Ask Your Question
0

live management no longer shows any nodes

asked 2013-02-28 14:48:02 -0500

Ash gravatar image

updated 2013-02-28 16:28:10 -0500

llowder gravatar image

For some unknown reason, live management stop showing all nodes, including the puppet master node. Puppet runs are still reporting ok. mcollective logs are showing they're connecting ok and I can mco ping all nodes and run agents on them from the master. I've tried restarting the mcollective service on the clients and even restarted the master server, but no joy.

Puppet Enterprise 2.7 on RedHat.

edit retag flag offensive close merge delete

Comments

What changed? Is there anything in any of the logs? Is there free space on the disk?

llowder gravatar imagellowder ( 2013-02-28 16:27:01 -0500 )edit

There is nothing in any of the puppet logs that indicates a problem and all of the filesystems have plenty of free space.

Can't pinpoint when they stopped showing ...(more)

Ash gravatar imageAsh ( 2013-03-06 12:48:04 -0500 )edit

All nodes are part of the default group which has the pecompliance, peaccounts & pe_mcollective classes.

Ash gravatar imageAsh ( 2013-03-10 17:04:01 -0500 )edit

4 Answers

Sort by ยป oldest newest most voted
1

answered 2013-04-06 02:58:31 -0500

jonn gravatar image

There is similar problem reported in PE 2.8.0 - there's a 2.8.1 release coming next week.

edit flag offensive delete link more
0

answered 2013-06-28 09:09:34 -0500

I had the same issue and it turns out that firewall was cutting my stomp connections. I've experienced those entries in mcollective.log:

`on_miscerr' Unexpected error on connection stomp://mcollective@puppet:61613: receive failed: Connection timed out

I've set registerinterval to 600 in mcollective/server.cfg and it solved my issue.

edit flag offensive delete link more
0

answered 2013-03-08 08:17:36 -0500

Daenney gravatar image

As far as I'm aware for all features of PE Console to work properly every node needs to have at least the pemcollective, pecompliance and pe_accounts modules coupled to them and be classified with at least the 'default' group.

These are normally automatically added to every host through Dashboard's External Node Classifier. Classifying every node in de default group is done through the default-add-all-nodes cron which gets installed by the Dashboard package.

The default group has the three pe_* modules assigned to it and then through the fact that PE uses Dashboard as an external node ... (more)

edit flag offensive delete link more
0

answered 2013-04-04 11:59:16 -0500

Hello,

As I'm trying puppet enterprise 2.8, after a flawless installation, I have the same problem. Everything is working OK but only the live management web interface is not showing my 4 nodes. Sometimes it does and some time it doesn't. When it does, and I try making some actions with the live management, nothing happens.

All the clients have pe-mcollective running (and are olso in the defualt group and have the pecompliant + pemcollective + pe_accounts classes) and the master server has the activemq service running also. I forgot to mention that the master is on ... (more)

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

Stats

Asked: 2013-02-28 14:48:02 -0500

Seen: 1,205 times

Last updated: Jun 28 '13