>"Carroll Kong" wrote, >That is one of the sentences I wrote. You seemed to have missed that >one. Although I suppose I was not as clear. Sorry. > >I agree with what you said or at least speculated that yes, there can only >be ONE component of the particular layer. Which would imply that NFS is >indeed at the application layer since it uses RPC which RPC itself is in >the Session layer. > >NFS itself has synchronization issues which some have considered to be a >Session Layer characteristic. > >I never said once that the USE of RPC means that it should be in the >session layer. I did mention that the fact that NFS has synchronization >primitives, which is considered a characteristic of the Session layer.
From a formal OSI perspective, the synchronization primitives in the Session _service_ deal first with establishing the two-way alternating/two-way simultaneous relationship of the application stream of records, the delimiting of records, and the ability to checkpoint records or, in some cases, groups of records. The session service, however, does not contain the abstraction of a file. File synchronization would be an application service, typically in a transaction processing protocol with commitment features. Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=28780&t=28378 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
