On Nov 25, 2009, at 12:53 PM, Nigel Kersten wrote:

> On Wed, Nov 25, 2009 at 12:21 PM, Luke Kanies <[email protected]>  
> wrote:
>
>> Well, the environment doesn't matter for the files, because it's
>> included in the URI itself.
>
> But not in localconfig.yaml right? So if you fail on your initial
> connection, and continue to apply a cached catalog, don't you fall
> back to the same problem?

I'm not actually sure.  My point was that the server doesn't need to  
know the client's env for the client to do a call, as long as the  
client knows it.  The problem with the facts, and I believe this is  
unique to the fact/catalog coupling, is that the compiling server  
needs access to the client's facts, and we'd prefer to do that over  
two queries rather than one.

>>>> Is this a reasonable approach?  It's obviously not sufficient for
>>>> 0.25.2, but I think it's a better long term direction.
>>>
>>> So for me this all depends on how well it scales... happy to do  
>>> tests.
>>>
>>> Is it perhaps feasible to have the server tell the client to  
>>> resubmit
>>> the fact values to it if it doesn't have a fresh cache? I have this
>>> nagging feeling I already brought up this in the past and it wasn't
>>> feasible.
>>
>>
>> You have the same problem, don't you?  If we push new data, it could
>> again end up at the wrong host.
>
> I assumed if a specific server requested the info, it would be
> delivered to that server.


That only works, I believe, for people who use DNS round robin or its  
equivalents, rather than people who use a load balancer, right?

-- 
Normal is getting dressed in clothes that you buy for work and driving
through traffic in a car that you are still paying for - in order to
get to the job you need to pay for the clothes and the car, and the
house you leave vacant all day so you can afford to live in it.
     -- Ellen DeGeneres
---------------------------------------------------------------------
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