From: Bas van Schaik Subject: Re: Reoccurring ext3 errors: attempt to access beyond end of device, freeing blocks not in datazone Date: Fri, 23 May 2008 16:41:03 +0200 Message-ID: <4836D77F.30304@tuxes.nl> References: <4832941A.70806@tuxes.nl> <20080520123505.GP15035@mit.edu> <48334A82.6020508@tuxes.nl> <20080522181827.GA3516@webber.adilger.int> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: linux-ext4@vger.kernel.org To: Andreas Dilger Return-path: Received: from a-eskwadraat.nl ([131.211.39.72]:47817 "EHLO a-eskwadraat.nl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750820AbYEWOlP (ORCPT ); Fri, 23 May 2008 10:41:15 -0400 In-Reply-To: <20080522181827.GA3516@webber.adilger.int> Sender: linux-ext4-owner@vger.kernel.org List-ID: Andreas Dilger wrote: > On May 21, 2008 00:02 +0200, Bas van Schaik wrote: > >> Theodore Tso wrote: >> >>> 15629775440 / 8 = 0x74736E4A (or in ascii 'Inst') >>> 13075964688 / 8 = 0x616C6C62 (or in ascii 'blla') >>> 15354014352 / 8 = 0x72657552 (or in ascii 'Ruer') >>> >>> Converting these numbers to hex: >>> >>> 1953721929 = 0x74736E49 (or in ascii 'Jnst') >>> 1634495585 = 0x616C6C61 (or in ascii 'alla' >>> 543517044 = 0x20656974 (or in ascii 'tie ') >>> 1919251793 = 0x72657551 (or in ascii 'Quer') >>> >>> Given that it's all ascii, it looks like the indirect block somehow >>> was overwritten, or was substituted by text. >>> >> Ah, such a lead was exactly what I was looking for, now I at least know >> where those bogus numbers were coming from. Maybe a very dump question: >> you seem to have reverse the ascii "translation", why? And shouldn't >> "Jnst" be "Inst"? Note also that the "translations" seem to resemble >> each other a little bit: "Jnst" = "Jnst", "alla" looks like "blla" and >> "Ruer" looks like "Quer". Coincidence? >> > > Or it is possible you are getting single-bit errors somewhere along > your IO path, and another single-bit error has resulted in this data > being written to the wrong block to begin with... > I don't think I understand your statement fully... Do you mean that some bitflip caused the filesystem to initiate a write to the wrong blocks? -- Bas