From: Boris Z. (boris@conley.com)
Date: 06/09/99-04:51:49 PM Z
Message-ID: <01BEB2A0.C09E64B0.boris@conley.com> From: "Boris Z." <boris@conley.com> Subject: RE: Some caching questions Date: Wed, 9 Jun 1999 17:51:49 -0400 On Wednesday, June 09, 1999 4:52 PM, Noveck, Dave [SMTP:dave.noveck@netapp.com] wrote: > I have some question about the weak-cache-consistency stuff > that went into v3 and how v3 clients use it and I also have > a question about directory caching in v3 clients. > > The first thing is that I notice that this stuff isn't in > v4 and I'm wondering why. Does it reflect some belief that > this stuff isn't useful? > I guess we intended to replace it with something more deterministic (byte range oplocks for example). > My other caching question has to do with caching of directories. Are > there any v3 clients out there that cache directories other than in the > normal name cache sense of caching name-to-vnode or name-to-handle > mappings for lookups, etc. that have succeeded. Specifically, does > anyone: > > Caching of directories without 'directory change notifications' is even more dangerous than caching of data base on wcc. On Wednesday, June 09, 1999 4:52 PM, Noveck, Dave [SMTP:dave.noveck@netapp.com] wrote: > I have some question about the weak-cache-consistency stuff > that went into v3 and how v3 clients use it and I also have > a question about directory caching in v3 clients. > > The first thing is that I notice that this stuff isn't in > v4 and I'm wondering why. Does it reflect some belief that > this stuff isn't useful? > I guess we intended to replace it with something more deterministic (byte range oplocks for example). > My other caching question has to do with caching of directories. Are > there any v3 clients out there that cache directories other than in the > normal name cache sense of caching name-to-vnode or name-to-handle > mappings for lookups, etc. that have succeeded. Specifically, does > anyone: > > Caching of directories without 'directory change notifications' is even more dangerous than caching of data base on wcc.
This archive was generated by hypermail 2.1.2 : 03/04/05-01:47:12 AM Z CST