From: Timo Sirainen Subject: Re: [NFS] Cache flushing Date: Wed, 21 Nov 2007 22:56:24 +0200 Message-ID: <1195678584.6039.436.camel@hurina> References: <1195258291.6039.189.camel@hurina> <1195328785.6999.5.camel@localhost.localdomain> <600549E3-82CF-44EB-8394-E57A3BB41118@iki.fi> <1195332062.6999.20.camel@localhost.localdomain> <1195337516.6039.239.camel@hurina> <1195524890.6039.315.camel@hurina> <1195602454.7234.100.camel@heimdal.trondhjem.org> <1195650851.6039.425.camel@hurina> <1195653389.8374.4.camel@heimdal.trondhjem.org> <1195677569.8374.18.camel@heimdal.trondhjem.org> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0330655882==" Cc: nfs@lists.sourceforge.net To: Trond Myklebust Return-path: Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.92] helo=mail.sourceforge.net) by sc8-sf-list2-new.sourceforge.net with esmtp (Exim 4.43) id 1Iuwcc-0001Wh-A2 for nfs@lists.sourceforge.net; Wed, 21 Nov 2007 12:56:34 -0800 Received: from dovecot.org ([82.118.211.50]) by mail.sourceforge.net with esmtp (Exim 4.44) id 1Iuwcf-0005Tp-LB for nfs@lists.sourceforge.net; Wed, 21 Nov 2007 12:56:40 -0800 In-Reply-To: <1195677569.8374.18.camel-rJ7iovZKK19ZJLDQqaL3InhyD016LWXt@public.gmane.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: --===============0330655882== Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-Vo7VInh71aL1v7G816pI" --=-Vo7VInh71aL1v7G816pI Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Wed, 2007-11-21 at 15:39 -0500, Trond Myklebust wrote: > On Wed, 2007-11-21 at 22:36 +0200, Timo Sirainen wrote: > > On 21.11.2007, at 15.56, Trond Myklebust wrote: > >=20 > > > On Wed, 2007-11-21 at 15:14 +0200, Timo Sirainen wrote: > > > > > >> Now that I think of it, I guess the reason is that my 2.6.22 setup =20 > > >> has > > >> Linux+ext3 as NFS server and 2.6.17-rc2 setup has NetApp as =20 > > >> server. So > > >> even though open()+close() flushes the directory's attribute =20 > > >> cache, it > > >> doesn't flush the file name -> NFS handle cache unless mtime changes= ? > > > > > > Right. You're hitting the principal limitation of ext3 as an NFS =20 > > > backend > > > filesystem. > > > As I said in an earlier mail, the resolution on the time is 1 =20 > > > second, so > > > client 2 can basically not see any changes that happen within < 1 =20 > > > second > > > on client 1. The reason is that the mtime stays the same. > >=20 > > With files it's possible to work around this limitation with fcntl =20 > > locking, but directories can't be locked. So back to my original =20 > > question: Would it be possible to have chmod(dir, (uid_t)-1, =20 > > (gid_t)-1) flush its cache? I don't see what harm there could be in =20 > > this, since normally no-one would do it anyway and the change is simple= - To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html --=-Vo7VInh71aL1v7G816pI-- --===============0330655882==--