From: Robert Thurlow (Robert.Thurlow@eng.sun.com)
Date: 05/29/02-05:30:49 PM Z
Date: Wed, 29 May 2002 16:30:49 -0600 (MDT) From: Robert Thurlow <Robert.Thurlow@eng.sun.com> Subject: Re: NFSv4 Replication and Migration: design team conference call Message-ID: <Roam.SIMC.2.0.6.1022711449.29602.thurlow@jurassic.eng> > I believe that the model of replication and migration that you > folks have in mind is more general, and does not assume > multi-hosted disks. > > I would like to somehow be involved (or at least listen in) > on your design conversations, to try to be aware of any > issues where what you are doing may actively interfere > with a multi-hosted disk failover model. We'd welcome your involvement. I'd summarize the current work as pertaining to what happens when you want to copy or (especially) move a filesystem over a network, where you may either be unable to have the new server handle the same filehandle or need to take pains to communicate the filehandle. Multi-hosted disks may not need attention from the server-to-server protocol we need to work out, but please do keep us honest in that area :-) The minutes will appear on the WG alias, and I'll include you in the review of an Internet-Draft-to-be we talked about in the meeting. Rob T
This archive was generated by hypermail 2.1.2 : 03/04/05-01:49:45 AM Z CST