From: Boris Z. (boris@conley.com)
Date: 10/06/98-06:49:07 PM Z
Message-ID: <01BDF149.3C99A210.boris@conley.com> From: "Boris Z." <boris@conley.com> Subject: RE: locking errors Date: Tue, 6 Oct 1998 16:49:07 -0700 On Tuesday, October 06, 1998 12:55 PM, Mike Eisler [SMTP:mre@Eng.Sun.COM] wrote: > It comes down to: > > - do want to use NFS V4 locking through firewalls? > - do you want to require the server to maintain state > through server reboot? > - do you want your server to offer service quickly after > a reboot? > > The 4th question is do you want clients to lose locks during a network > paritition? Obviously no one wants that, but depending on how you answer the > above 3 questions, you may get a distasteful answer. I think the compared the > 1st 3 questions, the 4th question pales. I believe the protocol should not limit servers to answer these questions as they want. Some of them may want to answer "yes, yes, yes, no" and declare themselves ready for 'mission critical applications'. The protocol just needs to address possible differences in client/server interaction in environments with high and low lease timeouts.
This archive was generated by hypermail 2.1.2 : 03/04/05-01:46:30 AM Z CST