From: Andres Freund Subject: Re: EXT4 ENOSPC Bug Date: Mon, 16 Feb 2009 12:37:19 +0100 Message-ID: <49994FEF.2020908@anarazel.de> References: <200811291418.24672.andres@anarazel.de> <200812100108.04163.andres@anarazel.de> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: adilger@sun.com, LKML , linux-ext4@vger.kernel.org, Jonathan Bastien-Filiatrault , "Aneesh Kumar K.V" To: Theodore Tso Return-path: Received: from mail.anarazel.de ([217.115.131.40]:34717 "EHLO smtp.anarazel.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751170AbZBPLru (ORCPT ); Mon, 16 Feb 2009 06:47:50 -0500 In-Reply-To: <200812100108.04163.andres@anarazel.de> Sender: linux-ext4-owner@vger.kernel.org List-ID: Hi All, "Luckily", to prove that I am not completly mad I recently found this problem again. On a more recent kernel: 2.6.29-rc3-andres-00498-g68e80d5 (upstream 2.6.29 + Theodore's ext4 debug patches) Again I got: open("/home/andres/tt", O_WRONLY|O_CREAT|O_NOCTTY|O_NONBLOCK, 0666) = -1 ENOSPC (No space left on device) When doing `touch /home/andres/tt` Again df does not show something suspicious: Filesystem Inodes IUsed IFree IUse% Mounted on /dev/mapper/root_crypt 19234816 826296 18408520 5% / tmpfs 491025 3 491022 1% /lib/init/rw varrun 491025 96 490929 1% /var/run varlock 491025 3 491022 1% /var/lock udev 491025 5517 485508 2% /dev tmpfs 491025 1 491024 1% /dev/shm /dev/sda2 61312 110 61202 1% /boot Filesystem 1K-blocks Used Available Use% Mounted on /dev/mapper/root_crypt 302855628 202819548 100036080 67% / tmpfs 1964100 0 1964100 0% /lib/init/rw varrun 1964100 300 1963800 1% /var/run varlock 1964100 0 1964100 0% /var/lock udev 1964100 92 1964008 1% /dev tmpfs 1964100 0 1964100 0% /dev/shm /dev/sda2 964532 634052 281484 70% /boot Delayed allocations are not showing anything suspicious: [72586.050538] ext4 debug delalloc of dm-0 [72586.050547] ext4: dirty blocks 22 free blocks 25008749 [72586.050551] s_dirty list: [72586.050556] ino 534064: 2 2 [72586.050559] ino 539981: 2 2 [72586.050563] ino 524506: 5 2 [72586.050566] ino 184479: 1 2 [72586.050571] ino 167172: 2 2 [72586.050574] ext4 debug delalloc done Any debugging ideas? I got this twice in the last two months, so its not really easy to reproduce :-( Andres PS: All information is taken out of the running kernel while having the problem meminfo: MemTotal: 3928200 kB MemFree: 23312 kB Buffers: 239564 kB Cached: 2383280 kB SwapCached: 0 kB Active: 1490948 kB Inactive: 2149360 kB Active(anon): 879880 kB Inactive(anon): 294356 kB Active(file): 611068 kB Inactive(file): 1855004 kB Unevictable: 52 kB Mlocked: 52 kB SwapTotal: 0 kB SwapFree: 0 kB Dirty: 52 kB Writeback: 4 kB AnonPages: 1017356 kB Mapped: 129440 kB Slab: 123316 kB SReclaimable: 94512 kB SUnreclaim: 28804 kB PageTables: 26936 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 1964100 kB Committed_AS: 1169664 kB VmallocTotal: 34359738367 kB VmallocUsed: 341580 kB VmallocChunk: 34359392035 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB DirectMap4k: 86016 kB DirectMap2M: 4003840 kB