Return-Path: Received: from mail-yh0-f43.google.com ([209.85.213.43]:34262 "EHLO mail-yh0-f43.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1030276AbbC0PB6 (ORCPT ); Fri, 27 Mar 2015 11:01:58 -0400 Received: by yhch68 with SMTP id h68so39598285yhc.1 for ; Fri, 27 Mar 2015 08:01:57 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <2000185644.567074.1427466785581.JavaMail.zimbra@desy.de> References: <2000185644.567074.1427466785581.JavaMail.zimbra@desy.de> Date: Fri, 27 Mar 2015 11:01:57 -0400 Message-ID: Subject: Re: pNFS layout, ROC and multiple opens From: Trond Myklebust To: "Mkrtchyan, Tigran" Cc: linux-nfs list Content-Type: text/plain; charset=UTF-8 Sender: linux-nfs-owner@vger.kernel.org List-ID: On Fri, Mar 27, 2015 at 10:33 AM, Mkrtchyan, Tigran wrote: > which means that even I I set roc flag, I still need to recall layout. No. It means that the layout is automatically returned on last close and/or delegreturn. If you want it returned earlier then you need to send a cb_layoutrecall. > ----- Original Message ----- >> From: "Trond Myklebust" >> To: "Mkrtchyan, Tigran" >> Cc: "linux-nfs list" >> Sent: Friday, March 27, 2015 3:08:39 PM >> Subject: Re: pNFS layout, ROC and multiple opens > >> On Fri, Mar 27, 2015 at 7:24 AM, Mkrtchyan, Tigran >> wrote: >>> >>> >>> Hi folks, >>> >>> we run into following issue: >>> >>> on a client host we have two processes - A and B. >>> Process A opens a file and reads some data: >>> >>> OPEN -> LAYOUTGET -> READ. >>> Notice, that layout has a roc set to true. >>> >>> Process B opens the same file: >>> >>> OPEN -> READ. >>> >>> Client obviously re-used layout information. >>> Now, process A closes the file and as a result, >>> layout becomes invalid. >> >> No it doesn't. See the errata. >> >> Cheers >> Trond -- Trond Myklebust Linux NFS client maintainer, PrimaryData trond.myklebust@primarydata.com