Unattached inode

brianL · 06-24-2012, 07:10 AM

I switch off my desktop at bedtime, and on in the morning. On the past few bootups, there have been filesystem with errors warnings, and the need to do a manual fsck. This results in an unattached inode being sent to /lost+found. It turns out to be /etc/mtab:

Code:

root@slackdesk:~# cat /lost+found/#9177219
/dev/sda1 / ext4 rw,commit=0 0 0
proc /proc proc rw 0 0
sysfs /sys sysfs rw 0 0
tmpfs /dev/shm tmpfs rw 0 0

Any ideas what is causing this, and permanent solution?

TobiSGD · 06-24-2012, 07:13 AM

Looks more like /etc/fstab to me. Any ways, I recommend to download the disk manufacturer's diagnosis toll and test the disk for errors.

brianL · 06-24-2012, 07:24 AM

No, it's definitely mtab:

Code:

bash-4.1$ cat /etc/mtab
/dev/sda1 / ext4 rw,commit=0 0 0
proc /proc proc rw 0 0
sysfs /sys sysfs rw 0 0
tmpfs /dev/shm tmpfs rw 0 0

Thanks for the quick reply, I'll try your suggestion.

brianL · 06-24-2012, 07:39 AM

Went to the Samsung site, and got redirected to Seagate site. And apparently their diagnostic tools are Windows only. Won't smartctl do just as well?

TobiSGD · 06-24-2012, 08:25 AM

Samsung sold their harddisk part to Seagate some time ago. You can find the utility to test the disk here: http://www.seagate.com/support/downloads/seatools/#
Select any OS, it doesn't matter and download the DOS version, it will come as bootable ISO. I always trust the manufacturer's tool more than other tools.

brianL · 06-24-2012, 08:47 AM

OK, thanks.

brianL · 06-24-2012, 10:07 AM

It's passed both the short and long tests, using SeaTools.

TobiSGD · 06-24-2012, 11:19 AM

Does this error also occur after you have fixed the filesystem? May be it was a one-time hiccup.

brianL · 06-24-2012, 11:30 AM

It seems to happen randomly. The next time I boot, after the fix, it might be OK, or it might repeat the error. Found this in the smartctl results that looks ominous (in bold):

Code:

Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x000f   100   100   051    Pre-fail  Always       -       0
  3 Spin_Up_Time            0x0007   253   253   025    Pre-fail  Always       -       4416
  4 Start_Stop_Count        0x0032   098   098   000    Old_age   Always       -       2232
  5 Reallocated_Sector_Ct   0x0033   253   253   010    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000f   253   253   051    Pre-fail  Always       -       0
  8 Seek_Time_Performance   0x0025   253   253   015    Pre-fail  Offline      -       0
  9 Power_On_Hours          0x0032   100   100   000    Old_age   Always       -       14819
 10 Spin_Retry_Count        0x0033   253   253   051    Pre-fail  Always       -       0
 11 Calibration_Retry_Count 0x0012   253   100   000    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   099   099   000    Old_age   Always       -       1163
 13 Read_Soft_Error_Rate    0x000e   100   100   000    Old_age   Always       -       1969049
184 End-to-End_Error        0x0033   253   253   099    Pre-fail  Always       -       0
187 Reported_Uncorrect      0x0032   253   253   000    Old_age   Always       -       0
188 Command_Timeout         0x0032   253   253   000    Old_age   Always       -       0
190 Airflow_Temperature_Cel 0x0022   142   121   000    Old_age   Always       -       32 (Min/Max 11/39)
194 Temperature_Celsius     0x0022   139   118   000    Old_age   Always       -       33 (Min/Max 13/40)
195 Hardware_ECC_Recovered  0x001a   100   100   000    Old_age   Always       -       1969049
196 Reallocated_Event_Count 0x0032   253   253   000    Old_age   Always       -       0
197 Current_Pending_Sector  0x0012   253   253   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0030   253   253   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always       -       0
200 Multi_Zone_Error_Rate   0x000a   100   100   000    Old_age   Always       -       0
201 Soft_Read_Error_Rate    0x000a   253   100   000    Old_age   Always       -       0
202 Data_Address_Mark_Errs  0x0032   253   253   000    Old_age   Always       -       0

Most other results for that that I've found websearching read 0

unSpawn · 06-24-2012, 02:04 PM

brianL · 06-24-2012, 02:45 PM

Thanks, unSpawn.

brianL · 06-28-2012, 04:21 AM

Two out of the last 5 bootups have been OK, the other 3 showing the same error. Might be due to an unclean shutdown, so I'm trying this: editing /etc/rc.d/rc.6. Changing this:

Code:

# sleep 3 fixes problems with some hard drives that don't
# otherwise finish syncing before reboot or poweroff
/bin/sleep 3

To this:

Code:

# sleep 3 fixes problems with some hard drives that don't
# otherwise finish syncing before reboot or poweroff
/bin/sleep 10

See what happens over the next few boots.

unSpawn · 06-28-2012, 04:49 AM

Do the errors always pertain the same partition or directory?
Could it be interesting to see which processes hold what files open right before umount?
As in '/path/to/lsof -Pwln +D/victim/path/ > /path/to/mountpoint/logfile 2>&1'?

brianL · 06-28-2012, 05:15 AM

Yeah, I've only / and swap: /dev/sda1 is the root partition. And it's always /etc/mtab that is the unattached inode that ends up in /lost+found. So, in my case, the command you suggest would be:

Code:

/usr/bin/lsof -Pwln +/etc/mtab > /dev/sda1/logfile 2>&1

Is that right? And that would have to be run just before shutdown or reboot?
EDIT
No, I've got this bit wrong, logfile:

Code:

/dev/sda1/logfile

brianL · 06-29-2012, 04:44 AM

Changing that sleep value from 3 to 10 didn't work.