can run the exact command the exec type is
using and everything works as expected. I''ve added --debug and --trace
and have no additional information to report as the output just tells
me everything is working as expected.

Does anyone have any ideas?

Best,
Adam

-- 
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.

No fix yet but I see a small typo in my original post.
> Daemon.run(''/tmp/foo.rb'')
should be

Daemons.run(''/tmp/foo.rb'')

-- 
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.

For reasons that are clearly beyond me, it''s now clear that /tmp/
foo.rb is being run as root despite the controller script running as
''akosmin''. I don''t get it :/

-- 
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.

Looks like this is related to rvm and the fact that the user''s
environment is not sourced when puppet changes user context.

-- 
You received this message because you are subscribed to the Google Groups
"Puppet Users" group.
To post to this group, send email to puppet-users@googlegroups.com.
To unsubscribe from this group, send email to
puppet-users+unsubscribe@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/puppet-users?hl=en.