autoload puppet/type/domain

[scarby]

Hi

I'm seeing the following when attempting to create a cluster resource using the glassfish module (interestingly enough i have seen this succeed - and the cluster exists and was created by the module)

Error: Could not retrieve catalog from remote server: Error 400 on SERVER: Could not autoload puppet/type/domain: Could not autoload puppet/provider/domain/asadmin: no such file to load -- puppet/provider/asadmin on node vid03-prod-zonza.zonza.prv

have checked and these providers exist (obviouslt) and have been synced to the agent

Any ideas?

Thanks
Adam

[fatmcgav]

@scarby Hmm, you're the second person who's reported that issue, but I've still not been able to replicate... See #34 for other issue.

Can you confirm the following:

• Module version/source
• Puppet version
• OS version
• A sample manifest that produces the above error.

I'll then try and re-produce it again.

[scarby]

module is taken from the master branch of this github repo - i have forked it and introduced a JMScluster provider, but at the moment that is not being called

The error looks slightly different - yet very similar.
Whats interesting is that all my versions are identical to the previous user

(puppet 3.7.1, ubuntu 12.04, ruby 1.8.7)

  class { 'glassfish':
    create_domain => true,
    domain_name   => 'domain1'
  }
  glassfish::create_cluster { 'vidispine':
    cluster_name => 'vidispine',
  }

(i know domain1 is a terrible name but its replicating config that already exists)

Initially i encountered this error, however it seemed intermittent - and would fail only on some runs. However now its pretty consistent

the below shows that the cluster has actually been created at some point

./asadmin list-clusters
Enter admin user name> admin
Enter admin password for user "admin">
vidispine not running
Command list-clusters executed successfully.

Also if i stop calling create cluster everything works

[fatmcgav]

Might be worth rebasing off the develop branch, as there's a fair few changes in there that aren't on the master yet...

I'll try and replicate it again over the next couple of days and come back to you...

I'd be interested in reviewing/adding the JMSCluster provider to this module if necessary aswell :)

Cheers
Gav

[scarby]

Thanks

I'll send you a pull request when i get everything working

Adam

On Wed, Oct 22, 2014 at 5:26 PM, Gavin Williams notifications@github.com
wrote:

Might be worth rebasing off the develop branch, as there's a fair few
changes in there that aren't on the master yet...

I'll try and replicate it again over the next couple of days and come back
to you...

I'd be interested in reviewing/adding the JMSCluster provider to this
module if necessary aswell :)

Cheers
Gav

—
Reply to this email directly or view it on GitHub
#38 (comment)
.

[scarby]

Ok, i have rebased from the develop branch and receive the same errors (seemingly only when i reference defined types such as create cluster)

Adam

[fatmcgav]

Hmm, ok... Possibly gives me something a bit more specific to focus on...

[fatmcgav]

@scarby I've not been able to replicate the issue on a vagrant Ubuntu 12.04 machine running Puppet 3.7.1...

One other question - are you using puppet apply or puppet agent against a master?

[scarby]

having now returned from some time away, it would appear that having rebuilt the puppet master i'm not getting the error again. I'm not entirely certain what caused it :(.

[fatmcgav]

Cheers for letting me know...