PDA

View Full Version : Failure to mount a HDD partition



prabhuvspj
29-May-2013, 09:17
Hi,

One of my HDD partition named /dev/sda2 failed to mount in any OS stating that “operation not supported”, when I tried to mount it through kernel mode, it posts the following error message.

Invalid format found in block 0
ReiserFS read locked i node
I/O failure finding stat data

Kindly requesting for suitable help if possible.

Magic31
30-May-2013, 16:06
Hi,

One of my HDD partition named /dev/sda2 failed to mount in any OS stating that “operation not supported”, when I tried to mount it through kernel mode, it posts the following error message.

Invalid format found in block 0
ReiserFS read locked i node
I/O failure finding stat data

Kindly requesting for suitable help if possible.


Do you know what the filesystem type is on that partition (e.g. is it really ReiserFS)?

Seems a filesystem check would be a first thing to do (reiserfsck), but with ReiserFS that can mean dataloss... (it's not the most stable filesystem in that regard).

Also, which OS is this, and which version/service pack level is it on?

-Willem

prabhuvspj
31-May-2013, 07:43
Hello Willem,

Thanks for your reply.

in /dev/sda2 is partitioned with ReiserFS only.
The suse OS installed in it was SLED 10 and SP2

regards
Prabhu V.

Magic31
31-May-2013, 10:31
in /dev/sda2 is partitioned with ReiserFS only.
The suse OS installed in it was SLED 10 and SP2


Hi Prabhu V,

Then it sounds like running a reiserfsck (using the SLES 10 install disc with the rescue mode option, or an something like an openSUSE live CD) is the way to get things running again.

Again a reminder that ReiserFS is not know for it's stability and recovery... so depending on how important the data on that partition is to you, you might want to first make an image (using dd or CloneZilla for example) before running the check.
There is a chance that running the check is all that's needed and your sever will boot normally again.

Fingers crossed,
Willem

prabhuvspj
31-May-2013, 12:03
Willem,

Thanks for your message.

During rescure operation using reiserfsck, i have the following message.

"The Problem has occured looks like a hardward problem. If you have bad blocks, we advice you to get a new hard drive, because once you get one bad block that the disk drive internals can't hide from your sight, the chances of getting more are generally said to become much higher (precise statistics are unknown to us, and this disk drive is probably not expensive enough for you to you to risk your time and data on it. if you don't want to follow that advice then if you have just a few bad blocks, try writing to the bad blocks and see if the drive remaps the bad blocks (that means it takes a block it has in reserve and allocates it for use for of that block number. If it cannot remap the block, use badblock option (-B) with reiserfs utils to handle this block correctly.

bread: cannot read the block 4738385: (Input/Output) error."

Kindly help on it

Magic31
31-May-2013, 12:43
Willem,

Thanks for your message.

During rescure operation using reiserfsck, i have the following message.

"The Problem has occured looks like a hardward problem. If you have bad blocks, we advice you to get a new hard drive, because once you get one bad block that the disk drive internals can't hide from your sight, the chances of getting more are generally said to become much higher (precise statistics are unknown to us, and this disk drive is probably not expensive enough for you to you to risk your time and data on it. if you don't want to follow that advice then if you have just a few bad blocks, try writing to the bad blocks and see if the drive remaps the bad blocks (that means it takes a block it has in reserve and allocates it for use for of that block number. If it cannot remap the block, use badblock option (-B) with reiserfs utils to handle this block correctly.

bread: cannot read the block 4738385: (Input/Output) error."

Kindly help on it

I think the -B option is your only option here.... again, if there is important data on this partition, you might first want to make a partition image before continuing.

What kind of disk is this that's holding the sda2 partition? Is it a single drive? Or is there a hardware/software raid configured over multiple disks?

-Willem

Magic31
31-May-2013, 12:48
I think the -B option is your only option here.... again, if there is important data on this partition, you might first want to make a partition image before continuing.

What kind of disk is this that's holding the sda2 partition? Is it a single drive? Or is there a hardware/software raid configured over multiple disks?

-Willem

These links might also be of help :

http://www.titov.net/2005/12/31/how-to-restore-reiserfs-partition-from-crashed-disk/
http://www.krazyworks.com/dealing-with-reiserfs-bad-blocks/

-Willem

prabhuvspj
31-May-2013, 13:14
Hi Willem,

Thank you for your help.

It is a single drive only.

(and regarding the other two posts will keep you posted on the progress)

thanks

prabhuvspj
01-Jun-2013, 06:56
while attempt to recovering reiserfs file system. following error followed.(ps: we tried rebulid tree before your suggestion. but that not completed.do you think that affected now?) Reiserfck --fix-fixable replaying journal: done reiserfs'/dev/sda2' in block [18..8211]:0 transactions replayed checking internal tree.. bad root block 0. (--rebuild-tree did not complete) Aborted.

Kindly do the needful

Magic31
01-Jun-2013, 09:52
while attempt to recovering reiserfs file system. following error followed.(ps: we tried rebulid tree before your suggestion. but that not completed.do you think that affected now?) Reiserfck --fix-fixable replaying journal: done reiserfs'/dev/sda2' in block [18..8211]:0 transactions replayed checking internal tree.. bad root block 0. (--rebuild-tree did not complete) Aborted.

Kindly do the needful

Do you have data on the sda2 partition? Or just the OS related bits?

If the reiserfsck can't get through due to bad blocks.... The only option I see is reinstalling the OS and retrieving what's needed from a backup. The times I've had to deal with fixing a corrupt filesystem on ReiserFS, outcome was usually to format and reinstall. :(
If reinstalling, using ext3 is still what I'd recommend with SLES 10. If you have the option, SLES 11 would be better as SLES 10 is almost EOL in terms of general support.

-Willem

prabhuvspj
03-Jun-2013, 10:22
Willem,

Thank you for your timely support.

I would like to provide some details for you to decide on this issue.

My sda2 is only data partition is not an OS data drive (my actual OS data drive is sda3 which file system is clean).

(I have received the bad root block 0 error message during the reiserfsck –fix-fixable – badblocks badblocksfile <device> option only, I can successfully complete the previous step i.e to complete badblocks –n –b 4096 -0 badblocksfile <device>)

Since I have sufficient back up of my critical files, data recovery is not prime aim.

My current OS is SLED (not SLES) 10 with SP2.

As you have recommended for ext3 up gradation from ReiserFS, is it possible to do it only for sda2 as the other hard drive’s FS (Sda1 & Sda3) are currently ok.

Kindly give me suggestion on this situation.

With great regards

Prabhu V.

Magic31
03-Jun-2013, 22:09
..
Since I have sufficient back up of my critical files, data recovery is not prime aim.
...
As you have recommended for ext3 up gradation from ReiserFS, is it possible to do it only for sda2 as the other hard drive’s FS (Sda1 & Sda3) are currently ok.


Sure, it should be no problem to reformat the sda2 partition with ext3 and restore the data to it. You don't need to touch the system partitions on sda1 and/or sda3.

What I would be careful about though, is reusing the existing disk. Are you sure the disk is in good health (and not the trigger for this corruption issue in the first place)?

Replacing the disk, as I understand it's just a single disk with no raid or other fancy stuff in the mix, is not that big a deal. There are good tools, Clonezilla being one of them, that let you copy over the working OS partitions to a new disk. To keep things simple, make sure the partitioning layout stays the same (sdX1 and sdX3 for the system partitions) as you probably them won't need to have to deal with the boot configuration.

If you are sure your current disk is in good health and working other, simply reformat the sda2 partition with ext3 (either using mkfs.ext3 on the prompt, or with YaST's partitioning utility) and restore the data needed. Change the mount point in /etc/fstab to reflect the changed filesystem.

Hope that helps,
Willem

prabhuvspj
10-Jun-2013, 10:02
Willem,

Thank you so much on your timely support.

I am having an issue with Laptop LCD screen which need to be fixed, in order to continue executing your suggestions.

Hence I will get back to you soon once my Laptop is operational.

Thanks once again.

Regards

Prabhu V.

prabhuvspj
13-Jun-2013, 10:52
Hi Willem,

On your above post, I would like to ask for more clarification on how to proceed the /etc/fstab update on reflecting the ext3 on my sda2 instead of reiserFS.

Since /etc/fstab editing is the critical part and I am little bit not sure on how to do it.

Kindly requesting for the exact procedure of the same.

Thanks.