hard drive problem during booting in Mageia 5

[aselluza]

Hi, I've done a clean install of Mageia 5 in a hard drive I use when I need to work on a computer that it's not mine. It's working fine -only Network Manager seems useless-, but past day when booting it stopped and showed an error page finishing with something like (repair:#
It created a rdsosreport.txt file. It's very long, so I copy only the last part, where it seems to appear the information about the problem:

Code: Select all

[    8.731132] EXT4-fs (sdb1): INFO: recovery required on readonly filesystem
[    8.731136] EXT4-fs (sdb1): write access will be enabled during recovery
[    8.830490] EXT4-fs (sdb1): recovery complete
[    8.831595] EXT4-fs (sdb1): mounted filesystem with ordered data mode. Opts: (null)
[    9.009204] dracut: Checking ext4: /dev/disk/by-uuid/7375073d-7bd2-4c77-95b3-7b22baec43c9
[    9.009437] dracut: issuing e2fsck -a  /dev/disk/by-uuid/7375073d-7bd2-4c77-95b3-7b22baec43c9
[    9.052184] dracut: /dev/disk/by-uuid/7375073d-7bd2-4c77-95b3-7b22baec43c9: clean, 5600/540672 files, 239655/2158478 blocks
[    9.054099] dracut: Mounting /dev/disk/by-uuid/7375073d-7bd2-4c77-95b3-7b22baec43c9 with -o noatime,acl
[    9.120119] EXT4-fs (sdb1): mounted filesystem with ordered data mode. Opts: acl
[    9.145045] dracut: Mounted root filesystem /dev/sdb1
[    9.175578] dracut: Checking ext4: /dev/disk/by-uuid/52470cb7-d745-4e48-9a16-d1e7c3cbe2d3
[    9.175845] dracut: issuing e2fsck -a  /dev/disk/by-uuid/52470cb7-d745-4e48-9a16-d1e7c3cbe2d3
[    9.307073] dracut Warning: e2fsck returned with 4
[    9.308234] dracut Warning: /dev/disk/by-uuid/52470cb7-d745-4e48-9a16-d1e7c3cbe2d3: Superblock needs_recovery flag is clear, but journal has data.
[    9.308469] dracut Warning: /dev/disk/by-uuid/52470cb7-d745-4e48-9a16-d1e7c3cbe2d3: Run journal anyway
[    9.308872] dracut Warning: *** An error occurred during the file system check.
[    9.309012] dracut Warning: *** Dropping you to a shell; the system will try
[    9.309140] dracut Warning: *** to mount the filesystem(s), when you leave the shell.
[    9.309469] dracut Warning: filesystem)
+ '[' -f /run/initramfs/init.log ']'

First times I thought there was nothing I could do but shutting down manually there as it didn't let me use commands as reboot. I reinstalled Mageia 5 formatting sdb1, where the problem seems to be. It booted and worked fine for a couple of sessions and then the problem appeared again. I tried to reinstall but not formatting sdb1 again, and it also worked, it booted again fine, but after shutting down and booting back the problem is again there. Now I discovered I only need to type "exit" in that command line it offers me, and then it continues booting and working, but I wonder why that problem appears all the time, if it can be something physical in the hard drive or something that can be repair.

Thanks!!

[nigelc]

Hello,
try a

Code: Select all

e2fsck  -y  -f  /dev/disk/by-uuid/52470cb7-d745-4e48-9a16-d1e7c3cbe2d3
exit


[aselluza]

I tried and this is the result...

Code: Select all

e2fsck 1.42.12 (29-Aug-2014)
/dev/disk/by-uuid/52470cb7-d745-4e48-9a16-d1e7c3cbe2d3 is mounted.
e2fsck: Cannot continue, aborting.


[jiml8]

I would be viewing that hard drive with deep suspicion. Try running badblocks on it and see what is returned. Make sure you don't use the -w option, unless you want to have to reload the system again.

Badblocks will take a long time to run. Could be many hours if the hard drive is large.

[aselluza]

I don't know how to do that, could yo halp me? (I also guess the problem is in the hard drive but I don't know anything about how to fix it....) thanks!!

[jiml8]

Actually, smartctl would be a lot quicker and might give an indication of the drive's condition.

Download the smartmontools package, and run (as root) the command "smartctl -a /dev/sdN" where sdN is the specific drive you are testing (sda I presume). Copy/paste the results here.

Then, enter the command "smartctl -t long /dev/sdN" to run the long self-test. Wait for this test to finish, then again run "smartctl -a /dev/sdN" to get the results.

To run badblocks on the device, just run (as root) the command "badblocks /dev/sdN" and wait for the result, which probably will take hours.

[filip]

I tried and this is the result...

Code: Select all

e2fsck 1.42.12 (29-Aug-2014)
/dev/disk/by-uuid/52470cb7-d745-4e48-9a16-d1e7c3cbe2d3 is mounted.
e2fsck: Cannot continue, aborting.


While I suspect the hard drive too this output only means that the drive is mounted. By no means there's any information about it's state in this output only. So I recommend repeating that with the drive unmounted.

And BTW there's a GUI for smartctl called GSmartControl if you're more used to that.

[doktor5000]

the problem is that I can't unmount it, as it's the root partition of Mageia 5... I'll try the other options and let's see what happen... I'll do it tomorrow, I must leave now. thanks!!

[filip]

the problem is that I can't unmount it, as it's the root partition of Mageia 5...

You can boot with LiveCD or Live DVD.

[aselluza]

I did first a short self-test, no errors, and then tried three times long ones, but all of them were interrupted, this is the result -only last part of it-:

Code: Select all

SMART Error Log Version: 1
No Errors Logged

SMART Self-test log structure revision number 1
Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Extended offline    Interrupted (host reset)      90%     10395         -
# 2  Extended offline    Interrupted (host reset)      90%     10394         -
# 3  Extended offline    Interrupted (host reset)      80%     10394         -
# 4  Short offline       Completed without error       00%     10394         -

SMART Selective self-test log data structure revision number 1
 SPAN  MIN_LBA  MAX_LBA  CURRENT_TEST_STATUS
    1        0        0  Not_testing
    2        0        0  Not_testing
    3        0        0  Not_testing
    4        0        0  Not_testing
    5        0        0  Not_testing
Selective self-test flags (0x0):
  After scanning selected spans, do NOT read-scan remainder of disk.
If Selective self-test is pending on power-up, resume after 0 minute delay.


any idea? maybe I'll try badblocks at night... thanks again!!

[doktor5000]

The short self-test is not that relevant, simply run smartctl -a /dev/sdN like jiml8 suggest in viewtopic.php?p=57969#p57969

[aselluza]

The short self-test is not that relevant, simply run smartctl -a /dev/sdN like jiml8 suggest in viewtopic.php?p=57969#p57969

I did it and there's no relevant information about any error in the har drive, and when I tried doing a long self-test I got the result I copied above. If you read what I wrote, I tried three times long self-test as jiml8 suggested, and the three times it was interrumpted (host reset) between 10 and 20%, but no error information there...

[doktor5000]

If you think smartctl -a does only show irrelevant information, so be it. If you don't share them nobody can check that. And nobody can help you further.
Best hint is: Get a new harddrive, safest bet.

[jiml8]

Host reset on the long test generally means the machine rebooted before the test finished. Is this what happened? Why the reboots?

And, yes. If you want our help you have to copy/paste the results of smartctl -a