From: Boris Z. (boris@conley.com)
Date: 07/16/98-03:55:50 PM Z
Message-ID: <01BDB0C1.71612310.boris@conley.com> From: "Boris Z." <boris@conley.com> Subject: RE: Atomic operations - Comments Please Date: Thu, 16 Jul 1998 13:55:50 -0700 It looks like this problem is similar by nature to the one that made you declare NLM SHARE and NLM LOCK advisory. This was fine for UNIX centric environment. Should we indicate in FS Attributes if the system supports mandatory NLM? On Thursday, July 16, 1998 10:32 AM, Peter Staubach [SMTP:Peter.Staubach@Eng.Sun.COM] wrote: > > > > On Thu Jul 16 17:45:36 1998, Boris Z. wrote: > > > Is BEGIN/END ATOMIC an attempt to add server based transaction > > > atomicity to compound requests? > > > If it just an isolation of client requests, how is it different from > > > adding > > > "NLM_SHARE(?, DENY_RW),?, restore Share level" to a compound request? > > > > > > > > NLM_SHARE is used by PC clients to perform a very specific SHARE > > request which is in force over the lifetime that the client has a file > > open. What I am suggesting is a way to allow a compound request to be > > atomic with having to assume that for all compound requests. The > > second option of the MUTEX is a more cooperative and very specific > > method which would be less of a burden then the ATOMIC operator. > > > > So, what happens if the server can not implement this atomic operator? > Does it return an error? > > I am asking this because this seem like it could be a difficult thing > to implement on a server which also allows local access to files. > > ps >
This archive was generated by hypermail 2.1.2 : 03/04/05-01:45:58 AM Z CST