[chef] Re: Re: Cryptic error message on first run of deploy_branch

[Loïc Antoine-Gombeaud < >]

To further your answer: it would be understandable that resource names
are evaluated at an earlier point of the run than their options, and
before the variable node itself is (set if nil).
Looks like a possible bug, I'll look at the code and report my finds here.

On 12/20/12, Jeffrey Hulten 
<
 >
 wrote:
> If I had to guess I would say that node['plinga-admintool'] wasn't set as a
> Hash for some reason. Once you look up node['plinga-admintool']['app_dir']
> it fails.
> --
> Jeffrey Hulten
> Principal Consultant at Automated Labs, LLC
> 
 
>   206-853-5216
> Skype: jeffhulten
>
> On Dec 20, 2012, at 7:36 AM, Cassiano Leal wrote:
>
>> This error means that you tried to index (using method []) a Nil object.
>> Usually it means that what you thought was going to return an Array or a
>> Hash returned nil instead.
>>
>> This:
>>
>>     NoMethodError: deploy_branch[/var/admintool]
>>
>> means that deploy_branch is nil, meaning that either:
>>
>> - It is supposed to be nil; or
>> - It had not yet been set when it was called
>>
>> On a side note, are you sure that deploy_branch is supposed to be a Hash,
>> rather than, say, a method? And if so, is this Hash supposed to have a key
>> '/var/admintool'? I'm just guessing about the semantics here since I'm not
>> familiar with the rest of the cookbook.
>>
>> Cheers,
>> --
>> Cassiano Leal
>>
>> On Thursday, December 20, 2012 at 09:47, Loïc Antoine-Gombeaud wrote:
>>
>>> Ohai there,
>>>
>>> I'm getting the following error message when I use the deploy resource
>>> for the first time on a clean Vagrant machine:
>>>
>>> FATAL: NoMethodError: deploy_branch[/var/admintool]
>>> (plinga-admintool::deploy line 14) had an error: NoMethodError: undefined
>>> method `[]' for nil:NilClass
>>>
>>> And here's the relevant recipe snippet:
>>>
>>> deploy_branch node['plinga-admintool']['app_dir'] do
>>>   user                        "root"
>>>   group                       "root"
>>>   migrate                     false
>>>   purge_before_symlink([])
>>>   create_dirs_before_symlink([])
>>>   symlink_before_migrate({})
>>>   symlinks({})
>>>   repository                  File.join(repo, branch)
>>>   revision                    node['plinga-admintool']['svn_revision']
>>>   scm_provider                Chef::Provider::Subversion
>>>   svn_username                node['plinga-admintool']['svn_username']
>>>   svn_password                node['plinga-admintool']['svn_password']
>>>   action                      :deploy
>>> end
>>>
>>> Has anyone had a similar or related issue before? The only resource I
>>> could find is this IRC log (search for dbalatero), and... it doesn't
>>> really help, since no solution was suggested there.
>>> I'm at a loss regarding how to debug, since I've triple-checked my code,
>>> all hints or pointers greatly appreciated! Otherwise my next step will be
>>> to hard-code logs in the deploy provider, but avoiding that pain would be
>>> better :)
>>>
>>> More info:
>>> - failed-run-data.json contains all the correct attributes (at least
>>> concerning this deployment)
>>> - any subsequent chef-client run on the same machine works perfectly.
>>>
>>> Cheers and thanks in advance,
>>>
>>> --
>>> Loic ANTOINE-GOMBEAUD
>>> IT contact & DevOps Engineer
>>>
>>>
>>> Plinga GmbH | Saarbrücker Straße 20/21 | 10405 Berlin | Germany
>>> E-Mail: 
>>> 
 
>>>  | skype: loic.plinga
>>> Cell: +49 (0) 160 922 86753
>>>
>>> www.plinga.com
>>>
>>> Geschäftsführer: Johannes Kreibohm, Florian Schmidt-Amelung
>>> Eingetragen beim Amtsgericht Charlottenburg, HRB 119994
>>>
>>
>
>


-- 
*Loic ANTOINE-GOMBEAUD*
IT contact & DevOps Engineer


Plinga GmbH | Saarbrücker Straße 20/21 | 10405 Berlin | Germany
E-Mail: 

 
 | skype: loic.plinga
Cell: +49 (0) 160 922 86753

www.plinga.com

Geschäftsführer: Johannes Kreibohm, Florian Schmidt-Amelung
Eingetragen beim Amtsgericht Charlottenburg, HRB 119994