HDD GURU FORUMS

Hi,

im trying to recover a slow read WD hd (applyed all the fixes possibles, mods, etc) and tryed using ddrescue (but so slowwwww take ages) and now trying hddsuperclone.

I get a decent speed,

Source: /dev/sdb Destination: /mnt/2/root/intento2/160gb.img Position: 293631866 Status: Phase1
Total LBA: 312581808 LBA to read: 312581808 Non-tried: 185873840 (7545 areas 59.464062%)
Run time: 0:12:34:21 Remaining: 1:08:39:37 n-divided: 0 (0 areas 0.000000%)
Rate: 2858 KB/s Recent: 808 KB/s Total: 1345 KB/s Bad: 0 (0 areas 0.000000%)
Skip size: 5457 Skips: 7338 Runs: 1049 Resets: 0 Recent: 5457 18 3a 55 c5 f1 45 b9 07 3e f6 4a a2 b3 f0 61 \.:U..E..>.J...a
Position: 301274982 Status: Phase1 bb 98 1d ef 94 6f 2a 89 ca 80 d2 22 03 e3 df U.....o*...."...
Finished: 129028864 (2331 areas 41.278431%) Total LBA: 312581808 LBA to read: 312581808
Non-tried: 182695472 (7833 areas 58.447250%) Rate: 30 MB/s Recent: 974 KB/s Total: 1370 KB/s
Non-trimmed: 857472 (6699 areas 0.274319%) Position: 293691002 Status: Phase1
Non-divided: 0 (0 areas 0.000000%) Non-tried: 185814704 (7546 areas 59.445145%)
Non-scraped: 0 (0 areas 0.000000%) n-divided: 0 (0 areas 0.000000%)
Bad: 0 (0 areas 0.000000%)

(sorrry for this little bad copy paste)

My question, the skips are big i think, is normal? Phase1 is making, next phases coming can check again the skipped areas?
I run with default setting.

Thx for any tip , greets!

Have you tried to send an email to Scott?

Just a guess, but I'd say you've likely got a bad/weak head and will only get a really partial recovery. Probably will ultimately need a head swap to get a good result.

Hi thx for replys, can be a damaged head, but now im phase 2, hope i can get some files,

Source: /dev/sdb Destination: /mnt/2/root/intento2/160gb.img
Total LBA: 312581808 LBA to read: 312581808 Run time: 0:22:21:48 Remaining: 4:14:11:55
Rate: 0 B/s Recent: 210 KB/s Total: 915 KB/s Skip size: 8068 Skips: 13792 Runs: 1976 Resets: 0 Recent: 21344
Position: 103587042 Status: Phase2 N Finished: 154060744 (3988 areas 49.286537%)
Non-tried: 156997205 (13354 areas 50.225956%) nNon-trimmed: 1523859 (11908 areas 0.487507%)
Non-divided: 0 (0 areas 0.000000%) Non-scraped: 0 (0 areas 0.000000%)
Bad: 0 (0 areas 0.000000%)

pclab wrote:Have you tried to send an email to Scott?

whats him email?

thx

enito wrote:

pclab wrote:Have you tried to send an email to Scott?

whats him email?

thx

Try the contact section of the HDDSuperClone website
http://www.sdcomputingservice.com/contact

The skip size is fine, the head skipping algorithm is adjusting it. From what I can tell from the display output the drive looks like it has 2 heads and one is weak/damaged/bad. After phase2 (about 50% recovery in this case) it will get slow. If the rate drops to zero after phase2 then the head is likely dead, but if it still slowly reads data then the head is weak/damaged. But a weak/damaged head could take a long time (days, weeks, months) to image, and very possibly die during the attempt. How valuable is the data?

maximus wrote:

enito wrote:

pclab wrote:Have you tried to send an email to Scott?

whats him email?

thx

Try the contact section of the HDDSuperClone website
http://www.sdcomputingservice.com/contact

The skip size is fine, the head skipping algorithm is adjusting it. From what I can tell from the display output the drive looks like it has 2 heads and one is weak/damaged/bad. After phase2 (about 50% recovery in this case) it will get slow. If the rate drops to zero after phase2 then the head is likely dead, but if it still slowly reads data then the head is weak/damaged. But a weak/damaged head could take a long time (days, weeks, months) to image, and very possibly die during the attempt. How valuable is the data?

Thx for the info. The data is not really valuable, but wanna try to get most data, i think i left the tool try to make some works. We see in phase 3 if i get more info, total are 4 phases and tool stop?

greets

Also, the fact that there are more skips than non-trimmed areas indicates that some of the skips were triggered by a slow read (2 seconds default). That is a likely sign the bad head is reading, but without seeing the log I can't be sure. It does not indicate how slow the bad head will read though, or how many errors it has. You won't know that until it gets past phase2.

Thx for the info. The data is not really valuable, but wanna try to get most data, i think i left the tool try to make some works. We see in phase 3 if i get more info, total are 4 phases and tool stop?

There are actually 6 total phases, after phase4 is trimming, then scraping. See the Theory of Operation section in the manual for more information about the different phases.
http://www.sdcomputingservice.com/hddsu ... -Operation

The thing to watch is the remaining time. It is not in any way fully accurate in situations like this, but it can give you an idea of what to expect. Once you get past phase2 it will give a better idea of the time it may take to finish. Usually the remaining time gets a bit more accurate every time it gets to the next phase, but not until the trimming phase will it give the best estimate.

Hi,now in phase 3 and so so slowww,

7a 2a 21 60 b1 b4 55 99 96 a7 80 1d 20 bd c7 c0 z*!`..U..... ... 00 00 00 00 00 00 40 00 00 00 c0 03 00 00 01 .......@........
0d 1e 28 00 67 00 c3 05 bd 15 5e 65 5e 00 61 e0 ..(.g.....^e^.a. Total LBA: 312581808 LBA to read: 312581808
06 89 d1 40 03 38 06 0d 77 31 09 91 e0 06 05 e8 ...@.8..w1...... Rate: 2807 B/s Recent: 1148 B/s Total: 595 KB/s
Source: /dev/sdb Destination: /mnt/2/root/intento2/160gb.img Position: 5635221 Status: Phase3
Total LBA: 312581808 LBA to read: 312581808 Non-tried: 153885879 (15867 areas 49.230595%)
Run time: 1:17:10:22 Remaining: 185:18:49:56 n-divided: 0 (0 areas 0.000000%)
Rate: 2811 B/s Recent: 3641 B/s Total: 507 KB/s Bad: 0 (0 areas 0.000000%)
Skip size: 4096 Skips: 17294 Runs: 2296 Resets: 0 Recent: 4096 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Position: 10939757 Status: Phase3 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Finished: 156979339 (5648 areas 50.220242%) Total LBA: 312581808 LBA to read: 312581808
Non-tried: 153723611 (16436 areas 49.178680%) Rate: 0 B/s Recent: 906 B/s Total: 595 KB/s
Non-trimmed: 1878858 (14376 areas 0.601077%) Position: 5635349 Status: Phase3
Non-divided: 0 (0 areas 0.000000%) Non-tried: 153885751 (15867 areas 49.230553%)
Non-scraped: 0 (0 areas 0.000000%) n-divided: 0 (0 areas 0.000000%)
Bad: 0 (0 areas 0.000000%)

Not sure if i left some days? or with this, nothing i cant do?

thx!

Remaining: 185:18:49:56

According to this number it would take 6 months to finish. I said it would be slow after phase2. Phase3 will do smaller skips trying to find areas that read faster, but I think that in this case you will not get much more data from it unless you let it run for a few months, assuming the drive would even last that long. I think it is time to stop and see what files you can recover from the partial image. HDDSuperClone has done what it is designed to do, get the most data from the good head(s) first, but now the bad head is reading so slow that it is not worth it to keep going in this case.

maximus wrote:I think it is time to stop and see what files you can recover from the partial image.

Judging by log the OP provided it has 2 heads, which 1 of them is bad/weak.
Unless its small files the OP is looking for, there is no point on trying.

jermy wrote:Judging by log the OP provided it has 2 heads, which 1 of them is bad/weak.
Unless its small files the OP is looking for, there is no point on trying.

That depends on what files. The disk might read about 100MB per head at the beginning and maybe 65MB at the end (basing this on the skip size, although it is an educated guess, maybe someday I will come up with the reverse formulas to correctly calculate this). That could mean possible recovery of some files such as pictures which are only a few MB in size each and would fit in the good head. But I do admit that recovery chances are not good without professional recovery, would be very lucky to get 30% of files at best, likely much less depending on what software is used for the file recovery and the nature of the files, not to mention disk fragmentation. If the OP has a spare computer and the time (months) then they could let the recovery run and see what happens...

If the OP does wish to keep going with the recovery, in this case I would suggest to change the cluster size to 1 for the rest of the recovery. With read speeds that slow there is no reason to read larger chunks, only to have to trim and scrape them later.

maximus wrote:If the OP does wish to keep going with the recovery, in this case I would suggest to change the cluster size to 1 for the rest of the recovery. With read speeds that slow there is no reason to read larger chunks, only to have to trim and scrape them later.

Hi i can try some time more, and i add a command to my line and resume?

hddsuperclone -s /dev/sdb -t /mnt/2/root/intento2/160gb.img -f /mnt/2/root/intento2/lala.log

this line im using right now, i scan the file generated:

file 160gb.img
160gb.img: x86 boot sector

dont see partitions table

greets!

Hi i can try some time more, and i add a command to my line and resume?

hddsuperclone -s /dev/sdb -t /mnt/2/root/intento2/160gb.img -f /mnt/2/root/intento2/lala.log

Yes, to continue with a cluster size of 1, add "-c 1" to the command line:

hddsuperclone -s /dev/sdb -t /mnt/2/root/intento2/160gb.img -f /mnt/2/root/intento2/lala.log -c 1

Is this drive connected as USB or directly as SATA/PATA? If connected directly as SATA/PATA then add "--ata" to the command line also. This can help some with error processing time. If USB don't bother with this option, it won't work.

maximus wrote:

Hi i can try some time more, and i add a command to my line and resume?

hddsuperclone -s /dev/sdb -t /mnt/2/root/intento2/160gb.img -f /mnt/2/root/intento2/lala.log

Yes, to continue with a cluster size of 1, add "-c 1" to the command line:

hddsuperclone -s /dev/sdb -t /mnt/2/root/intento2/160gb.img -f /mnt/2/root/intento2/lala.log -c 1

Is this drive connected as USB or directly as SATA/PATA? If connected directly as SATA/PATA then add "--ata" to the command line also. This can help some with error processing time. If USB don't bother with this option, it won't work.

Right, with c1 --ata i see more activity now, the hd is sata directly conencteroot@lubuntu:/home/lubuntu# hddsuperclone -s /dev/sdb -t /mnt/2/root/intento2/160gb.img -f /mnt/2/root/intento2/lala.log -c 1 --ata
HDDSuperClone 1.3_20161015
Logfile exists, reading it in
Checking progress log file...



65 6c 6c 61 6e 65 6f 75 73 53 79 6d 62 6f 6c 73 ellaneousSymbols
00 78 6d 6c 55 43 53 49 73 4d 69 73 63 65 6c 6c .xmlUCSIsMiscell
61 6e 65 6f 75 73 53 79 6d 62 6f 6c 73 61 6e 64 aneousSymbolsand
Source: /dev/sdb Destination: /mnt/2/root/intento2/160gb.img
Total LBA: 312581808 LBA to read: 312581808
Run time: 0:00:02:39 Remaining: 343:22:56:21
Rate: 735 B/s Recent: 2745 B/s Total: 2502 B/s
Skip size: 4096 Skips: 59 Runs: 0 Resets: 0 Recent: 0
Position: 21026668 Status: Phase3
Finished: 157049018 (6131 areas 50.242535%)
Non-tried: 153436177 (17553 areas 49.086727%)
Non-trimmed: 2096607 (15625 areas 0.670739%)
Non-divided: 0 (0 areas 0.000000%)
Non-scraped: 0 (0 areas 0.000000%)
Bad: 6 (6 areas 0.000002%)

But the strange thing i dont see the file image is growing, up..

-rw-r--r-- 1 root root 160041885184 nov 11 03:40 160gb.img

-rw-r--r-- 1 root root 160041885184 nov 11 03:41 160gb.img

Looks the time, and not growing up, insted i see activity in hddsuperclone ..

thx!!

enito wrote:
(...)

But the strange thing i dont see the file image is growing, up..

-rw-r--r-- 1 root root 160041885184 nov 11 03:40 160gb.img

-rw-r--r-- 1 root root 160041885184 nov 11 03:41 160gb.img

Looks the time, and not growing up, insted i see activity in hddsuperclone ..

thx!!

The image shouldn't grow because the tool should already had created the image to the full size with zeros or "bad" or whatever pattern filling the space on the image that could not be read on the source drive.

What are you doing now is attempting to read the data from the source drive to fill (by replacing/over/writting) the "empty" paces on the image file. You are not appending data to the end of it, you are just replacing the zeros or pattern on the middle of it with the correct data that should be there if that data can be retrieved from the source.

It would be interesting to see how this drive would image/clone using an hardware assisted tool !

The image shouldn't grow because the tool should already had created the image to the full size with zeros or "bad" or whatever pattern filling the space on the image that could not be read on the source drive.

Technically it uses sparse writes. If you write the last sector to the file Linux will show it as full size. But if you use a command that shows actual disk usage, you will see that it is only taking up the space that has been actually written.

Code:

du -ah

Did you made any progress with this drive ?