From owner-freebsd-fs@FreeBSD.ORG Wed Aug 31 00:17:58 2011 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 0A32B1065672; Wed, 31 Aug 2011 00:17:58 +0000 (UTC) (envelope-from geo.liaskos@gmail.com) Received: from mail-yi0-f54.google.com (mail-yi0-f54.google.com [209.85.218.54]) by mx1.freebsd.org (Postfix) with ESMTP id 4C1478FC1A; Wed, 31 Aug 2011 00:17:57 +0000 (UTC) Received: by yib19 with SMTP id 19so203312yib.13 for ; Tue, 30 Aug 2011 17:17:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=txEr8UwW/UN26zN9v8G1HaWBotqxGa+j0bIv/W1a3ec=; b=vbzkAg/o50XUter5QAinTAecKgFe9+y88Yifln/tTNHs/S/WqTbUIlhr7x8oXYWFUL jZ8woSynsuuqIVHnFNhlXsESRGPvkmrOEGZ4oZfiiBYSaD4tuVYQ2wOvZPUJWs3HAYKp HKJdHeiPRTYm+rVa4D7IxI2OKG6WoNW8ihDuY= MIME-Version: 1.0 Received: by 10.101.3.40 with SMTP id f40mr5713633ani.89.1314749876693; Tue, 30 Aug 2011 17:17:56 -0700 (PDT) Received: by 10.100.42.15 with HTTP; Tue, 30 Aug 2011 17:17:56 -0700 (PDT) In-Reply-To: <1005169645.540203.1314717014356.JavaMail.root@erie.cs.uoguelph.ca> References: <1005169645.540203.1314717014356.JavaMail.root@erie.cs.uoguelph.ca> Date: Wed, 31 Aug 2011 03:17:56 +0300 Message-ID: From: George Liaskos To: Rick Macklem Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Cc: freebsd-fs@freebsd.org, Pawel Jakub Dawidek Subject: Re: NFSv4: After upgrade to 9 users can no longer list files. (sounds like a ZFS issue?) X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 31 Aug 2011 00:17:58 -0000 > You could try this patch and see what effect it has (applied to the > server). It just disables the access check for readdir. > --- nfs_nfsdport.c.sav2 2011-08-30 10:35:58.000000000 -0400 > +++ nfs_nfsdport.c =C2=A0 =C2=A0 =C2=A02011-08-30 10:36:54.000000000 -040= 0 > @@ -1838,10 +1838,12 @@ nfsrvd_readdirplus(struct nfsrv_descript > =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0nd->nd_repstat =3D= NFSERR_NOTDIR; > =C2=A0 =C2=A0 =C2=A0 =C2=A0if (!nd->nd_repstat && cnt =3D=3D 0) > =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0nd->nd_repstat =3D= NFSERR_TOOSMALL; > +#ifdef notnow > =C2=A0 =C2=A0 =C2=A0 =C2=A0if (!nd->nd_repstat) > =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0nd->nd_repstat =3D= nfsvno_accchk(vp, VEXEC, > =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0nd->= nd_cred, exp, p, NFSACCCHK_NOOVERRIDE, > =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0NFSA= CCCHK_VPISLOCKED, NULL); > +#endif > =C2=A0 =C2=A0 =C2=A0 =C2=A0if (nd->nd_repstat) { > =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0vput(vp); > =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0if (nd->nd_flag & = ND_NFSV3) > > This wouldn't be suitable for a production system, but whether or > not it "fixes" the problem would give us an indication of where the > problem is. > > Also, if you could clarify when your 8/stable was downloaded, whether > your 9.0 upgrade was to vanilla Beta1 or ??? and details w.r.t. your > ZFS setup, that might help. I use svn, unfortunately i don't remember exactly when i moved from 8.2 to stable. I synced with CURRENT last week and this issue appeared, i did a second update to beta 2 [r225237] with the same results. The patch didn't make any difference. I downloaded an ISO with BETA-1 and made a VM installation, i was not able to reproduce this. Updated one of the clients to r225237, setup some nfs exports on top of ZFS and ls does not work for non root users. I created a new pool on top of a memory fs to test this. Next, i "downgraded" the server to BETA-1 [r224413] and everything is back to normal. So there's a bug which was introduced somewhere between BETA-1 && BETA-2 :p Thank you for your help! Regards, George