JMP> Instantiation

JMP> Instantiation

Ron Bergman rbergma at dpc.com
Thu May 22 20:20:08 EDT 1997


Harry,


I agree with Jay.  deviceAlertCode should be conditionally-mandatory.
Where the condition is jobState is NeedsAttention.




	Ron




On Thu, 22 May 1997, JK Martin wrote:


> Based on your analysis, I'd go for instantiating deviceAlertCode *only*
> if it is applicable (ie, jobState is NeedsAttention, etc).
> 
> 	...jay
> 
> ----- Begin Included Message -----
> 
> From jmp-owner at pwg.org Thu May 22 19:27 EDT 1997
> From: Harry Lewis <harryl at us.ibm.com>
> To: <jmp at pwg.org>
> Subject: JMP> Instantiation
> Date: Thu, 22 May 1997 19:28:42 -0400
> 
> I would like to clarify something I think we agreed to in San Diego, but I'm
> not sure I understand it.
> 
> I think we said something like... all mandatory attributes must be
> "instantiated" as soon as a job becomes Pending. The idea, here, is that we
> can't predict what combinations of attributes an application might stuff
> together in a varbind so there must be a valid response to (at least) the
> mandatory set.
> 
> Meanwhile, we moved the alert code which would have been associated with state
> NeedsAttention out of the Job State Table and into the Attribute table... so,
> now I want to test the assumption by asking the question:
> 
>    Is the deviceAlertCode attribute only valid while the state is
>    needsAttention?  Should the agent return unknown(-1) while in other
>    states? Or would the row disappear from the Attribute table?
> 
> I think the answer is the row must be instantiated and the value -1 be
> returned, according to our agreement. This really doesn't seem very
> efficient, however, because there will be a deviceAlertCode row for every
> job in the Attribute table even though the vast majority of jobs will
> never need attention!
> 
> Harry Lewis - IBM Printing Systems
> 
> 
> ----- End Included Message -----
> 



More information about the Jmp mailing list