Skip to contentSkip to conversation

Broken nilfs2 filesystem

DISCUSSION Anton Eliasson 2 participants 2 messages May 27, 2013 4 messages not in the archive NILFS2 filesystem

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): 713B7146-DC0C-45AE-9ED2-30EB8F84FA57-yeENwD64cLxBDgjK7y7TUQ@public.gmane.org

#3 lore raw

Vyacheslav Dubeyko skrev 2013-05-26 14:59:

Quoted 7 lines

Hi Anton,

Quoted 1 line · Anton Eliasson

Thank you for additional details.

Quoted 9 lines

But, as I remember, Ryusuke asked to try such commands too:

$ sudo nilfs-tune -l /dev/dm-3
$ sudo dumpseg /dev/dm-3 7007
$ lssu -a /dev/dm-3

Could you share output of these commands?

My messages are being silently swallowed! Maybe your list doesn't like my attachments? This is the third attempt and this time without attachments.

Ryusuke Konishi skrev 2013-05-23 03:40:

Quoted 18 lines

Hi,

Quoted 16 lines · Anton Eliasson

Anton Eliasson skrev 2013-05-22 22:33:

Quoted 15 lines

Greetings!
It pains me to report that my /home filesystem broke down today. My system is running Arch Linux 64-bit. The filesystem resides on a Crucial M4 256 GB SSD, on top of a LVM2 volume. The drive and filesystem are both around six months old. Partition table and error log excerpts are at the bottom of this e-mail. Full logs are available upon request.

I am providing this information as a bug report. I have no reason to suspect the hardware but I cannot exclude it either. If you (the developers) are interested in troubleshooting this for prosperity, I can be your hands and run whatever tools are required. If not, I'll reformat the filesystem, restore the data from backup and forget that this happened.

Quoted 5 lines
Quoted 2 lines
Quoted 1 line

In case the formatting gets mangled, this e-mail is also available at

Right here: http://paste.debian.net/5841/

Thank you for the report.

According to the log, btree of a regular file is destroyed for some

reason.

Quoted 3 lines

I think we should look into how the btree block is broken.

Could you try the following commands to inspect the broken disk

segment ?

Quoted 2 lines
  $ sudo dd if=/dev/dm-3 bs=4k count=2048 skip=14350336 iflag=direct 

2>/dev/null | hexdump -C
There's some semi-private stuff in there so I'll e-mail it separately to Ryusuke Konishi and Vyacheslav Dubeyko.

Quoted 8 lines

This will print out blocks of the segment 7007 which includes the broken btree block.

The following commands are also useful to get debug information. Could you try them, too ?

  $ sudo nilfs-tune -l /dev/dm-3

Today (May 23) it's called dm-2 but I don't think that should matter.

nilfs-tune 2.1.5

Filesystem volume name:      home
Filesystem UUID:      e4e8bd9a-12f6-4c2a-b32f-9471f1b321fc
Filesystem magic number:  0x3434
Filesystem revision #:      2.0
Filesystem features:      (none)
Filesystem state:      invalid or mounted,error
Filesystem OS type:      Linux
Block size:          4096
Filesystem created:      Sat Oct  6 15:52:11 2012
Last mount time:      Sat May 25 10:42:30 2013
Last write time:      Sat May 25 10:42:30 2013
Mount count:          143
Maximum mount count:      50
Reserve blocks uid:      0 (user root)
Reserve blocks gid:      0 (group root)
First inode:          11
Inode size:          128
DAT entry size:          32
Checkpoint size:      192
Segment usage size:      16
Number of segments:      14039
Device size:          117771862016
First data block:      1
# of blocks per segment:  2048
Reserved segments %:      5
Last checkpoint #:      1260585
Last block address:      430080
Last sequence #:      1557848
Free blocks count:      10317824
Commit interval:      0
# of blks to create seg:  0
CRC seed:          0xfb8deb0b
CRC check sum:          0x0db18bf2
CRC check data size:      0x00000118
Quoted 1 line
  $ sudo dumpseg /dev/dm-3 7007

http://antoneliasson.se/publicdump/dumpseg-home-Anton_Eliasson-20130525.gz

Quoted 1 line
  $ lssu -a /dev/dm-3

I ran this on May 23 but haven't had time to compose this e-mail until two days ago. During that period I mounted the filesystem as rw once or twice and I unfortunately forgot to kill nilfs_cleanerd so some of the segments might have moved around. So I have rerun lssu and uploaded both outputs here:
http://antoneliasson.se/publicdump/lssu-Anton_Eliasson-20130523.gz
http://antoneliasson.se/publicdump/lssu-Anton_Eliasson-20130525.gz

Quoted 6 lines

The third command requires the device is mounted, so /home should be mounted previously with a readonly option and a norecovery option:

  $ sudo mount -t nilfs2 -o ro,norecovery /dev/dm-3 /home

Additionally, I have uploaded /var/log/everything.log spanning May 19-22 here:
http://antoneliasson.se/publicdump/everything.log.gz

The first system crash is on line 14748. On line 15829 onwards nilfs warns that an fs is unchecked and has a bad checksum. On line 16206 is the first bad btree node error. I copied the entire /var/log tree a reboot or two after I figured out that I had a bad fs. Please tell me if you need any other log files from there.

Signature · 8 lines
Best Regards
Anton Eliasson


--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

A message that is not in the archive (sent off-list or not yet received): [email protected]

#5 lore raw

Hi Anton,

On Mon, 2013-05-27 at 14:45 +0200, Anton Eliasson wrote:

[snip]

Quoted 10 lines

Additionally, I have uploaded /var/log/everything.log spanning May 19-22 here:
http://antoneliasson.se/publicdump/everything.log.gz

The first system crash is on line 14748. On line 15829 onwards nilfs warns that an fs is unchecked and has a bad checksum. On line 16206 is the first bad btree node error. I copied the entire /var/log tree a reboot or two after I figured out that I had a bad fs. Please tell me if you need any other log files from there.

Thank you for all additional details and, especially, for the system log. Your system log is really interesting. It is first time when I can see crash dump before error messages about broken bmap. It is really important detail, I suppose. So, I need to investigate all your information more deeply. Maybe, I will ask about additional information later. But, currently, I have enough and I need to think over about known issue's environment.

Thanks,
Vyacheslav Dubeyko.

--
To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html