Sarah Jelinek wrote:
> Hi Dave and Jack,
> 
>>>>>
>>>>
>>>> I'd view it as an important requirement that we be able to perform 
>>>> full semantic verification on the server.  Why is this regarded as 
>>>> optional?
>>> It's not that it's optional on the server, it's that the server 
>>> cannot provide the full client context to do full validation.  Of 
>>> course we want to do as much validation as we can on both the server 
>>> and the client.
>>>
>>
>> I would expect that a user could supply whatever context we desire to 
>> use, either synthetically or via importation of some client context 
>> that was generated from a real client.  From the point of view of the 
>> validator, the client is just a bunch of parameters, which can also be 
>> supplied via other avenues.  Requiring real, live clients in order to 
>> validate thus seems unnecessary.
> 
> I don't think in all validation scenarios we require real, live clients. 
> But, we have to know that some validation is not possible if we don't 
> have the context for which we are doing the validation. The issue with 
> some of the semantic validation, specific to the client, is that beyond 
> simple 'syntax' checking or format checking of something like a target 
> device specification, without a live client or a set of client data that 
> provides this information, it is hard to validate the schema. So, are 
> you proposing we provide for 'importing' the client data if the user so 
> desires so we can do more contextual validation when setting up a 
> service? I am not sure why we would want to do that actually.
> 
> Seems to me a service is independent of the client. Clients discover 
> services, and one service will be used to produce an installed system, 
> but the manifests that are provided in that service may apply to many 
> clients. Providing this contextual semantic validation as a user is 
> setting up a service seems to break what our model is. As you note, we 
> would have to have some data regarding the client to do this type of 
> validation, which implies that the user setting up the service would 
> have to know which clients the service applied to, if they wanted to 
> make use of it.
> 
> During specific client setup, that is installadm create-client, we could 
> do this.
> 

Apparently there's some confusion.  I wasn't suggesting that this be 
done specifically at the time a manifest is installed into the service, 
but as a manifest development tool used outside that process.  It's 
fairly unlikely that you'd have access to a client against which to 
validate even in the create-client case.

Dave

Reply via email to