Fix-disk output meaning...

[phd]

hi,
I had a problem last night where the hard disk contents could not be viewed from the media menu, with a processing dialog that would not disappear.
Trying to access the diagnostics section from the web-interface seemed to lock it up as well.
The hard disk is visible, as it shows the free space.
I did a fix-disk a few weeks ago, but had to stop it due to it taking too long.
I've just done it again over night, hoping to get viewable access again.
The output seems to have got to a similar place as it did before when I stopped it, (see picture).
It says there there are over 3600 multiply-claimed blocks, and is starting to clone them, in batches.
It's done less than 20, and seems to take forever to clone.
What exactly is this supposed to be doing?
Is it really going to need to do another 3600 before it completes?
Should I just stop it now, and hope it has recovered sufficiently?
Are there other commands I should run to try and help sort the hard disk out?
What command should I run to reinstall the custom packages to make sure they are not corrupted?

Any help would be appreciated.

Many thank,
Phd.

 

[Black Hole]

I can't comment on the rest, but:

What command should I run to reinstall the custom packages to make sure they are not corrupted?

That's the least of your worries!

Any idea how full the disk is? If the fix process is trying to disentangle corrupted pointers but there's little to no room to work in...

 

[phd]

It's a 2tb hard disk with over 500gb free.

 

[phd]

Can I connect a second HDD internally via Sata cable in order to transfer file from one to the other as a backup, then reformat original?
If so what's the best way to copy the stuff off? cp -R?

Do you think I should just abort the e2fsck now?

 

[Black Hole]

I would wait for the experts to advise.

No, you can't connect a second drive internally, but if you are in a hurry I suggest replacing the existing drive with a new one, then pulling stuff over by USB.

 

[prpr]

Do you think I should just abort the e2fsck now?

Probably. It might be easier (less time consuming) to copy everything off and reformat.
You need to post your SMART stats. though first.

 

[phd]

It would seem that there is one specific directory that seems to be causing the main locking up problem. If I try to ls in that directory is just hangs, same when thing trying to access via ftp. Is there a way to recover that one specific directory via it's inode, using superblocks / alternate hard disk journal?
How do I get the SMART stats?

 

[Black Hole]

If the WebIF is functioning, go to the Diagnostics page.

 

[prpr]

How do I get the SMART stats?

From a command line session (telnet/ssh/Webshell):

humax# smartctl -a /dev/sda

assuming no USB disks plugged in (try sdb instead if so).