From mboxrd@z Thu Jan 1 00:00:00 1970 From: Theodore Ts'o Subject: Re: ext4 fix for interaction between i_size, fallocate, and delalloc after a crash Date: Thu, 30 Nov 2017 09:22:19 -0500 Message-ID: <20171130142219.qeh3a7hleyvxkmkf@thunk.org> References: <20171127161137.4ghjcxfklpurk2eo@thunk.org> <20171128204525.ijpis74t75f4bbsc@thunk.org> <20171129061307.f6hekryi3bppo26y@thunk.org> <20171130004816.pdzfqjluchtvfq6u@thunk.org> <20171130044649.xfgwhd2wzdc5mwql@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Amir Goldstein , Vijay Chidambaram , Ext4 To: Ashlie Martinez Return-path: Received: from imap.thunk.org ([74.207.234.97]:47102 "EHLO imap.thunk.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751125AbdK3OWX (ORCPT ); Thu, 30 Nov 2017 09:22:23 -0500 Content-Disposition: inline In-Reply-To: <20171130044649.xfgwhd2wzdc5mwql@thunk.org> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Wed, Nov 29, 2017 at 11:46:49PM -0500, Theodore Ts'o wrote: > > Can you send the block I/O trace that was observed when you did the > following, a complete output of dumpe2fs on the file system, and a > debugfs stat output on the test file? Oh, and the block I/O trace after CrashMonkey reordered the I/O operations after the last CACHE FLUSH / barrier operation? I'm ***really*** curious what I/O operations happened *after* the last barrier operation if you had waited 120 secounds before triggering the crash. - Ted