Hi Alessandra,
This is a fair point - I thought about this myself but came to the
conclusion that even if noone else uses the modules, if there's more
flexibility built in from the off then that helps me in the future
if/when software changes, etc. Having said that, this is part of the
reason why I haven't gone 'all the way' with heira, etc. and just used
templates as these are easy to do (just template each config file so you
can add stuff later).
On the wider issue of general use of group provided modules, is it worth
having a dedicated meeting of interested parties at all? I have looked
at the CERN ones but they seem a little like overkill for most things I
need them for and still a bit CERN centric (though that may just be me
not giving them a chance!). Maybe just an informal meeting where each
site/user gives an overview of what modules they use/can provide might
reduce duplication of work...?
Just my 2 cents :)
Thanks,
Mark
On 16/05/13 12:34, Alessandra Forti wrote:
> Here is the dilemma... we are discussing the structure of our own
> repository and there are of course different ways of doing things. One
> of these is to avoid site related values in the modules directory in
> order to simplify sharing them with other sites. So the real question
> here is how many sites would benefit from this? Because if we are
> adopting a more complicated way and then nobody is using it is just a
> wast of time.
>
> cheers
> alessandra
>
> On 16/05/2013 12:27, Christopher J. Walker wrote:
>> On 16/05/13 11:56, Mark Slater wrote:
>>> Hi Alessandra,
>>>
>>> In my puppet scripts I use templates extensively. I haven't gone to
>>> hiera yet but I am thinking about it...
>>>
>> The CERN templates do.
>>
>> Ben Jones gave a talk at Hepix on what they've done and he's leading a
>> puppet taskforce I think.
>>
>> Chris
>>
>>
>>> Thanks,
>>>
>>> Mark
>>>
>>> On 16/05/13 11:51, Alessandra Forti wrote:
>>>> Hi,
>>>>
>>>> how many people are using hiera/templates mechanism in puppet to
>>>> contain site related values?
>>>>
>>>> cheers
>>>> alessandra
>>>>
>
>
|