[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-ddrescue] Tip: Improving recovery chances with Seagate "F3" mod

From: Franc Zabkar
Subject: Re: [Bug-ddrescue] Tip: Improving recovery chances with Seagate "F3" models
Date: Wed, 14 May 2014 04:29:34 +1000

At 02:07 AM 14/05/14, you wrote:

or it could be implemented as a small helper application. However, it requires a little knowledge of WD's Vendor Specific Commands, or access to certain commercial tools.

It would be ideal if every drive came with a complete and free firmware tool, but meanwhile probably the most practical solution is to manually apply the fix as described in your link, and buying a drive without this problem next time.

OK, let's not bother with incorporating such a feature in ddrescue. I'm now just asking for someone to write a small external helper application that prepares the drive in much the same way that a specialised hardware imager does.

As you are no doubt aware, cloning speed can slow down to a crawl. Some drives may require 1000 hours or more with ddrescue, but only 10 hours with a hardware tool such as DeepSpar Disk Imager. Moreover, constant thrashing by software tools risks accelerating the drive's total failure, either by killing a weak head or by corrupting a firmware module. Every time that a drive has difficulty with a bad sector, it adds it to the pending list. Such lists can grow to the point where they overflow. The drive then goes offline and never comes back. A subsequent reboot reports a capacity of 0MB and the drive is essentially inaccessible. The solution is then to clear the G-lists and re-lo lists, which are firmware operations.

BTW, "ROYL" drives are not limited to "recent" models. They have been around for 10 years, so "buying a drive without this problem" is not an option. EVERY WD drive exhibits this behaviour. The firmware modification I'm proposing involves modifying one byte and clearing the re-lo list. I'm not even sure that the latter is necessary. There is already an open source tool called idle3ctl which illustrates the nature of WD's Vendor Specific Commands (VSC), and I already have an open source tool to recalculate WD's checksums. "Manually applying the fix", as you suggest, requires some significant effort and knowledge, so it is far from being a "practical solution". In fact practical commercial tools such as PC3000 (US$10,000) have a single-click solution for this problem.

Ddrescue is a great tool, and I recommend it in every storage forum, but without an appropriate helper application it will never achieve its full potential. I'm disappointed that you appear to be unwilling or incapable of accepting and understanding its limitations.

-Franc Zabkar

reply via email to

[Prev in Thread] Current Thread [Next in Thread]