From owner-freebsd-arch Sun Mar 17 9:44:54 2002 Delivered-To: freebsd-arch@freebsd.org Received: from bsdone.bsdwins.com (www.bsdwins.com [192.58.184.33]) by hub.freebsd.org (Postfix) with ESMTP id 190AD37B43D; Sun, 17 Mar 2002 09:44:52 -0800 (PST) Received: (from jwd@localhost) by bsdone.bsdwins.com (8.11.6/8.11.0) id g2HHipo08019; Sun, 17 Mar 2002 12:44:51 -0500 (EST) (envelope-from jwd) Date: Sun, 17 Mar 2002 12:44:51 -0500 From: John De Boskey To: "David O'Brien" Cc: Arch List Subject: Re: ftpd ESTALE recovery patch Message-ID: <20020317124451.A7636@bsdwins.com> References: <20020317084153.A3942@FreeBSD.org> <20020317085350.E10393@dragon.nuxi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20020317085350.E10393@dragon.nuxi.com>; from obrien@FreeBSD.ORG on Sun, Mar 17, 2002 at 08:53:50AM -0800 Sender: owner-freebsd-arch@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG ----- David O'Brien's Original Message ----- > On Sun, Mar 17, 2002 at 08:41:53AM -0800, John De Boskey wrote: > > In a busy cluster, a generated file being handed out by > > ftp is failing due to an ESTALE condition. The following > > patch fixes the problem. Failure to open the file is also > > logged when -l is specified twice (see ftpd(8)). > > This really does seem like a specific problem of your environment; and > this patch is something that really only scratches an itch of yours due > to how you've setup your environment. I'm sorry, but I have to disagree. ESTALE is an errno that indicates that an application can retry an operation with success. > It sounds like you need to copy the generated file to a place it will > remain thru out the ftp session. I am not sure I understand what you are trying to say. The target file to be transfered has already been stored in it's permanent location. However, it takes a small finite amount of time for NFS to propogate the information for the (new/replaced) file. Our clusters are able to attempt access during that short time at which point the process fails without the patch. > I do not think the patch should go in. And why `3' tries? Why not `10', > etc. If you can provide a better loop/timeout mechanism, I will be glad to review and incorporate it. Thanks! John http://people.freebsd.org/~jwd/ftpd.estale.patch To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message