From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ee0-f46.google.com ([74.125.83.46]:51913 "EHLO mail-ee0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751049Ab3CGKX0 (ORCPT ); Thu, 7 Mar 2013 05:23:26 -0500 Received: by mail-ee0-f46.google.com with SMTP id e49so171612eek.5 for ; Thu, 07 Mar 2013 02:23:24 -0800 (PST) Message-ID: <51386A98.90302@gmail.com> Date: Thu, 07 Mar 2013 11:23:20 +0100 From: Michael Schmitt MIME-Version: 1.0 To: Chris Mason , "linux-btrfs@vger.kernel.org" Subject: Re: weird kernel-oopses while deleting files on btrfs References: <513356D5.5030301@gmail.com> <20130303235243.GA26002@shiny.masoncoding.com> In-Reply-To: <20130303235243.GA26002@shiny.masoncoding.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Sender: linux-btrfs-owner@vger.kernel.org List-ID: Am 04.03.2013 00:52, schrieb Chris Mason: > On Sun, Mar 03, 2013 at 06:57:41AM -0700, Michael Schmitt wrote: >> Hi list, >> >> some rather unexpected btrfs-oopses for my taste. I use btrfs for some >> time now (mostly on external harddisks) and these "oopses" happened >> during some simple file and folder deletion operation on that device. It >> is a luks-encrypted 80GB drive. Anything like that known? And the fs was >> created just yesterday, how come there is a message like... >> >> [91491.919358] btrfs: mismatching generation and generation_v2 found in root item. This root was probably mounted with an older kernel. Resetting all new fields. > This may be from first mount after mkfs. It depends on your tools. During mkfs, mount and that message the same kernel and tools were used. At least I am quite sure about that (but I wouldn't swear on my life). >> ... but the kernel used (3.7.3 from Debian experimental on Debian sid) >> was installed several days ago. What kind of oopses are these? As of now >> there is no real data on that device. But if there were, would I need to >> be concerned about the integrity of those files? >> >> [93283.762006] WARNING: at /build/buildd-linux_3.7.3-1~experimental.1-i386-eX5kUQ/linux-3.7.3/fs/btrfs/extent-tree.c:6297 btrfs_alloc_free_block+0xcd/0x2a4 [btrfs]() > These are not oopsen but warnings. It's an ENOSPC warning as we try to > delete the extents. It did happen sometimes in this kernel, but it is > only a warning. Ah, 3.7.x had issues there? Do you have by any chance the bug-# or URL at hand? Or could you elaborate a bit what issues a user might have there and if he should be concerned? I just wonder how important a kernel up- or downgrade might be. But as it is only a "warning" (sorry that I did get that wrong), I suppose nothing "bad" could happen? ENOSPC on an almost empty filesystem would be at least somewhat "bad". :) But fs was almost full and I deleted it completely as the message appeared the first time. I copied another batch of files to it (only a few GB, maybe 10% of drivespace used) and deleted them again and got the next "warning"-message. regards Michael