Ask Your Question

puppet agent fails to run automatically

asked 2015-04-27 21:34:08 -0600

Awmusic12635 gravatar image

updated 2015-04-27 21:51:17 -0600

I have two nodes besides the puppet master listed in my puppet master. The two are currently listed as unresponsive. It appears the puppet agents are failing to run automatically.

In my logs on the agents it lists:

Pastie link as its a bit long:

However if I run puppet agent -t manually:

   root@ex:/var/log# puppet agent -t
   Info: Retrieving pluginfacts
   Info: Retrieving plugin
   Info: Loading facts
   Info: Caching catalog for
   Info: Applying configuration version '1430058342'
   Notice: Finished catalog run in 1.00 seconds

It runs without issue and then appears responsive in the master. Until it fails to run automatically on its next run.

I have tried clearing the ssl keys on both the master and the nodes however the issue still remains.

This is right from an install from the puppet master:

curl -k | sudo bash

OS is Debian 7

Would appreciate any help that could be provided.

edit retag flag offensive close merge delete


Looks like this is Puppet Enterprise, which I have no experience with. Simple things: Check to make sure you're running ntp on master and both nodes.

JohnsonEarls gravatar imageJohnsonEarls ( 2015-04-27 21:42:46 -0600 )edit

Based on this snippet from the error log -- Could not retrieve file metadata for puppet:// -- make sure the name the agent is talking to your puppet master with is the puppet master's name or an alt name, since your domain is not

JohnsonEarls gravatar imageJohnsonEarls ( 2015-04-27 21:43:56 -0600 )edit

@JohnsonEarls It is trying to connect to the right puppet master installation. I am more confused by why if manually run it works, and automatically it fails.

Awmusic12635 gravatar imageAwmusic12635 ( 2015-04-27 21:48:31 -0600 )edit

In addition this is a fresh installation installed directly off of my puppet master: curl -k | sudo bash

Awmusic12635 gravatar imageAwmusic12635 ( 2015-04-27 21:50:37 -0600 )edit

I'm guessing from the log that P.E. runs the agent via an MCollective invocation. Is it possible that it's MCollective that's having the issue? Or that MCollective is invoking the agent with invalid parameters?

JohnsonEarls gravatar imageJohnsonEarls ( 2015-04-28 13:14:05 -0600 )edit

1 Answer

Sort by ยป oldest newest most voted

answered 2015-04-29 16:47:53 -0600

Awmusic12635 gravatar image

I was able to solve the problem by clearing the certs of all of the nodes out of the master,as well as the master cert and fulling deleting the SSL directory off the agent. I had cleared the agents before but it seemed clearing the master one was required as well.

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools

1 follower


Asked: 2015-04-27 21:34:08 -0600

Seen: 1,335 times

Last updated: Apr 29 '15