On Nov 18, 2008, at 9:09 AM, Sven Mueller wrote:

>
> Luke Kanies schrieb:
> [overriding variables in derived nodes (and classes?)]
>>
>> This is a common problem I haven't been able to find a good solution
>> for.  It's not so much related to the write-once aspect of variables
>> as it is related to the fact that classes are evaluated immediately,
>> so the subnode's overriding of a variable has no affect.
>
> Hmm, is there a specific reason why classes are evaluated immediately?
> IMHO evaluation should start with global entries, followed by the
> specific node (following down inheritance), but I'm not sure at all  
> how
> this could be achieved with (current) puppet architecture. Actually I
> know this won't be easy at all, but it would make puppet so much  
> easier
> to use for us.
> (Actually, if you had an idea how to solve this, I'm pretty (ie. 75%)
> sure I could arrange for you to get paid to implement it)

I have some ideas, but I'm not sure it's possible without either  
changing backward compatibility or largely redefining what nodes mean  
-- they're currently very similar to classes, and we'd need to  
differentiate them considerably, probably by limiting the statements  
they allow (no resources) and changing how 'include' works to make it  
late-binding somehow.

>
>>> Any suggestion for a more elegant solution would be appreciated.
>>
>> An external node solution would work much better.
>
> So you mean an external program providing the "host" entries in YAML
> format? Hmm, might be a solution.... Will think about it.

If, rather than fixing the node type, your company might be willing to  
sponsor a straightforward external nodes app, that might be a better  
place to start.  I'd like to get one going, and I'll never be able to  
deprecate our node syntax until there's a good replacement.

-- 
True Terror is to wake up one morning and discover that your high
school class is running the country. -- Kurt Vonnegut
---------------------------------------------------------------------
Luke Kanies | http://reductivelabs.com | http://madstop.com


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Puppet Developers" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/puppet-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to