From guolin at alexa.com Wed Nov 3 22:00:17 2004 From: guolin at alexa.com (Guolin Cheng) Date: Wed, 3 Nov 2004 14:00:17 -0800 Subject: Help! how to infer failed fileName|inode from LBA sector for ext3 on Fedora Linux?? Message-ID: <41089CB27BD8D24E8385C8003EDAF7AB018CE30A@karl.alexa.com> Hi, I have an ext3 file system question for a while: How to obtain failed file (inode) from LBA section reported by Linux kernel? The following lines are a simple example: ... FCbox1: Nov 1 04:08:59 FCbox1 kernel: hda: dma_intr: status=0x51 { DriveReady SeekComplete Error } FCbox1: Nov 1 04:08:59 FCbox1 kernel: hda: dma_intr: error=0x40 { UncorrectableError }, LBAsect=165843413, high=9, low=14848469, sector=129761360 ... I'd like to know the tools|commands(&subcommands) used to do the translation below, any help are greatly appreciated. LBA section == ?? => logical sector in partition == ?? => block == ?? => inode in ext3 fileSystem == find|ls => filename. Thanks a lot. --Guolin -------------- next part -------------- An HTML attachment was scrubbed... URL: From A.Apprich at science-computing.de Thu Nov 4 14:40:17 2004 From: A.Apprich at science-computing.de (Alexander Apprich) Date: Thu, 04 Nov 2004 15:40:17 +0100 Subject: Help! how to infer failed fileName|inode from LBA sector for ext3 on Fedora Linux?? In-Reply-To: <41089CB27BD8D24E8385C8003EDAF7AB018CE30A@karl.alexa.com> References: <41089CB27BD8D24E8385C8003EDAF7AB018CE30A@karl.alexa.com> Message-ID: <418A3F51.80907@science-computing.de> Guolin, Guolin Cheng wrote: > Hi, > > I have an ext3 file system question for a while: How to obtain failed > file (inode) from LBA section reported by Linux kernel? The following > lines are a simple example: > > ... > FCbox1: Nov 1 04:08:59 FCbox1 kernel: hda: dma_intr: status=0x51 { > DriveReady SeekComplete Error } > FCbox1: Nov 1 04:08:59 FCbox1 kernel: hda: dma_intr: error=0x40 { > UncorrectableError }, LBAsect=165843413, high=9, > low=14848469, sector=129761360 > ... > I could be wrong, but to me it seems your harddisk is dying :-( > I'd like to know the tools|commands(&subcommands) used to do the > translation below, any help are greatly appreciated. > > LBA section == ?? => logical sector in partition == ?? => block == ?? => > inode in ext3 fileSystem == find|ls => filename. Can't be much of a help w/this, but maybe smartmontools will tell you more http://smartmontools.sourceforge.net/ > > Thanks a lot. > > --Guolin > Alex From guolin at alexa.com Thu Nov 4 21:48:53 2004 From: guolin at alexa.com (Guolin Cheng) Date: Thu, 4 Nov 2004 13:48:53 -0800 Subject: Help! how to infer failed fileName|inode from LBA sector for ext3 on Fedora Linux?? Message-ID: <41089CB27BD8D24E8385C8003EDAF7AB018CE394@karl.alexa.com> Alexander, Thanks for your reply. You are right, my hard disk is dying, although it is only 3 years old. Your suggestions in former email helps me a lot. In fact, The page at http://smartmontools.sourceforge.net/BadBlockHowTo.txt answers exactly my question. Thanks. --Guolin -----Original Message----- From: ext3-users-bounces at redhat.com [mailto:ext3-users-bounces at redhat.com] On Behalf Of Alexander Apprich Sent: Thursday, November 04, 2004 6:40 AM To: For users of Fedora Core releases Cc: ext3-users at redhat.com Subject: Re: Help! how to infer failed fileName|inode from LBA sector for ext3 on Fedora Linux?? Guolin, Guolin Cheng wrote: > Hi, > > I have an ext3 file system question for a while: How to obtain failed > file (inode) from LBA section reported by Linux kernel? The following > lines are a simple example: > > ... > FCbox1: Nov 1 04:08:59 FCbox1 kernel: hda: dma_intr: status=0x51 { > DriveReady SeekComplete Error } > FCbox1: Nov 1 04:08:59 FCbox1 kernel: hda: dma_intr: error=0x40 { > UncorrectableError }, LBAsect=165843413, high=9, > low=14848469, sector=129761360 > ... > I could be wrong, but to me it seems your harddisk is dying :-( > I'd like to know the tools|commands(&subcommands) used to do the > translation below, any help are greatly appreciated. > > LBA section == ?? => logical sector in partition == ?? => block == ?? => > inode in ext3 fileSystem == find|ls => filename. Can't be much of a help w/this, but maybe smartmontools will tell you more http://smartmontools.sourceforge.net/ > > Thanks a lot. > > --Guolin > Alex _______________________________________________ Ext3-users mailing list Ext3-users at redhat.com https://www.redhat.com/mailman/listinfo/ext3-users From chun_h_ng at yahoo.com Fri Nov 5 23:01:43 2004 From: chun_h_ng at yahoo.com (Chun H. Ng) Date: Fri, 5 Nov 2004 15:01:43 -0800 (PST) Subject: Urgent question Message-ID: <20041105230143.75232.qmail@web14627.mail.yahoo.com> Hi, Recently I have started seeing a file system corruption problem. A specific partition (ext3) is corrupted intermittenly and I suspect the problem is the root inode is lost after system reboot. Have anyone seen similar problems and can give me some pointers where to look for suggestions/solutions? Thanks, Chun -------------- next part -------------- An HTML attachment was scrubbed... URL: From jacques.duplessis at videotron.ca Fri Nov 12 12:22:35 2004 From: jacques.duplessis at videotron.ca (Jacques Duplessis) Date: Fri, 12 Nov 2004 07:22:35 -0500 Subject: Enlarge ext3 Logical Volume (Filesystem) in a volume group (LVM) Message-ID: <000601c4c8b2$4b2fbc50$4a01a8c0@maison.ca> Anybody know a way to enlarge a filesystem ext3 without having to unmounted it, when they are still space left in the volume group (when using LVM) ? I will be running large production linux system running Oracle. I can't stop the database everytime I have to enlarge a filesystem. We can do it with all others filesystems (JFS, REISERSFS and XFS) when they are created in a volume group. Why can't we do it with ext3 ? I cannot believe nobody ask for it ! Somebody must have a tools (they used to be ext2online) any other ? How all others folks are dealing with this kind of problem ? -------------- next part -------------- An HTML attachment was scrubbed... URL: From perbu at linpro.no Thu Nov 18 18:28:53 2004 From: perbu at linpro.no (Per Andreas Buer) Date: Thu, 18 Nov 2004 19:28:53 +0100 Subject: Enlarge ext3 Logical Volume (Filesystem) in a volume group (LVM) In-Reply-To: <000601c4c8b2$4b2fbc50$4a01a8c0@maison.ca> (Jacques Duplessis's message of "Fri, 12 Nov 2004 07:22:35 -0500") References: <000601c4c8b2$4b2fbc50$4a01a8c0@maison.ca> Message-ID: Jacques Duplessis writes: > Anybody know a way to enlarge a filesystem ext3 without having to > unmounted it, when they are still space left in the volume group (when > using LVM) ? > > I will be running large production linux system running Oracle. > I can't stop the database everytime I have to enlarge a filesystem. Ext3 cannot do this (yet). But why don't you just create a new volume and then let Oracle utilize this? -- Per Andreas Buer From jacques.duplessis at videotron.ca Fri Nov 19 01:00:49 2004 From: jacques.duplessis at videotron.ca (Pluton Jacques) Date: Thu, 18 Nov 2004 20:00:49 -0500 Subject: Enlarge ext3 Logical Volume (Filesystem) in a volume group (LVM) In-Reply-To: Message-ID: <0I7E00B69ITOC6@VL-MO-MR011.ip.videotron.ca> We have 4 environment on our Dev. Server (Acceptance, Test, Volume, Development). Each of them have 3 filesystems (/Acc/DBF, /Acc/ARChive, /Acc/App). We need top have them separated so we control the space allocated and used by our DBA and Developper. We try to stick with the structure that we all have on all our 30 servers. It is easier to maintain. Only size of filesystems vary from system to system not a new name each time we need more space. Some scripts and some applications required to be on 1 filesystem only. If you backup image of your server onto 1 main server, the same script is used on all servers. So the image is store on 1 filesystem, it would complicate things to have let say 10 servers image in one filesystem and 12 in another.... One mount point that can grow make this sysadmin life easier and earlier at home -----Original Message----- From: Per Andreas Buer [mailto:perbu at linpro.no] Sent: Thursday, November 18, 2004 1:29 PM To: Jacques Duplessis Cc: ext3-users at redhat.com Subject: Re: Enlarge ext3 Logical Volume (Filesystem) in a volume group (LVM) Jacques Duplessis writes: > Anybody know a way to enlarge a filesystem ext3 without having to > unmounted it, when they are still space left in the volume group (when > using LVM) ? > > I will be running large production linux system running Oracle. > I can't stop the database everytime I have to enlarge a filesystem. Ext3 cannot do this (yet). But why don't you just create a new volume and then let Oracle utilize this? -- Per Andreas Buer From adilger at clusterfs.com Fri Nov 19 05:23:27 2004 From: adilger at clusterfs.com (Andreas Dilger) Date: Thu, 18 Nov 2004 22:23:27 -0700 Subject: Enlarge ext3 Logical Volume (Filesystem) in a volume group (LVM) In-Reply-To: References: <000601c4c8b2$4b2fbc50$4a01a8c0@maison.ca> Message-ID: <20041119052327.GL1974@schnapps.adilger.int> Jacques Duplessis writes: > Anybody know a way to enlarge a filesystem ext3 without having to > unmounted it, when they are still space left in the volume group (when > using LVM) ? > > I will be running large production linux system running Oracle. > I can't stop the database everytime I have to enlarge a filesystem. I think the ext3 online resizing patches are now in the 2.6-mm kernel tree (thanks to Stephen Tweedie) and new patches/tools are available for download at http://sourceforge.net/projects/ext2resize/ Cheers, Andreas -- Andreas Dilger http://members.shaw.ca/adilger/ http://members.shaw.ca/golinux/ From jacques.duplessis at videotron.ca Sat Nov 20 02:52:22 2004 From: jacques.duplessis at videotron.ca (Pluton Jacques) Date: Fri, 19 Nov 2004 21:52:22 -0500 Subject: Enlarge ext3 Logical Volume (Filesystem) in a volume group (LVM) In-Reply-To: <20041119052327.GL1974@schnapps.adilger.int> Message-ID: <0I7G00HF0INNIC@VL-MO-MR011.ip.videotron.ca> Great News . At Last ! Hopefully they will include it in the Enterprise Server Thank you ! -----Original Message----- From: Andreas Dilger [mailto:adilger at clusterfs.com] Sent: Friday, November 19, 2004 12:23 AM To: Per Andreas Buer Cc: Jacques Duplessis; ext3-users at redhat.com Subject: Re: Enlarge ext3 Logical Volume (Filesystem) in a volume group (LVM) Jacques Duplessis writes: > Anybody know a way to enlarge a filesystem ext3 without having to > unmounted it, when they are still space left in the volume group (when > using LVM) ? > > I will be running large production linux system running Oracle. > I can't stop the database everytime I have to enlarge a filesystem. I think the ext3 online resizing patches are now in the 2.6-mm kernel tree (thanks to Stephen Tweedie) and new patches/tools are available for download at http://sourceforge.net/projects/ext2resize/ Cheers, Andreas -- Andreas Dilger http://members.shaw.ca/adilger/ http://members.shaw.ca/golinux/ From jeff at jettis.com Tue Nov 23 02:05:56 2004 From: jeff at jettis.com (Jeff Dinisco) Date: Mon, 22 Nov 2004 18:05:56 -0800 Subject: files missing Message-ID: I have a 1 TB ext3 filesystem mounted via iscsi on a redhat 9 system w/ kernel version - 2.4.20-30.9. I'm not sure when it happened, but today there appears to be about 7,000 files (600GB) missing. The output from df implies that the files are still there. It shows 861 GB utilized. But du shows only 300 GB of data. I'm sure that there are no processes holding onto deleted files because I have unmounted/mounted the filesystem several times, synced, etc. Here's an excerpt from e2fsck -nf /dev/sdf ... e2fsck 1.32 (09-Nov-2002) Pass 1: Checking inodes, blocks, and sizes Inode 674 has illegal block(s). Clear? no Illegal block #32780 (2552928151) in inode 674. IGNORED. Illegal block #32781 (443979519) in inode 674. IGNORED. Illegal block #32782 (2730682564) in inode 674. IGNORED. Illegal block #32783 (1341333000) in inode 674. IGNORED. Illegal block #32784 (864228082) in inode 674. IGNORED. Illegal block #32785 (1637408843) in inode 674. IGNORED. Illegal block #32786 (2702337062) in inode 674. IGNORED. Illegal block #32787 (399755839) in inode 674. IGNORED. Illegal block #32788 (2350927161) in inode 674. IGNORED. Illegal block #32789 (972130738) in inode 674. IGNORED. Illegal block #32790 (726004907) in inode 674. IGNORED. Too many illegal blocks in inode 674. Clear inode? no [ This message basically repeats itself and eventually e2fsck errors out with... ] Error while iterating over blocks in inode 675: Illegal indirect block found e2fsck: aborted /dev/sdf: ********** WARNING: Filesystem still has errors ********** Things look pretty grim right now. As soon as I run e2fsck -p, these inodes will be deleted and I will lose data, correct? I suspect the cause of this issue was multiple hosts mounting this filesystem r/w. Does anyone know of any method that could be used to recover this data? Any help would be greatly appreciated. - Jeff -------------- next part -------------- An HTML attachment was scrubbed... URL: From sudheer at svw.com Tue Nov 23 05:24:48 2004 From: sudheer at svw.com (Sudheer Divakaran) Date: Tue, 23 Nov 2004 10:54:48 +0530 Subject: EXT3 + ACL + SAMBA problem Message-ID: <41A2C9A0.9040805@svw.com> Hi, I 've setup a SAMBA server under ext3 file system which uses the 'ACL' package ( http://acl.bestbits.at/) for restricting users. I've set 'default' permission for some directories using 'setfacl'. Everything works fine but, at times the default permission is not applied to newly created directories or files. I think that SAMBA is innocent in this case (i.e., it is purely an ACL or EXT3 filesystem problem). Because sometimes when I create directories by ssh-ing to the server, the same problem occurs. Is this a bug of EXT3 filesystem? Or Is there any workaround for this? Thanks Sudheer From tytso at mit.edu Tue Nov 23 20:47:34 2004 From: tytso at mit.edu (Theodore Ts'o) Date: Tue, 23 Nov 2004 15:47:34 -0500 Subject: files missing In-Reply-To: References: Message-ID: <20041123204734.GA11561@thunk.org> On Mon, Nov 22, 2004 at 06:05:56PM -0800, Jeff Dinisco wrote: > I have a 1 TB ext3 filesystem mounted via iscsi on a redhat 9 system w/ > kernel version - 2.4.20-30.9. I'm not sure when it happened, but today > there appears to be about 7,000 files (600GB) missing. The output from > df implies that the files are still there. It shows 861 GB utilized. > But du shows only 300 GB of data. I'm sure that there are no processes > holding onto deleted files because I have unmounted/mounted the > filesystem several times, synced, etc. Here's an excerpt from e2fsck > -nf /dev/sdf ... The output from df is calculated by using the free blocks field in the superblock, and is only as accurate as the superblock statistics will be. It does look like garbage has been written into your filesystem, and the situation is probably pretty grim; there will almost certainly be some data loss. It may not be as bad as you think, however, since there may be disconnected inodes (and possibly entire directory hierarchies) which e2fsck can recover and link into the lost+found directory. > Things look pretty grim right now. As soon as I run e2fsck -p, these > inodes will be deleted and I will lose data, correct? Actually, e2fsck -y will be needed to fix the data. E2fsck -p will make "safe" fixes, but anything that might require human judgement will cause e2fsck -p to abort (since it is intended to be used in unattended boot scripts). You will need to answer each of e2fsck's questions manually, or use e2fsck -y to cause e2fsck to automatically assume an answer of "yes" for each question. I'd strongly suggest making a bit-for-bit image backup of the filesystem before proceeding, since it might be possible for an expert to try other means of recoverying data after getting back what data you can with e2fsck -y. > I suspect the > cause of this issue was multiple hosts mounting this filesystem r/w. Ah, yup, that would do it. How long was the filesystem mounted by multiple hosts? > Does anyone know of any method that could be used to recover this data? > Any help would be greatly appreciated. You can try e2fsck -y and hope for the best. Aside from that, you can grep the disk looking for a specific text pattern if there is a few extremely valuable files that have to be recovered at all costs. Failing that, it will be restore from backup times..... (you did keep regular backups, right? :-) - Ted From jeff at jettis.com Tue Nov 23 20:59:22 2004 From: jeff at jettis.com (Jeff Dinisco) Date: Tue, 23 Nov 2004 12:59:22 -0800 Subject: files missing Message-ID: Thanks for the reply. I'm already in the process of copying the filesystem to an image using "cp /dev/sdf /another_fs/sdf.image" and I plan to fsck -y that image. Is there another method I should use? I'm not sure how long it was mounted r/w on the other host but it's pretty clear that that's the cause. Is there particular service you can recommmend for this type of data recovery (does that answer your backup question)? Thanks. -----Original Message----- From: Theodore Ts'o [mailto:tytso at thunk.org] On Behalf Of Theodore Ts'o Sent: Tuesday, November 23, 2004 12:48 PM To: Jeff Dinisco Cc: ext3-users at redhat.com Subject: Re: files missing On Mon, Nov 22, 2004 at 06:05:56PM -0800, Jeff Dinisco wrote: > I have a 1 TB ext3 filesystem mounted via iscsi on a redhat 9 system > w/ kernel version - 2.4.20-30.9. I'm not sure when it happened, but > today there appears to be about 7,000 files (600GB) missing. The > output from df implies that the files are still there. It shows 861 GB utilized. > But du shows only 300 GB of data. I'm sure that there are no > processes holding onto deleted files because I have unmounted/mounted > the filesystem several times, synced, etc. Here's an excerpt from > e2fsck -nf /dev/sdf ... The output from df is calculated by using the free blocks field in the superblock, and is only as accurate as the superblock statistics will be. It does look like garbage has been written into your filesystem, and the situation is probably pretty grim; there will almost certainly be some data loss. It may not be as bad as you think, however, since there may be disconnected inodes (and possibly entire directory hierarchies) which e2fsck can recover and link into the lost+found directory. > Things look pretty grim right now. As soon as I run e2fsck -p, these > inodes will be deleted and I will lose data, correct? Actually, e2fsck -y will be needed to fix the data. E2fsck -p will make "safe" fixes, but anything that might require human judgement will cause e2fsck -p to abort (since it is intended to be used in unattended boot scripts). You will need to answer each of e2fsck's questions manually, or use e2fsck -y to cause e2fsck to automatically assume an answer of "yes" for each question. I'd strongly suggest making a bit-for-bit image backup of the filesystem before proceeding, since it might be possible for an expert to try other means of recoverying data after getting back what data you can with e2fsck -y. > I suspect the > cause of this issue was multiple hosts mounting this filesystem r/w. Ah, yup, that would do it. How long was the filesystem mounted by multiple hosts? > Does anyone know of any method that could be used to recover this data? > Any help would be greatly appreciated. You can try e2fsck -y and hope for the best. Aside from that, you can grep the disk looking for a specific text pattern if there is a few extremely valuable files that have to be recovered at all costs. Failing that, it will be restore from backup times..... (you did keep regular backups, right? :-) - Ted From jeff at jettis.com Wed Nov 24 18:39:08 2004 From: jeff at jettis.com (Jeff Dinisco) Date: Wed, 24 Nov 2004 10:39:08 -0800 Subject: files missing Message-ID: So I copied the image then ran e2fsck -fy on it. Good news is that it restored over 6,000 files. Bad news is it restored them to lost+found w/ no file names. It's my understanding that I should be looking for directory inodes w/ tables mapping file names to inode #'s. My guess is that these inodes have been destroyed at some point which is the reason that I discovered missing files in the 1st place. They don't appear to be in lost+found. Is there a method to recover them? Thanks. -----Original Message----- From: Theodore Ts'o [mailto:tytso at thunk.org] On Behalf Of Theodore Ts'o Sent: Tuesday, November 23, 2004 12:48 PM To: Jeff Dinisco Cc: ext3-users at redhat.com Subject: Re: files missing On Mon, Nov 22, 2004 at 06:05:56PM -0800, Jeff Dinisco wrote: > I have a 1 TB ext3 filesystem mounted via iscsi on a redhat 9 system w/ > kernel version - 2.4.20-30.9. I'm not sure when it happened, but today > there appears to be about 7,000 files (600GB) missing. The output from > df implies that the files are still there. It shows 861 GB utilized. > But du shows only 300 GB of data. I'm sure that there are no processes > holding onto deleted files because I have unmounted/mounted the > filesystem several times, synced, etc. Here's an excerpt from e2fsck > -nf /dev/sdf ... The output from df is calculated by using the free blocks field in the superblock, and is only as accurate as the superblock statistics will be. It does look like garbage has been written into your filesystem, and the situation is probably pretty grim; there will almost certainly be some data loss. It may not be as bad as you think, however, since there may be disconnected inodes (and possibly entire directory hierarchies) which e2fsck can recover and link into the lost+found directory. > Things look pretty grim right now. As soon as I run e2fsck -p, these > inodes will be deleted and I will lose data, correct? Actually, e2fsck -y will be needed to fix the data. E2fsck -p will make "safe" fixes, but anything that might require human judgement will cause e2fsck -p to abort (since it is intended to be used in unattended boot scripts). You will need to answer each of e2fsck's questions manually, or use e2fsck -y to cause e2fsck to automatically assume an answer of "yes" for each question. I'd strongly suggest making a bit-for-bit image backup of the filesystem before proceeding, since it might be possible for an expert to try other means of recoverying data after getting back what data you can with e2fsck -y. > I suspect the > cause of this issue was multiple hosts mounting this filesystem r/w. Ah, yup, that would do it. How long was the filesystem mounted by multiple hosts? > Does anyone know of any method that could be used to recover this data? > Any help would be greatly appreciated. You can try e2fsck -y and hope for the best. Aside from that, you can grep the disk looking for a specific text pattern if there is a few extremely valuable files that have to be recovered at all costs. Failing that, it will be restore from backup times..... (you did keep regular backups, right? :-) - Ted From jehan.procaccia at int-evry.fr Wed Nov 24 18:52:02 2004 From: jehan.procaccia at int-evry.fr (jehan.procaccia) Date: Wed, 24 Nov 2004 19:52:02 +0100 Subject: Externalize journal Message-ID: <41A4D852.1050303@int-evry.fr> Hello Is it possible to externalize the journal of an already existing (journal inside) ext3 FS ? Here's what I did to create a new FS with external journal for /dev/emcpowerl2 on /dev/emcpowerl2 mke2fs -O journal_dev /dev/sda10 mke2fs -J device=/dev/sda10 /dev/emcpowerl2 it works perfectly , but can I do the same whitout reformating the original FS; /dev/emcpowerl1 which is in production for exemple thanks. From adilger at clusterfs.com Wed Nov 24 19:14:20 2004 From: adilger at clusterfs.com (Andreas Dilger) Date: Wed, 24 Nov 2004 12:14:20 -0700 Subject: Externalize journal In-Reply-To: <41A4D852.1050303@int-evry.fr> References: <41A4D852.1050303@int-evry.fr> Message-ID: <20041124191420.GA23661@schnapps.adilger.int> On Nov 24, 2004 19:52 +0100, jehan.procaccia wrote: > Is it possible to externalize the journal of an already existing > (journal inside) ext3 FS ? > > Here's what I did to create a new FS with external journal for > /dev/emcpowerl2 on /dev/emcpowerl2 > mke2fs -O journal_dev /dev/sda10 > mke2fs -J device=/dev/sda10 /dev/emcpowerl2 > it works perfectly , but can I do the same whitout reformating the > original FS; /dev/emcpowerl1 which is in production for exemple Use "tune2fs -O ^has_journal" on unmounted filesystem to remove the existing journal. Then "tune2fs -O has_journal -J device=/dev/sda10" to add the journal to the new device (unmounted also of course). Cheers, Andreas -- Andreas Dilger http://sourceforge.net/projects/ext2resize/ http://members.shaw.ca/adilger/ http://members.shaw.ca/golinux/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: