VOGONS


First post, by AlessandroB

User metadata
Rank Oldbie
Rank
Oldbie

I have always used to check Western Digital disks (large from 250GB to 3TB and beyond) Wester Digital software called "Data Lifeguard Diagnostic" but sometimes I perceived that it was not reliable as I could hear the disk making strange noises but at the same time the software told me that there were no errors… maybe the software only checks the possibility to read the sectors and from the vaults the continuous reading noises return a correct reading value?

Reply 1 of 17, by ediflorianUS

User metadata
Rank Member
Rank
Member

use hirensboot cd dvd it has evrything you may need. MHDD and Victoria , HDAT2 , vivard , etc , if you think your plate PCB is coroded , after hdd is cooled down , clean it , let it dry and remount , for main pins contacts on pcb, you can use a pecil's eraser , and for full pcb-board you can use contact cleaner. may help for many things and interruptions. (saved a 1TB samsung recently, had crud on main contact to pcb that was not visible).

My 80486-S i66 Project

Reply 3 of 17, by weedeewee

User metadata
Rank l33t
Rank
l33t

the only benefit with using official tools is that they will often give you an error code that you can use to request an RMA.
That aside. the only official tool I had some confidence in, in finding a faulty drive was the samsung tool I once used for testing a 4GB drive.
All other tools I used at some point in time just tend to rely on the hard drive firmware smart status tables and to show they do something just do a linear read on the drive.
Some have the option to do a Read/write/read test, it's better than nothing I guess.
Some tools allow you to do a secure erase.
But mostly the tools rely on the drive to tell the software if the drive is still good or not.

If you hear the drive make odd noises, chances are high that it has some or a lot of "relocate" sectors.

Victoria is a great tool. Just getting the timing info on the sector reads can tell you a lot about the state of a hard drive.

Right to repair is fundamental. You own it, you're allowed to fix it.
How To Ask Questions The Smart Way
Do not ask Why !
https://www.vogonswiki.com/index.php/Serial_port

Reply 4 of 17, by rasz_pl

User metadata
Rank l33t
Rank
l33t

While I have no experience with WD tools, afaik Seagate SeaTools doesnt use any proprietary protocols and relies on SMART and good old read/seek tests.

https://github.com/raszpl/FIC-486-GAC-2-Cache-Module for AT&T Globalyst
https://github.com/raszpl/386RC-16 memory board
https://github.com/raszpl/440BX Reference Design adapted to Kicad
https://github.com/raszpl/Zenith_ZBIOS MFM-300 Monitor

Reply 5 of 17, by verysaving

User metadata
Rank Newbie
Rank
Newbie

Weedeewee is right, Official tools are just to check if drive is for RMA.
With Victoria and MHDD you can get lot of info on the health of the drive,
but if you want to get really inside it, download WDMARVEl demo from the
link below and take a look at the internal defect lists of the drive.
Since it's a demo you are allowed only to read from the drive, so, theoretically,
can't do any damage, but BE CAREFUL because some commands are active even
in demo mode and it'a matter of a click to render a good drive a brick!

https://wdmarvel.com/en/

This could be helpful too :

http://www.hddoracle.com/viewtopic.php?f=117&t=1393

Reply 6 of 17, by Zup

User metadata
Rank Oldbie
Rank
Oldbie

Just use MHDD / Victoria / WHDD. They read every block on disk and "time" every operation and show the results. Up to "green", usually is fine. "Orange" or worse means that a problem is developing in that zone, even before SMART catches that. A disk with "orange" blocks is not reliable.

Also, keep an eye on smart statistics... look at how many reallocated sectors (or pending reallocation sectors) has the disk and if it has increased before and after running one of those tools. SMART won't trip until those parameters reach a threshold, and sometimes that threshold is higher than desirable.

I have traveled across the universe and through the years to find Her.
Sometimes going all the way is just a start...

I'm selling some stuff!

Reply 7 of 17, by ediflorianUS

User metadata
Rank Member
Rank
Member

you can reset relocation table , after disabling smart , then format all with MHDD or HDAT2 (don't remember exactly). on old hdd-s I just use a 486 or previous to save it . (if not seen on more modern controllers than dos checkdisk-it to save the error mess). However try a smell of pcb and pcb clean before doing anything else , may have a failing part on the pcb.

My 80486-S i66 Project

Reply 8 of 17, by AlessandroB

User metadata
Rank Oldbie
Rank
Oldbie

I downloaded Victoria and it looks great, I prefer graphical interfaces over dos for equal functions and victoria seems pleasant to use. Basically I just check if the drives are trustworthy, when they start failing they are untrustworthy and trash them.

Reply 9 of 17, by Zup

User metadata
Rank Oldbie
Rank
Oldbie
AlessandroB wrote on 2023-03-14, 22:28:

I downloaded Victoria and it looks great, I prefer graphical interfaces over dos for equal functions and victoria seems pleasant to use. Basically I just check if the drives are trustworthy, when they start failing they are untrustworthy and trash them.

The advantage of victoria is that it can test USB disks (USB support in MHDD does not always work). The disadvantage is that "timing" of the blocks depends on CPU load. If your CPU is busy (i.e.: another program is working hard) it will register as longer reading times, so it will be less accurate. You can see that the first time a test is launched: some blocks will be "green", but if you stop some 5 seconds later and relaunch the test those blocks will appear as "grey".

BTW, I consider:
- "Green" or better safe (unless there are many "green" blocks).
- "Orange" are dangerous. Make a backup now, don't use that disk for anything important.
- Failed blocks: make a backup, then trash that disk.

ediflorianUS wrote on 2023-03-14, 11:00:

you can reset relocation table , after disabling smart , then format all with MHDD or HDAT2 (don't remember exactly). on old hdd-s I just use a 486 or previous to save it . (if not seen on more modern controllers than dos checkdisk-it to save the error mess). However try a smell of pcb and pcb clean before doing anything else , may have a failing part on the pcb.

IMHO, not a good suggestion. "Recovering" a disk via HDAT2 only hides the problems and some people may think that the disk is reliable again. You will only squeeze some time before that disk fails again... and your data will be gone.

I have traveled across the universe and through the years to find Her.
Sometimes going all the way is just a start...

I'm selling some stuff!

Reply 10 of 17, by AlessandroB

User metadata
Rank Oldbie
Rank
Oldbie

Some result… what you think? Is smart reliable as a test? should it be used together with the block check or is it redundant and does it matter more to do the block check?

Attachments

  • 2.jpg
    Filename
    2.jpg
    File size
    361.61 KiB
    Views
    1304 views
    File license
    CC-BY-4.0
  • 1.jpg
    Filename
    1.jpg
    File size
    489.75 KiB
    Views
    1304 views
    File license
    CC-BY-4.0

Reply 11 of 17, by ediflorianUS

User metadata
Rank Member
Rank
Member
Zup wrote on 2023-03-15, 05:58:
The advantage of victoria is that it can test USB disks (USB support in MHDD does not always work). The disadvantage is that "ti […]
Show full quote
AlessandroB wrote on 2023-03-14, 22:28:

I downloaded Victoria and it looks great, I prefer graphical interfaces over dos for equal functions and victoria seems pleasant to use. Basically I just check if the drives are trustworthy, when they start failing they are untrustworthy and trash them.

The advantage of victoria is that it can test USB disks (USB support in MHDD does not always work). The disadvantage is that "timing" of the blocks depends on CPU load. If your CPU is busy (i.e.: another program is working hard) it will register as longer reading times, so it will be less accurate. You can see that the first time a test is launched: some blocks will be "green", but if you stop some 5 seconds later and relaunch the test those blocks will appear as "grey".

BTW, I consider:
- "Green" or better safe (unless there are many "green" blocks).
- "Orange" are dangerous. Make a backup now, don't use that disk for anything important.
- Failed blocks: make a backup, then trash that disk.

ediflorianUS wrote on 2023-03-14, 11:00:

you can reset relocation table , after disabling smart , then format all with MHDD or HDAT2 (don't remember exactly). on old hdd-s I just use a 486 or previous to save it . (if not seen on more modern controllers than dos checkdisk-it to save the error mess). However try a smell of pcb and pcb clean before doing anything else , may have a failing part on the pcb.

IMHO, not a good suggestion. "Recovering" a disk via HDAT2 only hides the problems and some people may think that the disk is reliable again. You will only squeeze some time before that disk fails again... and your data will be gone.

It's for personal use on vintage stuff you already have on dvd's or backed-up . or archive web page. if you use it daily than no. I suggest not using windows apps only dos and not over USB , just on pc mb controller.

Bad hdd's never sit at 21C/23 C they never go under 35C, never, that hdd seems to be in great shape , you may need to reset the smart because of the many errors , reseat the PCB on back of hdd and disable smart from bios if you can. You are good to go. if you like win apps just try hdd tune on quick if it has many problems it will have red squares than you can stop testing. And I suggest change-ing the SATA cable of hdd. it may have caused the errors. Or SATA controller. if not than the PCB has crack joining's on data signal transmission edge of the SATA port and bettwen pcb.

My 80486-S i66 Project

Reply 12 of 17, by rasz_pl

User metadata
Rank l33t
Rank
l33t
AlessandroB wrote on 2023-03-15, 08:47:

Some result… what you think? Is smart reliable as a test? should it be used together with the block check or is it redundant and does it matter more to do the block check?

both look fine, you can run whole disk read scan manually or initialize internal SMART self tests. Dont know if victoria can do this, but https://hddscan.com does under tools/smart tests/short or extended.

https://github.com/raszpl/FIC-486-GAC-2-Cache-Module for AT&T Globalyst
https://github.com/raszpl/386RC-16 memory board
https://github.com/raszpl/440BX Reference Design adapted to Kicad
https://github.com/raszpl/Zenith_ZBIOS MFM-300 Monitor

Reply 13 of 17, by AlessandroB

User metadata
Rank Oldbie
Rank
Oldbie
rasz_pl wrote on 2023-03-15, 20:30:
AlessandroB wrote on 2023-03-15, 08:47:

Some result… what you think? Is smart reliable as a test? should it be used together with the block check or is it redundant and does it matter more to do the block check?

both look fine, you can run whole disk read scan manually or initialize internal SMART self tests. Dont know if victoria can do this, but https://hddscan.com does under tools/smart tests/short or extended.

Are you sure? Victoria say "unideal" in one of the two HD, the other one is rated "good".

Reply 14 of 17, by weedeewee

User metadata
Rank l33t
Rank
l33t

it probably says that due to the hardware ecc recovered entry in the smart table.

Can't recall if I mentioned it but smart entries aren't foolproof.
Sometimes a bad cable can cause smart failures to appear.

Right to repair is fundamental. You own it, you're allowed to fix it.
How To Ask Questions The Smart Way
Do not ask Why !
https://www.vogonswiki.com/index.php/Serial_port

Reply 15 of 17, by rasz_pl

User metadata
Rank l33t
Rank
l33t

from experience some drives just generate a ton of internal crc errors while working fine
not to be confused with dma crc, those re cable related

https://github.com/raszpl/FIC-486-GAC-2-Cache-Module for AT&T Globalyst
https://github.com/raszpl/386RC-16 memory board
https://github.com/raszpl/440BX Reference Design adapted to Kicad
https://github.com/raszpl/Zenith_ZBIOS MFM-300 Monitor

Reply 17 of 17, by ediflorianUS

User metadata
Rank Member
Rank
Member

for a full reset you can use the MHDD / HDAT2 or PartedMagic - DiskEraser*will reset the disk.
Victoria & Vivard-for scan also good (dos version) but not as great as above apps.

My 80486-S i66 Project