Return-Path: Received: from dkim2.fusionio.com ([66.114.96.54]:47375 "EHLO dkim2.fusionio.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760983Ab3DCNFo convert rfc822-to-8bit (ORCPT ); Wed, 3 Apr 2013 09:05:44 -0400 Received: from mx2.fusionio.com (unknown [10.101.1.160]) by dkim2.fusionio.com (Postfix) with ESMTP id 7DF2A9A0680 for ; Wed, 3 Apr 2013 07:05:43 -0600 (MDT) Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 From: Chris Mason To: Tsutomu Itoh , Linux Btrfs References: <515B838A.7020000@jp.fujitsu.com> In-Reply-To: <515B838A.7020000@jp.fujitsu.com> Message-ID: <20130403130541.27335.23925@localhost.localdomain> Subject: Re: (3.9-rc5) WARNING: at fs/btrfs/super.c:255 __btrfs_abort_transaction Date: Wed, 3 Apr 2013 09:05:41 -0400 Sender: linux-btrfs-owner@vger.kernel.org List-ID: Quoting Tsutomu Itoh (2013-04-02 21:19:06) > Hi, > > When inode_cache was specified for the mount option, I encountered the > following messages at umount. > > Step to reproduce: > > mkfs.btrfs -f /dev/sdc4 > mount -o compress=lzo,autodefrag,inode_cache /dev/sdc4 /test1 > dd if=/dev/zero of=/test1/file1 bs=1M count=8192 > dd if=/test1/file1 of=/dev/null bs=1M > rm -f /test1/file1 > umount /test1 > Apr 3 09:43:12 luna kernel: [ 554.639149] BTRFS error (device sdc4) in btrfs_save_ino_cache:471: error 28 We're running out of space while saving the inode cache. Strictly speaking this should not abort the FS. I'll take a look. -chris