Return-Path: linux-nfs-owner@vger.kernel.org Received: from mx2.netapp.com ([216.240.18.37]:7421 "EHLO mx2.netapp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758672Ab2IMSMh convert rfc822-to-8bit (ORCPT ); Thu, 13 Sep 2012 14:12:37 -0400 From: "Adamson, Andy" To: "J. Bruce Fields" CC: "Adamson, Andy" , "Myklebust, Trond" , Jeff Layton , "" Subject: Re: [PATCH 0/4] RFC Avoid expired credential keys for buffered writes Date: Thu, 13 Sep 2012 18:12:36 +0000 Message-ID: <7851A663D3BF5E41B9E166C2C7A42145102C06E1@SACEXCMBX02-PRD.hq.netapp.com> References: <1346961251-2554-1-git-send-email-andros@netapp.com> <20120910145732.4d800f5b@corrin.poochiereds.net> <7851A663D3BF5E41B9E166C2C7A42145102B2F5A@SACEXCMBX02-PRD.hq.netapp.com> <20120910160858.552bec1b@corrin.poochiereds.net> <7851A663D3BF5E41B9E166C2C7A42145102B48E8@SACEXCMBX02-PRD.hq.netapp.com> <4FA345DA4F4AE44899BD2B03EEEC2FA908F9DD4F@SACEXCMBX04-PRD.hq.netapp.com> <7851A663D3BF5E41B9E166C2C7A42145102B4C9F@SACEXCMBX02-PRD.hq.netapp.com> <20120913175754.GA6566@fieldses.org> In-Reply-To: <20120913175754.GA6566@fieldses.org> Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Sender: linux-nfs-owner@vger.kernel.org List-ID: On Sep 13, 2012, at 1:57 PM, J. Bruce Fields wrote: > On Wed, Sep 12, 2012 at 04:14:38PM +0000, Adamson, Andy wrote: >> >> On Sep 12, 2012, at 11:21 AM, Myklebust, Trond wrote: >> >>> On Wed, 2012-09-12 at 15:13 +0000, Adamson, Andy wrote: >>>> After doing more test verification, here are the reasons for the low watermark. Reason #2 is the strongest justification. > > 1 and 2 don't sound right. What exactly were the test failures? > > The client and server's gss code already check the context expiry for > us--we don't want an extra check in an upper layer on the client. Yeah, I agree that another upcall is not really the right way to go. What we want is to be notified (on the client) when a user's credential has changed - either destroyed or renewed. I have some ideas which I will propose after these patches are resolved. > > The context *will* expire unexpectedly sometimes, and we do have to > handle that. (The server's clock could be a tad faster than the > server's, or the server could reboot, etc., etc.) Agreed, and the patch set needs to recognize (handle) this. But this is not the normal case. > > I agree with all the suggestions for trying to anticipate expiry in the > normal cases and preparing to minimize the damage, that's fine. But > once the expiry finally comes we should leave the existing mechanisms to > do their job. I'm not changing what happens at GSS context expiry, but what happens just before the context expires. I do hear the arguments against the low watermark, and will submit patches with a single (the so-called high) watermark. We can then proceed from there. :) -->Andy > > --b.