Hi, I have a strange problem with reiserfs and SuSE 9.1. There seem to be pairs of filenames which are not different in the same directory. An example is nb3001/mm3001: /tmp> mkdir nb3001 /tmp> mkdir mm3001 mkdir: kann Verzeichnis „mm3001“ nicht anlegen: Die Datei existiert bereits /tmp> rm -r mm3001
ls nb3001 /bin/ls: nb3001: Datei oder Verzeichnis nicht gefunden
This problem exists on any reiserfs partition an any of my SuSE 9.1 systems. If I change to ext3 everything is OK. Even reiserfs on SuSE 9.0 ist OK. No matter whether you create files ore directories. ls always shows the created file, but rm, chown, ... may use both! Some other pairs are bb3001/am3001 gb3001/fm3001 mb4001/lm4001. Any hint where the problem or better the solution is? Best Uwe
/ 2004-08-03 11:40:46 +0200 \ Uwe Debacher:
Hi,
I have a strange problem with reiserfs and SuSE 9.1. There seem to be pairs of filenames which are not different in the same directory.
An example is nb3001/mm3001: /tmp> mkdir nb3001 /tmp> mkdir mm3001 mkdir: kann Verzeichnis ?mm3001? nicht anlegen: Die Datei existiert bereits /tmp> rm -r mm3001
ls nb3001 /bin/ls: nb3001: Datei oder Verzeichnis nicht gefunden
This problem exists on any reiserfs partition an any of my SuSE 9.1 systems. If I change to ext3 everything is OK. Even reiserfs on SuSE 9.0 ist OK.
No matter whether you create files ore directories. ls always shows the created file, but rm, chown, ... may use both!
Some other pairs are bb3001/am3001 gb3001/fm3001 mb4001/lm4001.
Any hint where the problem or better the solution is?
please give exact kernel version output of debugreiserfs /dev/whatever for the working, and non-working reiserfs. just in case: did you recently run reiserfsck? does it report something before such collisions? after such collisions? Lars Ellenberg
Am Di, den 03.08.2004 schrieb Lars Ellenberg um 12:13:
please give exact kernel version 2.6.5-7.95-default
output of debugreiserfs /dev/whatever for the working, and non-working reiserfs.
debugreiserfs 3.6.13 (2003 www.namesys.com) Filesystem state: consistency is not checked after last mounting Reiserfs super block in block 16 on 0x346 of format 3.6 with standard journal Count of blocks on the device: 263056 Number of bitmaps: 9 Blocksize: 4096 Free blocks (count of blocks - used [journal, bitmaps, data, reserved] blocks): 159992 Root block: 30686 Filesystem is NOT clean Tree height: 4 Hash function used to sort names: "r5" Objectid map size 8, max 972 Journal parameters: Device [0x0] Magic [0x1280274f] Size 8193 blocks (including 1 for journal header) (first block 18) Max transaction length 1024 blocks Max batch size 900 blocks Max commit age 30 Blocks reserved by journal: 0 Fs state field: 0x0: sb_version: 2 inode generation number: 63600 UUID: d85a24d7-c02d-4009-8776-47b1073e6c0f LABEL: Set flags in SB: ATTRIBUTES CLEAN
just in case: did you recently run reiserfsck? does it report something before such collisions? after such collisions? I once checked the partition manually without any reported errors. If you have a 9.1-System with reiserfs test it
Up to now I could not find any reiserfs-partition an a SuSE 9.1-System without this problem. But while searching fo such an system I found an other thing. The problem only exists in the root of a partition. So on my systems /tmp has an own partition. If not you may have both directories in /tmp but not in / Best Uwe
/ 2004-08-03 12:49:39 +0200 \ Uwe Debacher:
If you have a 9.1-System with reiserfs test it
Up to now I could not find any reiserfs-partition an a SuSE 9.1-System without this problem.
But while searching fo such an system I found an other thing. The problem only exists in the root of a partition. So on my systems /tmp has an own partition. If not you may have both directories in /tmp but not in /
Unfortunately I have to confirm this problem. Cc feedback, secutity etc. Lars Ellenberg ------- PROBLEM ------- Confirmed name clashes in the base directory of any reiserfs mount with SuSE kernel 2.6.5+. ls -l *3001 # no such file ... mkdir mm3001 # create one ... ls -lid mm3001 nb3001 # ... get both ! 332106 drwxr-xr-x 2 root root 48 Aug 3 13:37 mm3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:37 nb3001 ----------- SYSTEM INFO ----------- Base install SuSE 9.0 // 9.1 To verify reported strange behaviour, installed one of the most recent SuSE 2.6.X kernels. Linux minna 2.6.5-7.61-lge01 #1 Wed Jun 2 21:52:29 CEST 2004 i686 i686 i386 GNU/Linux ,-- Information about the reiserfs in question: | |debugreiserfs 3.6.9 (2003 www.namesys.com) |Filesystem state: consistency is not checked after last mounting | |Reiserfs super block in block 16 on 0xfd0d of format 3.6 with standard journal |Count of blocks on the device: 1048576 |Number of bitmaps: 32 |Blocksize: 4096 |Free blocks (count of blocks - used [journal, bitmaps, data, reserved] blocks): 281844 |Root block: 597731 |Filesystem is NOT clean |Tree height: 5 |Hash function used to sort names: "r5" |Objectid map size 112, max 972 |Journal parameters: | Device [0x0] | Magic [0x2117d8e4] | Size 8193 blocks (including 1 for journal header) (first block 18) | Max transaction length 1024 blocks | Max batch size 900 blocks | Max commit age 30 |Blocks reserved by journal: 0 |Fs state field: 0x0: |sb_version: 2 |inode generation number: 110473 |UUID: 878e2a63-aa2a-4533-a2ab-e874a4df17e9 |LABEL: |Set flags in SB: | ATTRIBUTES CLEAN `--- --------- REPRODUCE --------- MNT=/usr/kernel-src mount | grep $MNT /dev/mapper/LVM--minna-kernel--src on /usr/kernel-src type reiserfs (rw) umount $MNT; reiserfsck $DEV [ no errors ] mount $MNT; cd $MNT ls -lid mm3001 nb3001 ls: mm3001: No such file or directory ls: nb3001: No such file or directory mkdir mm3001 ls -lid mm3001 nb3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:34 mm3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:34 nb3001 mkdir nb3001 mkdir: cannot create directory `nb3001': File exists rmdir nb3001 ls -lid mm3001 nb3001 ls: mm3001: No such file or directory ls: nb3001: No such file or directory # AGAIN. mkdir mm3001 ls -lid *3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:36 mm3001 ls -lid mm3001 nb3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:36 mm3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:36 nb3001 ls -lid *3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:36 mm3001 rmdir nb3001 mkdir nb3001 ls -lid *3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:37 nb3001 ls -lid mm3001 nb3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:37 mm3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:37 nb3001 ls -lid *3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:37 nb3001 rmdir nb3001 # AGAIN with plain files: same thing. set -o noclobber ls -lid *3001 ls: *3001: No such file or directory
mm3001 nb3001 bash: line 4: nb3001: cannot overwrite existing file ls -lid mm3001 nb3001 332110 -rw-r--r-- 1 root root 0 Aug 3 13:42 mm3001 332110 -rw-r--r-- 1 root root 0 Aug 3 13:42 nb3001 ls -lid *3001 332110 -rw-r--r-- 1 root root 0 Aug 3 13:42 mm3001 rm nb3001 ls -lid *3001 ls: *3001: No such file or directory
# AGAIN in subdirectory: works as should be. mkdir try; cd try set -o noclobber ls -lid *3001 ls: *3001: No such file or directory
mm3001 nb3001 ls -lid mm3001 nb3001 332109 -rw-r--r-- 1 root root 0 Aug 3 13:43 mm3001 332110 -rw-r--r-- 1 root root 0 Aug 3 13:43 nb3001 ls -lid *3001 332109 -rw-r--r-- 1 root root 0 Aug 3 13:43 mm3001 332110 -rw-r--r-- 1 root root 0 Aug 3 13:43 nb3001 rm nb3001 ls -lid *3001 332109 -rw-r--r-- 1 root root 0 Aug 3 13:43 mm3001 cd ..; rm -r try/
Unfortunately I have to confirm this problem. Cc feedback, secutity etc.
Lars Ellenberg
Thanks for the extensive report. Fixing this will take some days - by consequence, please don't expect the fix in the following kernel update. Thanks, Roman.
-----BEGIN PGP SIGNED MESSAGE----- On Dienstag, 3. August 2004 13:57 wrote Lars Ellenberg:
/ 2004-08-03 12:49:39 +0200
\ Uwe Debacher: Confirmed name clashes in the base directory of any reiserfs mount with SuSE kernel 2.6.5+.
ls -l *3001 # no such file ... mkdir mm3001 # create one ... ls -lid mm3001 nb3001 # ... get both ! 332106 drwxr-xr-x 2 root root 48 Aug 3 13:37 mm3001 332106 drwxr-xr-x 2 root root 48 Aug 3 13:37 nb3001
Seems to be a hashing problem in inode allocation procedure.
I'm currently not subscribed at the reiserfs-list, so I send it this way, but
as far as I know reiser makes extensive use of hashing. Due to the similar
spelling of the filenames I think these will generate the same hash value,
and all FS-operations relies on these hashes.
Maybe this will help to fix this bug in less time.
bye...
- --
Thomas Gaertner ( host leela )
Brandenburg Technical University at Cottbus
- ---------------------------------------------------------------
Please send only plain ASCII-Mail. In case your Mail will be turned down, use
On Tue, Aug 03, 2004 at 08:57:24PM +0200, Thomas Gaertner wrote:
Seems to be a hashing problem in inode allocation procedure.
Actually it seems to be a problem with the r5 hash method. All string pairs mentioned in previous posts produce hash collisions in the function "r5_hash" in /usr/src/linux/fs/reiserfs/hashes.c I could only check this against kernel-source-2.6.4-0 (and kernel-source-2.4.21-231, although it had no effect on the filesystem there) but according to http://www.linuxhq.com/kernel/file/fs/reiserfs/hashes.c there were no further changes. BTW, the first 2 characters of each filename are sufficient to produce a collision. -- Michel Messerschmidt lists@michel-messerschmidt.de antiVirusTestCenter, Computer Science, University of Hamburg
/ 2004-08-03 21:54:17 +0200 \ Michel Messerschmidt:
On Tue, Aug 03, 2004 at 08:57:24PM +0200, Thomas Gaertner wrote:
Seems to be a hashing problem in inode allocation procedure.
Actually it seems to be a problem with the r5 hash method. All string pairs mentioned in previous posts produce hash collisions in the function "r5_hash" in /usr/src/linux/fs/reiserfs/hashes.c I could only check this against kernel-source-2.6.4-0 (and kernel-source-2.4.21-231, although it had no effect on the filesystem there) but according to http://www.linuxhq.com/kernel/file/fs/reiserfs/hashes.c there were no further changes.
BTW, the first 2 characters of each filename are sufficient to produce a collision.
with other hash functions, the same problem will persist, though most likely with different collision strings. since the issue occurs only in the root of the reiserfs, it is likely related to the "hidden" .reiserfs_priv directory, which is the one thing that makes the root of the fs special (the other is that its parent belongs to a different superblock...). I'd investigated myself, but I'm in block devices, not a fs guy :) anyways, from a like three liner patch in the suse internal bugzilla, I guess they already tracked it down to some xattr patch related thingy, so it is likely to be in the next suse kernel update, though Roman suggested otherwise. Seems to be resolved, but needs to be confirmed, and possible further implications investigated (can you think of any?). cheers, Lars Ellenberg
On 2004-08-03 11:40, Uwe Debacher wrote:
I have a strange problem with reiserfs and SuSE 9.1. There seem to be pairs of filenames which are not different in the same directory.
An example is nb3001/mm3001: /tmp> mkdir nb3001 /tmp> mkdir mm3001 mkdir: kann Verzeichnis „mm3001“ nicht anlegen: Die Datei existiert bereits
Me too! nimrodel:~ # cd / nimrodel:/ # mkdir nb3001 nimrodel:/ # mkdir mm3001 mkdir: cannot create directory `mm3001': File exists nimrodel:/ # l *3001 total 1 drwxr-xr-x 2 root root 48 Aug 3 14:08 ./ drwxr-xr-x 31 root root 1024 Aug 3 14:08 ../ nimrodel:/ # rd mm3001 nimrodel:/ # rd nb3001 rmdir: `nb3001': No such file or directory nimrodel:/ # l nb3001 /bin/ls: nb3001: No such file or directory Surprising :-/ However, on my home partition, which is also a reiserfs partition, behaviour is different: cer@nimrodel:~> mkdir nb3001 cer@nimrodel:~> mkdir mm3001 cer@nimrodel:~> l /home/cer/??3001 /home/cer/mm3001: total 12 drwxr-xr-x 2 cer users 48 2004-08-03 14:11 ./ drwx------ 98 cer users 11832 2004-08-03 14:11 ../ /home/cer/nb3001: total 12 drwxr-xr-x 2 cer users 48 2004-08-03 14:11 ./ drwx------ 98 cer users 11832 2004-08-03 14:11 ../ cer@nimrodel:~> rd nb3001 cer@nimrodel:~> rd mm3001 cer@nimrodel:~> l /home/cer/??3001 /bin/ls: /home/cer/??3001: No such file or directory And, to answer Lars Ellenberg questions: nimrodel:~ # cat /proc/version Linux version 2.6.5-7.75-default (geeko@buildhost) (gcc version 3.3.3 (SuSE Linux)) #1 Mon Jun 14 10:44:37 UTC 2004 nimrodel:~ # mount /dev/hdb6 on / type reiserfs (rw,noatime) ... nimrodel:~ # debugreiserfs /dev/hdb6 debugreiserfs 3.6.13 (2003 www.namesys.com) Filesystem state: consistency is not checked after last mounting Reiserfs super block in block 16 on 0x346 of format 3.6 with standard journal Count of blocks on the device: 1588419 Number of bitmaps: 49 Blocksize: 4096 Free blocks (count of blocks - used [journal, bitmaps, data, reserved] blocks): 364894 Root block: 33875 Filesystem is NOT clean Tree height: 5 Hash function used to sort names: "r5" Objectid map size 8, max 972 Journal parameters: Device [0x0] Magic [0x5f4db5c4] Size 8193 blocks (including 1 for journal header) (first block 18) Max transaction length 1024 blocks Max batch size 900 blocks Max commit age 30 Blocks reserved by journal: 0 Fs state field: 0x0: sb_version: 2 inode generation number: 3894219 UUID: a1b70046-6aef-4895-a490-b7ec5ce12917 LABEL: Set flags in SB: ATTRIBUTES CLEAN I will run a reiserfsck on next reboot, but I don't expect anything there. -- Saludos Carlos Robinson
Hi, Am Di, den 03.08.2004 schrieb Carlos E. R. um 14:21:
However, on my home partition, which is also a reiserfs partition, behaviour is different:
cer@nimrodel:~> mkdir nb3001 cer@nimrodel:~> mkdir mm3001 cer@nimrodel:~> l /home/cer/??3001
maybe not /home/cer is a partition but /home Cd to /home an try again, if /home is a partition of its own. You may even try the following pairs bb3001/am3001 gb3001/fm3001 mb4001/lm4001 mb4002/lm4002 ma4001/ll4001 td3001/so3001 ma4001/ll4001 I found these files becaus this is my system for user accounts. Among my 1300 usernames I found about 10 pairs with this kind of clash! But will this be the only ones? Best Uwe
Hello,
You may even try the following pairs bb3001/am3001 gb3001/fm3001 mb4001/lm4001 mb4002/lm4002 ma4001/ll4001 td3001/so3001 ma4001/ll4001
I found these files becaus this is my system for user accounts. Among my 1300 usernames I found about 10 pairs with this kind of clash!
I also can confirm the problem (didn't test the hole bunch written above :-) Is anybody here knowing the sourcecode of reiserfs (At the moment I don't have installed the source nor I have the time to look through). I wonder if this names are reserved in some way by reiserfs, so you can't use'm. I also tried to find a correaltion between the pairs mentioned above, but I didn't see anything :-(
But will this be the only ones?
I fear there are more of this kind. Why should it be only 7 pairs, you found by chance. Somebody here who can test it on another Kernel < 2.6.5 In Kernel 2.4.x (no I did not test all of them :-) I didn't found the problem. I could imagine that it is a problem of kernel 2.6.x + reiserfs. guido
Best Uwe
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Tuesday 2004-08-03 at 15:24 +0200, Uwe Debacher wrote:
cer@nimrodel:~> l /home/cer/??3001
maybe not /home/cer is a partition but /home
True.
Cd to /home an try again, if /home is a partition of its own.
You may even try the following pairs bb3001/am3001 ...
Er... [crash] [...20 hours later] Sorry, I'm not testing anything else. Between the first test and the start of this email I had lunch, and when I started to write, the PC started to misbehave. An mc session refused to close, then Pine refused to close, then gnome refused to log out, then ctrl-alt-backspace [Ok, but not really], halt refused to halt... I could not kill anything! kill -9 did not work! I had a task tree full of defunct jobs, but the PC refused to power-off. I had to hit the power button. Rebooted, crashed again. Rescue, autorestore, crash. Rescue, crash (not booted). Install, crash. Even the DVD and CD (SuSE 9.1) refused to boot! I booted into my spare partition, 7.3, and tried to copy my '/' somewhere else. Many crashes - I think 7.3 used and older format, not fully compatible, both for reiserfs and ext3. I created a new partition, ext3. Tried to copy '/' there and reboot, impossible. Went to sleep. Awoke, memtest, nothing. Rebooted (init 1) , no problem. reiserfsck, problems detected, corrected, fix-fizable (and a few more yesterday). Then init 3, ok, copied the root partition to the new partition created yesterday, no problem. Now init 5, writing this, running ok, 3:32 hours so far. What happened!? :-/ Can a fault at the '/' partition render the rescue CD/DVD inoperable? This is the kernel log, at the time of the first crash - it is related to reiserfs - therefore I think we should all get as far of reiserfs as possible till this is fully solved and debugged: {lunch, check mail} Aug 3 16:53:31 nimrodel kernel: SFW2-INext-DROP-DEFLT IN=ppp0 OUT= MAC= SRC=81.41.86.11 DST=81.41.198.230 LEN=48 TOS=0x00 PREC=0x00 TTL=121 ID=20982 DF PROTO=TCP SPT=2931 DPT=445 WINDOW=8760 RES=0x00 SYN URGP=0 OPT (020405B401010402) Aug 3 16:53:42 nimrodel kernel: ip6table_filter: Unknown parameter `ip6table_mangle' {ppp connection down at 16:53:50} Aug 3 16:56:56 nimrodel kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000 Aug 3 16:56:56 nimrodel kernel: printing eip: Aug 3 16:56:56 nimrodel kernel: f28671fb Aug 3 16:56:56 nimrodel kernel: *pde = 00000000 Aug 3 16:56:56 nimrodel kernel: Oops: 0000 [#1] Aug 3 16:56:56 nimrodel kernel: CPU: 0 Aug 3 16:56:56 nimrodel kernel: EIP: 0060:[__crc_nf_setsockopt+2229782/3863135] Tainted: P U Aug 3 16:56:56 nimrodel kernel: EIP: 0060:[<f28671fb>] Tainted: P U Aug 3 16:56:56 nimrodel kernel: EFLAGS: 00010202 (2.6.5-7.75-default) Aug 3 16:56:56 nimrodel kernel: EIP is at reiserfs_commit_page+0xdb/0x200 [reiserfs] Aug 3 16:56:56 nimrodel kernel: eax: cc0d5b01 ebx: 00000001 ecx: 00001000 edx: cc0d5b01 Aug 3 16:56:56 nimrodel kernel: esi: 00000000 edi: 00001000 ebp: c175c200 esp: ea883d8c Aug 3 16:56:56 nimrodel kernel: ds: 007b es: 007b ss: 0068 Aug 3 16:56:56 nimrodel kernel: Process mc (pid: 18714, threadinfo=ea882000 task=c2575960) Aug 3 16:56:56 nimrodel kernel: Stack: 00000000 00000001 00000000 00000000 c711a3bc 00001000 00000000 00000000 Aug 3 16:56:56 nimrodel kernel: c1087100 c975874c 00000000 eae69fb8 00000000 eae69180 00000000 ee4dd320 Aug 3 16:56:56 nimrodel kernel: 0000000d 00000000 00000000 00000000 00000000 0000002c 00000fbc 00000fbc Aug 3 16:56:56 nimrodel kernel: Call Trace: Aug 3 16:56:56 nimrodel kernel: [__crc_nf_setsockopt+2230180/3863135] reiserfs_submit_file_region_for_write+0x69/0x2b0 [reiserfs] Aug 3 16:56:56 nimrodel kernel: [<f2867389>] reiserfs_submit_file_region_for_write+0x69/0x2b0 [reiserfs] Aug 3 16:56:56 nimrodel kernel: [__crc_nf_setsockopt+2239317/3863135] reiserfs_file_write+0x32a/0x59b [reiserfs] Aug 3 16:56:56 nimrodel kernel: [<f286973a>] reiserfs_file_write+0x32a/0x59b [reiserfs] Aug 3 16:56:56 nimrodel kernel: [do_no_page+391/1856] do_no_page+0x187/0x740 Aug 3 16:56:56 nimrodel kernel: [<c01473b7>] do_no_page+0x187/0x740 Aug 3 16:56:56 nimrodel kernel: [handle_mm_fault+267/2352] handle_mm_fault+0x10b/0x930 Aug 3 16:56:56 nimrodel kernel: [<c014863b>] handle_mm_fault+0x10b/0x930 Aug 3 16:56:56 nimrodel kernel: [do_page_fault+329/1266] do_page_fault+0x149/0x4f2 Aug 3 16:56:56 nimrodel kernel: [<c0119469>] do_page_fault+0x149/0x4f2 Aug 3 16:56:56 nimrodel kernel: [do_mmap2+197/320] do_mmap2+0xc5/0x140 Aug 3 16:56:56 nimrodel kernel: [<c010e515>] do_mmap2+0xc5/0x140 Aug 3 16:56:56 nimrodel kernel: [__crc_nf_setsockopt+2238507/3863135] reiserfs_file_write+0x0/0x59b [reiserfs] Aug 3 16:56:56 nimrodel kernel: [<f2869410>] reiserfs_file_write+0x0/0x59b [reiserfs] Aug 3 16:56:56 nimrodel kernel: [vfs_write+159/256] vfs_write+0x9f/0x100 Aug 3 16:56:56 nimrodel kernel: [<c01585ef>] vfs_write+0x9f/0x100 Aug 3 16:56:56 nimrodel kernel: [sys_write+129/208] sys_write+0x81/0xd0 Aug 3 16:56:56 nimrodel kernel: [<c0158821>] sys_write+0x81/0xd0 Aug 3 16:56:56 nimrodel kernel: [sysenter_past_esp+82/121] sysenter_past_esp+0x52/0x79 Aug 3 16:56:56 nimrodel kernel: [<c0107dc9>] sysenter_past_esp+0x52/0x79 Aug 3 16:56:56 nimrodel kernel: Aug 3 16:56:56 nimrodel kernel: Code: 8b 1e c1 eb 05 83 e3 01 0f ba 36 05 8b 44 24 14 8d 3c 08 3b {reboot, after manual reiserfsck from rescue dvd} Aug 3 17:44:41 nimrodel kernel: klogd 1.4.1, log source = /proc/kmsg started. Aug 3 17:44:41 nimrodel kernel: Inspecting /boot/System.map-2.6.5-7.75-default ... Aug 3 17:44:44 nimrodel kernel: ACPI: Power Button (FF) [PWRF] Aug 3 17:44:44 nimrodel kernel: ACPI: Sleep Button (CM) [SLPB] Aug 3 17:44:45 nimrodel kernel: ACPI: Processor [CPU0] (supports C1 C2, 2 throttling states) Aug 3 17:44:45 nimrodel kernel: ip6table_filter: Unknown parameter `ip6table_mangle' Aug 3 17:44:45 nimrodel kernel: powernow: This module only works with AMD K7 CPUs Aug 3 17:44:46 nimrodel kernel: hda: drive_cmd: status=0x51 { DriveReady SeekComplete Error } Aug 3 17:44:46 nimrodel kernel: hda: drive_cmd: error=0x04 { DriveStatusError } Aug 3 17:44:47 nimrodel kernel: hdb: drive_cmd: status=0x51 { DriveReady SeekComplete Error } Aug 3 17:44:47 nimrodel kernel: hdb: drive_cmd: error=0x04 { DriveStatusError } Aug 3 17:44:48 nimrodel kernel: Non-volatile memory driver v1.2 {those 4 hd error I get on every boot only, so I ignore.} Aug 3 17:44:49 nimrodel kernel: end_request: I/O error, dev fd0, sector 0 Aug 3 17:44:49 nimrodel kernel: end_request: I/O error, dev fd0, sector 0 Aug 3 17:44:49 nimrodel kernel: st: Version 20040318, fixed bufsize 32768, s/g segs 256 Aug 3 17:44:49 nimrodel kernel: BIOS EDD facility v0.13 2004-Mar-09, 2 devices found Aug 3 17:45:02 nimrodel kernel: bttv0: PLL: 28636363 => 35468950 .. ok Aug 3 17:45:02 nimrodel kernel: bttv0: add subdevice "remote0" Aug 3 17:45:02 nimrodel kernel: ip6table_filter: Unknown parameter `ip6table_mangle' ... Aug 3 17:49:30 nimrodel kernel: Kernel logging (proc) stopped. Aug 3 17:49:30 nimrodel kernel: Kernel log daemon terminating. ... {second crash? Not logged. Siesta pause around here} {Third crash} Aug 3 18:49:16 nimrodel k ernel: klogd 1.4.1, log source = /proc/kmsg started. Aug 3 18:49:16 nimrodel kernel: CSLIP: code copyright 1989 Regents of the University of California Aug 3 18:49:16 nimrodel kernel: PPP generic driver version 2.4.2 Aug 3 18:49:16 nimrodel kernel: ip6table_filter: Unknown parameter `ip6table_mangle' Aug 3 18:49:16 nimrodel kernel: Unable to handle kernel paging request at virtual address 00c420e8 Aug 3 18:49:16 nimrodel kernel: printing eip: Aug 3 18:49:16 nimrodel kernel: c016ebbb Aug 3 18:49:16 nimrodel kernel: *pde = 00000000 Aug 3 18:49:16 nimrodel kernel: Oops: 0000 [#1] Aug 3 18:49:16 nimrodel kernel: CPU: 0 Aug 3 18:49:16 nimrodel kernel: EIP: 0060:[__d_lookup+203/272] Not tainted Aug 3 18:49:16 nimrodel kernel: EIP: 0060:[<c016ebbb>] Not tainted Aug 3 18:49:16 nimrodel kernel: EFLAGS: 00010246 (2.6.5-7.75-default) Aug 3 18:49:16 nimrodel kernel: EIP is at __d_lookup+0xcb/0x110 Aug 3 18:49:16 nimrodel kernel: eax: 00000000 ebx: efc42080 ecx: efc439c0 edx: 00c420e4 Aug 3 18:49:16 nimrodel kernel: esi: 00000005 edi: 00000000 ebp: efc420f0 esp: eac61e58 Aug 3 18:49:16 nimrodel kernel: ds: 007b es: 007b ss: 0068 Aug 3 18:49:17 nimrodel kernel: Process getcfg (pid: 2384, threadinfo=eac60000 task=edebacd0) Aug 3 18:49:17 nimrodel kernel: Stack: eff36db4 eac61ec0 00000000 f28829c0 00000000 00000000 edcb500f 1334b807 Aug 3 18:49:17 nimrodel kernel: 00000005 efc439c0 eac61f18 eac61f18 eac61eb8 eac61ec0 c0164b58 effe4f40 Aug 3 18:49:17 nimrodel kernel: eac61f18 1334b807 edcb5014 eac61ec0 c016651d efc439c0 00000000 eac61f18 Aug 3 18:49:17 nimrodel kernel: Call Trace: Aug 3 18:49:17 nimrodel kernel: [__crc_nf_setsockopt+2342363/3863135] xattr_lookup_poison+0x0/0x50 [reiserfs] Aug 3 18:49:17 nimrodel kernel: [<f28829c0>] xattr_lookup_poison+0x0/0x50 [reiserfs] Aug 3 18:49:17 nimrodel kernel: [do_lookup+24/112] do_lookup+0x18/0x70 Aug 3 18:49:17 nimrodel kernel: [<c0164b58>] do_lookup+0x18/0x70 Aug 3 18:49:17 nimrodel kernel: [link_path_walk+1325/2480] link_path_walk+0x52d/0x9b0 Aug 3 18:49:17 nimrodel kernel: [<c016651d>] link_path_walk+0x52d/0x9b0 Aug 3 18:49:17 nimrodel kernel: [getname+107/160] getname+0x6b/0xa0 Aug 3 18:49:17 nimrodel kernel: [<c01674db>] getname+0x6b/0xa0 Aug 3 18:49:17 nimrodel kernel: [__user_walk_it+57/80] __user_walk_it+0x39/0x50 Aug 3 18:49:17 nimrodel kernel: [<c0168079>] __user_walk_it+0x39/0x50 Aug 3 18:49:17 nimrodel kernel: [vfs_lstat+84/304] vfs_lstat+0x54/0x130 Aug 3 18:49:17 nimrodel kernel: [<c0160f34>] vfs_lstat+0x54/0x130 Aug 3 18:49:17 nimrodel kernel: [filldir+0/240] filldir+0x0/0xf0 Aug 3 18:49:17 nimrodel kernel: [<c01695c0>] filldir+0x0/0xf0 Aug 3 18:49:17 nimrodel kernel: [sys_lstat64+15/48] sys_lstat64+0xf/0x30 Aug 3 18:49:17 nimrodel kernel: [<c016101f>] sys_lstat64+0xf/0x30 Aug 3 18:49:17 nimrodel kernel: [sysenter_past_esp+82/121] sysenter_past_esp+0x52/0x79 Aug 3 18:49:17 nimrodel kernel: [<c0107dc9>] sysenter_past_esp+0x52/0x79 Aug 3 18:49:17 nimrodel kernel: Aug 3 18:49:17 nimrodel kernel: Code: 39 72 04 75 c2 8b 02 39 f6 fc 89 c6 8b 7c 24 18 8b 4c 24 20 Aug 3 18:49:17 nimrodel kernel: <3>hw_random: RNG not detected Aug 3 18:49:17 nimrodel kernel: Linux agpgart interface v0.100 (c) Dave Jones Aug 3 18:49:17 nimrodel kernel: agpgart: Detected an Intel i845 Chipset. Aug 3 18:49:17 nimrodel kernel: agpgart: Maximum main memory to use for agp memory: 690M Aug 3 18:49:17 nimrodel kernel: agpgart: AGP aperture is 64M @ 0xd8000000 Aug 3 18:49:17 nimrodel kernel: 8139too Fast Ethernet driver 0.9.27 Aug 3 18:49:17 nimrodel kernel: eth0: RealTek RTL8139 at 0xf30ca000, 00:40:f4:2e:b1:21, IRQ 17 Aug 3 18:49:17 nimrodel kernel: eth0: Identified 8139 chip type 'RTL-8139C' Aug 3 18:49:17 nimrodel kernel: eth0: link down Aug 3 18:49:17 nimrodel kernel: ip6table_filter: Unknown parameter `ip6table_mangle' Aug 3 18:49:17 nimrodel kernel: NET: Registered protocol family 17 Aug 3 18:49:17 nimrodel kernel: PCI: Setting latency timer of device 0000:00:1f.5 to 64 Aug 3 18:49:17 nimrodel kernel: intel8x0_measure_ac97_clock: measured 49280 usecs Aug 3 18:49:17 nimrodel kernel: intel8x0: clocking to 48000 Aug 3 18:49:20 nimrodel kernel: ip6table_filter: Unknown parameter `ip6table_mangle' Aug 3 18:49:20 nimrodel kernel: ACPI: Power Button (FF) [PWRF] Aug 3 18:49:20 nimrodel kernel: ACPI: Sleep Button (CM) [SLPB] Aug 3 18:49:20 nimrodel kernel: ACPI: Processor [CPU0] (supports C1 C2, 2 throttling states) Aug 3 18:49:21 nimrodel kernel: powernow: This module only works with AMD K7 CPUs Aug 3 18:49:22 nimrodel kernel: hda: drive_cmd: status=0x51 { DriveReady SeekComplete Error } Aug 3 18:49:22 nimrodel kernel: hda: drive_cmd: error=0x04 { DriveStatusError } Aug 3 18:49:22 nimrodel kernel: hdb: drive_cmd: status=0x51 { DriveReady SeekComplete Error } Aug 3 18:49:22 nimrodel kernel: hdb: drive_cmd: error=0x04 { DriveStatusError } Aug 3 18:49:24 nimrodel kernel: Non-volatile memory driver v1.2 Aug 3 18:49:24 nimrodel kernel: end_request: I/O error, dev fd0, sector 0 Aug 3 18:49:24 nimrodel kernel: end_request: I/O error, dev fd0, sector 0 Aug 3 18:49:37 nimrodel kernel: bttv0: PLL: 28636363 => 35468950 .. ok Aug 3 18:49:37 nimrodel kernel: bttv0: add subdevice "remote0" Aug 3 18:49:37 nimrodel kernel: ip6table_filter: Unknown parameter `ip6table_mangle' Aug 3 18:49:51 nimrodel kernel: unmap_vmas: VMA list is not sorted correctly! Aug 3 18:50:05 nimrodel kernel: cdrom: This disc doesn't have any tracks I recognize! Aug 3 18:53:35 nimrodel kernel: Unable to handle kernel paging request at virtual address 28c7b3ca Aug 3 18:53:35 nimrodel kernel: printing eip: Aug 3 18:53:35 nimrodel kernel: f285f6fb Aug 3 18:53:35 nimrodel kernel: *pde = 00000000 Aug 3 18:53:35 nimrodel kernel: Oops: 0002 [#2] Aug 3 18:53:35 nimrodel kernel: CPU: 0 Aug 3 18:53:35 nimrodel kernel: EIP: 0060:[__crc_nf_setsockopt+2198294/3863135] Tainted: P U Aug 3 18:53:35 nimrodel kernel: EIP: 0060:[<f285f6fb>] Tainted: P U Aug 3 18:53:35 nimrodel kernel: EFLAGS: 00010212 (2.6.5-7.75-default) Aug 3 18:53:35 nimrodel kernel: EIP is at reiserfs_find_entry+0x11b/0x4b0 [reiserfs] Aug 3 18:53:35 nimrodel kernel: eax: 000001d8 ebx: ec396db0 ecx: 28c7b3b6 edx: 00000008 Aug 3 18:53:35 nimrodel kernel: esi: e75bdcfc edi: 00000000 ebp: 00000007 esp: d7847c48 Aug 3 18:53:35 nimrodel kernel: ds: 007b es: 007b ss: 0068 Aug 3 18:53:36 nimrodel kernel: Process wdm (pid: 5717, threadinfo=d7846000 task=d7864d60) Aug 3 18:53:36 nimrodel kernel: Stack: e75bdc8c e75bd090 00000007 e75bdcfc 00000006 e0d766c4 e6ff03cc 0000684d Aug 3 18:53:36 nimrodel kernel: 00023511 285accff 000001f4 00000000 00000003 e0c2dd30 ec8b7280 d7847ca8 Aug 3 18:53:36 nimrodel kernel: f288554c d7847cec e0d76648 f28615e1 d7847ca8 d7847cf8 00000000 e6ff03cc Aug 3 18:53:36 nimrodel kernel: Call Trace: Aug 3 18:53:36 nimrodel kernel: [__crc_nf_setsockopt+2206204/3863135] reiserfs_lookup+0x61/0x140 [reiserfs] Aug 3 18:53:36 nimrodel kernel: [<f28615e1>] reiserfs_lookup+0x61/0x140 [reiserfs] Aug 3 18:53:36 nimrodel kernel: [do_page_fault+329/1266] do_page_fault+0x149/0x4f2 Aug 3 18:53:36 nimrodel kernel: [<c0119469>] do_page_fault+0x149/0x4f2 Aug 3 18:53:36 nimrodel kernel: [d_alloc+27/432] d_alloc+0x1b/0x1b0 Aug 3 18:53:36 nimrodel kernel: [<c016eceb>] d_alloc+0x1b/0x1b0 Aug 3 18:53:36 nimrodel kernel: [d_lookup+27/64] d_lookup+0x1b/0x40 Aug 3 18:53:36 nimrodel kernel: [<c016ec1b>] d_lookup+0x1b/0x40 Aug 3 18:53:36 nimrodel kernel: [real_lookup+184/224] real_lookup+0xb8/0xe0 Aug 3 18:53:36 nimrodel kernel: [<c0164b18>] real_lookup+0xb8/0xe0 Aug 3 18:53:36 nimrodel kernel: [do_lookup+98/112] do_lookup+0x62/0x70 Aug 3 18:53:36 nimrodel kernel: [<c0164ba2>] do_lookup+0x62/0x70 Aug 3 18:53:36 nimrodel kernel: [link_path_walk+1325/2480] link_path_walk+0x52d/0x9b0 Aug 3 18:53:36 nimrodel kernel: [<c016651d>] link_path_walk+0x52d/0x9b0 Aug 3 18:53:36 nimrodel kernel: [open_exec+122/336] open_exec+0x7a/0x150 Aug 3 18:53:36 nimrodel kernel: [<c01624fa>] open_exec+0x7a/0x150 Aug 3 18:53:36 nimrodel kernel: [do_timer+47/1200] do_timer+0x2f/0x4b0 Aug 3 18:53:36 nimrodel kernel: [<c012629f>] do_timer+0x2f/0x4b0 Aug 3 18:53:36 nimrodel kernel: [do_execve+24/672] do_execve+0x18/0x2a0 Aug 3 18:53:36 nimrodel kernel: [<c0162b68>] do_execve+0x18/0x2a0 Aug 3 18:53:36 nimrodel kernel: [__crc_nf_setsockopt+2342363/3863135] xattr_lookup_poison+0x0/0x50 [reiserfs] Aug 3 18:53:36 nimrodel kernel: [<f28829c0>] xattr_lookup_poison+0x0/0x50 [reiserfs] Aug 3 18:53:36 nimrodel kernel: [__alloc_pages+801/864] __alloc_pages+0x321/0x360 Aug 3 18:53:36 nimrodel kernel: [<c013f771>] __alloc_pages+0x321/0x360 Aug 3 18:53:36 nimrodel kernel: [handle_mm_fault+1182/2352] handle_mm_fault+0x49e/0x930 Aug 3 18:53:36 nimrodel kernel: [<c01489ce>] handle_mm_fault+0x49e/0x930 Aug 3 18:53:36 nimrodel kernel: [handle_mm_fault+1212/2352] handle_mm_fault+0x4bc/0x930 Aug 3 18:53:36 nimrodel kernel: [<c01489ec>] handle_mm_fault+0x4bc/0x930 Aug 3 18:53:36 nimrodel kernel: [do_page_fault+329/1266] do_page_fault+0x149/0x4f2 Aug 3 18:53:36 nimrodel kernel: [<c0119469>] do_page_fault+0x149/0x4f2 Aug 3 18:53:36 nimrodel kernel: [filp_close+67/112] filp_close+0x43/0x70 Aug 3 18:53:36 nimrodel kernel: [<c01551c3>] filp_close+0x43/0x70 Aug 3 18:53:36 nimrodel kernel: [sys_execve+42/96] sys_execve+0x2a/0x60 Aug 3 18:53:36 nimrodel kernel: [<c010663a>] sys_execve+0x2a/0x60 Aug 3 18:53:36 nimrodel kernel: [sysenter_past_esp+82/121] sysenter_past_esp+0x52/0x79 Aug 3 18:53:36 nimrodel kernel: [<c0107dc9>] sysenter_past_esp+0x52/0x79 Aug 3 18:53:36 nimrodel kernel: Aug 3 18:53:36 nimrodel kernel: Code: 89 51 14 89 d1 8b 46 0e a8 01 74 03 8d 4a d4 8b 44 24 54 89 Aug 3 18:53:37 nimrodel kernel: <6>agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0. Aug 3 18:53:37 nimrodel kernel: agpgart: Putting AGP V2 device at 0000:00:00.0 into 4x mode Aug 3 18:53:37 nimrodel kernel: agpgart: Putting AGP V2 device at 0000:01:00.0 into 4x mode Aug 3 18:54:31 nimrodel kernel: cdrom: This disc doesn't have any tracks I recognize! Aug 3 18:55:38 nimrodel kernel: Unable to handle kernel paging request at virtual address 006442a4 Aug 4 09:41:06 nimrodel kernel: klogd 1.4.1, log source = /proc/kmsg started. Aug 4 09:41:06 nimrodel kernel: Inspecting /boot/System.map-2.6.5-7.75-default {Working now.} - -- Cheers, Carlos Robinson -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFBEMuUtTMYHG2NR9URAsebAJ9Tw6TSCEQVbc8hG0evu8Ho7wzUAgCfWa7A 4nY9sHFMpiWuUn9jKEUJMzo= =CzAq -----END PGP SIGNATURE-----
Hello, I'm not an expert for reiserfs (I try to avoid using it ;-) However, I have some hints for you, see below for details. Am Mittwoch, 4. August 2004 13:41 schrieb Carlos E. R.:
The Tuesday 2004-08-03 at 15:24 +0200, Uwe Debacher wrote:
You may even try the following pairs bb3001/am3001 ... Er... [crash] [...20 hours later] [...] Rebooted, crashed again. Rescue, autorestore, crash. Rescue, crash (not booted). Install, crash.
Even the DVD and CD (SuSE 9.1) refused to boot! [...] Went to sleep.
Awoke, memtest, nothing.
Hmm, otherwise this would have been my first question ;-)
Rebooted (init 1) , no problem. reiserfsck, problems detected, corrected, fix-fizable (and a few more yesterday). Then init 3, ok, copied the root partition to the new partition created yesterday, no problem. Now init 5, writing this, running ok, 3:32 hours so far.
What happened!? :-/
Maybe your harddisk starts to fail. Create a backup first und test your harddisk with "badblocks". Also have a look at /var/log/messages, if there are any interesting messages there (especially while running badblocks). AFAIK, ReiserFS can create "interesting effects" if there's a hardware failure.
Can a fault at the '/' partition render the rescue CD/DVD inoperable?
IMHO: no, it shouldn't do so. But it may do if it tries to mount the existing partitions.
This is the kernel log, at the time of the first crash - it is related to reiserfs - therefore I think we should all get as far of reiserfs as possible till this is fully solved and debugged:
Aug 3 16:56:56 nimrodel kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000 [...] Aug 3 16:56:56 nimrodel kernel: Oops: 0000 [#1]
This is a kernel oops. Not nice :-( Try to run ksymoops, this could make the output clearer.
{reboot, after manual reiserfsck from rescue dvd} Aug 3 17:44:46 nimrodel kernel: hda: drive_cmd: error=0x04 { DriveStatusError } [...] Aug 3 17:44:47 nimrodel kernel: hdb: drive_cmd: error=0x04 { DriveStatusError }
Hmm, errors with hda and hdb. Are you sure your IDE-cable is ok?
{Third crash} [...] Aug 3 18:49:16 nimrodel kernel: Oops: 0000 [#1]
Kernel oops again. See above. Gruß Christian Boltz -- Warning: Sleeping Sigmonster! Please do not disturb.
The Monday 2004-08-09 at 22:50 +0200, Christian Boltz wrote:
Even the DVD and CD (SuSE 9.1) refused to boot! [...] Went to sleep.
Awoke, memtest, nothing.
Hmm, otherwise this would have been my first question ;-)
Rebooted (init 1) , no problem. reiserfsck, problems detected, corrected, fix-fizable (and a few more yesterday). Then init 3, ok, copied the root partition to the new partition created yesterday, no problem. Now init 5, writing this, running ok, 3:32 hours so far.
What happened!? :-/
Maybe your harddisk starts to fail. Create a backup first und test your harddisk with "badblocks". Also have a look at /var/log/messages, if there are any interesting messages there (especially while running badblocks).
No, nothing there, and SMART would detect them as well. If the system finds a badblock while running, it is reported immediately on console 10, with the sector number - I know, I saw it happening time ago. There are indeed a few badblocks in one of my hard disk, but they were remapped by the HD firmware more than a year ago, and there are no new ones. That's acceptable: I can remember when all hard disks came with a few badblocks, listed on a paper sticker, when newly bought.
AFAIK, ReiserFS can create "interesting effects" if there's a hardware failure.
Yea :-( Ext2/3 is more resilient in that case.
Can a fault at the '/' partition render the rescue CD/DVD inoperable?
IMHO: no, it shouldn't do so. But it may do if it tries to mount the existing partitions.
That's my thought. It shouldn't try to check partitions till told to, because after all, it is a rescue thing.
This is the kernel log, at the time of the first crash - it is related to reiserfs - therefore I think we should all get as far of reiserfs as possible till this is fully solved and debugged:
Aug 3 16:56:56 nimrodel kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000 [...] Aug 3 16:56:56 nimrodel kernel: Oops: 0000 [#1]
This is a kernel oops. Not nice :-(
Try to run ksymoops, this could make the output clearer.
I just did. Lots of things, but useless for me - unless a kernel developer requests it here, I will not post it. What I see is problems with reiserfs, one of them from today. I still have not reformatted my root partitions, but I will have too. I'm seeing strange crashes (wdm, gnome-terminal, gkrellm), and they started just after I tried that name clashing reported on this thread, so I blame it.
{reboot, after manual reiserfsck from rescue dvd} Aug 3 17:44:46 nimrodel kernel: hda: drive_cmd: error=0x04 { DriveStatusError } [...] Aug 3 17:44:47 nimrodel kernel: hdb: drive_cmd: error=0x04 { DriveStatusError }
Hmm, errors with hda and hdb. Are you sure your IDE-cable is ok?
Yes, and I replaced it a month and a half ago. That error is reported only by SuSE 9.1, not by SuSE 8.2 (I updated in June), and only once during boot. I assume the kernel tried something, [X session just crashed on me. Nothing in the logs, except this: Aug 11 13:40:56 nimrodel gconfd (cer-5498): Received signal 15, shutting down cleanly Aug 11 13:40:56 nimrodel gconfd (cer-5498): Exiting and this in /var/log/XFree86.0.log.old Fatal server error: Caught signal 11. Server aborting As you can see, I did not even loose my email session in Pine...] What was I saying... ah. I assume the kernel tried something, failed, logged it, and continued some other way. Similar to automount testing the floppy device and failing, because there is no floppy. Ugh. Well... I'm copying my root partitio [CRASH!] I was typing the above in a text console (I hoped to write it up before switching off), but it crashed on me, badly. I could not halt. So... this night I booted to runlevel 3, and copied my root partition to a new one of ext3 type. Changed the fstab and grub files, and rebooted. It's working so far, I'll try it out harder tomorrow. If it doesn't crash, then I'll leave it as ext3 and say goodbye to reiser. Meanwhile, I'll have to get the kernel patches using SuSE 7.3, because, as I reported in SLE, modem speed in 9.1 is below 1.5Kbytes per second :-( . I want to see if that was perchance solved, and if not, I'll try another distro with a 2.6 kernel. -- Cheers, Carlos Robinson
participants (8)
-
Carlos E. R. -
Christian Boltz -
Guido Tschakert -
Lars Ellenberg -
Michel Messerschmidt -
Roman Drahtmueller -
Thomas Gaertner -
Uwe Debacher