Thanks for bringing up ensure_resources, I completely forgot about that 
one. Add it to the list of things I'd rather see solved more permanently.

On Wednesday, 19 February 2014 14:32:32 UTC+1, Trevor Vaughan wrote:
>
> Sorry for the double post, but I just remembered the new way of doing if 
> defined(). A function was added to stdlib called ensure_resource 
> https://github.com/puppetlabs/puppetlabs-stdlib/blob/master/lib/puppet/parser/functions/ensure_resource.rb
> .
>
> Looking at it, and how it is used, it's just another if defined().
>
> Trevor
>
>
> On Wed, Feb 19, 2014 at 8:28 AM, Trevor Vaughan 
> <[email protected]<javascript:>
> > wrote:
>
>> One of the things that was proposed in the past was allowing different 
>> modules to declare the same resource iff it was declared identically in all 
>> cases.
>>
>> It was shot down as promoting bad practices but so was if defined() and 
>> the fact is that we *need* that because I can't have service { 'foo': 
>> ensure => running } in two modules.
>>
>> Yes, of course the correct way to do things is to abstract out each 
>> portion that is not on your critical path and have parameters that allow 
>> you to selectively manage it thereby alleviating the issues with if 
>> defined(), etc...
>>
>> But, everyone has jobs to do and may not be able to spend much time 
>> refactoring for other users.
>>
>> I *do* think that all Puppet Labs sponsored/approved modules should allow 
>> for the following to be selectively managed/included by a module at a 
>> minimum:
>>
>> * Firewall
>> * PAM
>> * Service
>> * Web
>>
>> In general, I've found that if you make these optional, you stand a 
>> pretty good chance of staying out of people's way and letting them manage 
>> those portions their way while still taking advantage of your module.
>>
>> All that said, I still want to be able to declare multiple identical 
>> resources and have it "just work".
>>
>> Trevor
>>
>>
>> On Wed, Feb 19, 2014 at 6:16 AM, Daniele Sluijters 
>> <[email protected]<javascript:>
>> > wrote:
>>
>>> Hi everyone,
>>>
>>> As I was browsing through a few modules on the Forge I noticed that more 
>>> and more people seem to have taken to the `if ! defined(Package['name']) 
>>> {}` to ensure their module doesn't cause a duplicate resource declaration 
>>> with another module, third party or your own, that is also managing that 
>>> package.
>>>
>>> From a purist perception this should never happen in the first place as 
>>> there are suddenly two modules declaring the same resource. Unfortunately 
>>> this is often what it comes down to when you pull in a few modules from the 
>>> Forge or Github, especially with things like build-essential that a 
>>> multitude of different things might need. In turn this makes using 
>>> third-party modules more difficult because they might introduce issues like 
>>> these that, especially for newcomers to the platform, are troublesome to 
>>> fix.
>>>
>>> I was hoping we could devise a solution, at least for package resources, 
>>> that would allow multiple declarations of the package and have those 
>>> declarations resolve themselves. I'm specifically focussing on package 
>>> here because that's the most common source of trouble, things like an 
>>> rbenv, rvm and pyenv module all declaring that they need 'build-essential'.
>>>
>>> My initial thought was that a package could have two states, hard and 
>>> soft where a state => hard, definition of a package would always trump any 
>>> other definitions of a package in a soft state. The question then becomes, 
>>> how do you deal with multiple soft states of the same package which 
>>> basically boils down to you can't have multiple declarations of the same 
>>> thing, so we're back to square one.
>>>
>>> My other thought was that perhaps it should be possible to `require => 
>>> Package['name']` from any resource and if that resource isn't declared in 
>>> the catalog spawn the resource ourselves but if it is defined that 
>>> declaration is used. I'm not sure how we could deal with scenario's where 
>>> someone requires said package but we explicitly define it as absent in our 
>>> manifests.
>>>
>>> Please spare me the "they're being wrong on the internet" speech, I'm 
>>> well aware of it but the truth of the matter is you'll never be able to 
>>> coordinate the community in such a way that this issue doesn't pop up. I'm 
>>> not really concerned about experienced Puppet users/developers that can 
>>> solve this for themselves, more about the less developer oriented operator 
>>> that has to use Puppet for whatever reason. Another approach would be to 
>>> really curate the modules up on the Forge but I doubt anyone feels like 
>>> doing that or has the time for it.
>>>
>>> I'm not sure if it's even desirable or possible to solve this issue in 
>>> the first place but I was curious if anyone has given any more thought to 
>>> this?
>>>
>>> -- 
>>> Daniele Sluijters
>>>
>>> -- 
>>> You received this message because you are subscribed to the Google 
>>> Groups "Puppet Developers" group.
>>> To unsubscribe from this group and stop receiving emails from it, send 
>>> an email to [email protected] <javascript:>.
>>> To view this discussion on the web visit 
>>> https://groups.google.com/d/msgid/puppet-dev/234c5510-c2b2-4786-ac76-68a73bf55e7d%40googlegroups.com
>>> .
>>> For more options, visit https://groups.google.com/groups/opt_out.
>>>
>>
>>
>>
>> -- 
>> Trevor Vaughan
>> Vice President, Onyx Point, Inc
>> (410) 541-6699
>> [email protected] <javascript:>
>>
>> -- This account not approved for unencrypted proprietary information -- 
>>
>
>
>
> -- 
> Trevor Vaughan
> Vice President, Onyx Point, Inc
> (410) 541-6699
> [email protected] <javascript:>
>
> -- This account not approved for unencrypted proprietary information -- 
>

-- 
You received this message because you are subscribed to the Google Groups 
"Puppet Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/puppet-dev/c517efc7-2db7-43c0-b977-bbd4f129c730%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to