[DRMAA-WG] Conf call minutes - Jan 20th 2009

[Daniel Gruber]

Attendees:
   Peter Troeger
   Dan Templeton
   Hrabri Rajic
   Daniel Gruber

1. Meeting secretary for this meeting?
   Daniel Gruber

2. Next IDL language binding ?

   Python binding was finished within 3 months
   -> efficient
   
   Do we want to do it again for Java / C binding?
   -> low priority
   -> DRMAA2 is currently more important

3. DRMAA2 schedule (requested by Steven)

   -> doable within 6 months  
   -> some time for approval after that
   -> there are two OGF meetings in this time (26th and 27th)
   -> earlier agreement: fixing everything we identified

4. DRMAA2 feature discussion

   Do we need "System Hold"/ "User Hold" differentiation?
   -> idea: reason as exception?
   -> agreed on that it is not really necessary

   Undetermined - is it a valid job state?
   -> Yes! Undetermined = Error
   -> Condor: it is permanent some time
   -> Need to clarify if this means "don't try again"
      or "try it again"

   Distinction between state "failed" and "terminated"
   -> "Failed" := user can fix it (through changes on job template for 
example)
   -> "Terminated" := error the user can't fix

   TODO later: DRMAA wait state will be discussed other time

   Why should we support extensible state?
   -> basically for reporting
   -> problem: difficult to implement in C

   Solution for more fine grained states (and "User/System Hold" problem)
   -> add another output parameter (a DRM specific string) or constants
      for job state
   -> hence "Hold" could be one state
   -> fine grained states could additionally retrieved via such DRM
      specific string or constant
   -> proposal: one string and a reference to a generic object/struct
   -> will appear in draft later on!

   Problem with state transition in Condor from "Suspend" to "Queued 
Active"!
   -> should be supported (at least 3 people agreed)

5. Continue with email threads for discussion.

6. Next meeting: 3rd February 2009