Surface scan during Disk Utility's Erase?

Not to my knowledge. I can't find any documented evidence of it, anyway. AFAIK the last Apple utility that would perform a surface scan was Drive Setup, which ran in OS 9. If you have AppleCare, you have a copy of TechTool Deluxe which has a surface scan function for OS X.

AFAIK, DU does map out bad blocks when writing zeros to the HD. Whether or not that constitutes a surface scan, I'll leave to the experts.

AFAIK, DU does map out bad blocks when writing zeros to the HD.

Hi baltwo,

I had believed that as well, but after searching today I could find no documented evidence of it anywhere on Apple's site. The only thing I could find on Apple's support site referenced TechTool Deluxe for this task. If you have more info I'd love to hear about it. 🙂

Nope, but, IIRC, I've seen that comment associated with discussions about DU's writing zeros option. See http://forums.dealmac.com/read.php?4,2620022 for example.

I have seen similar comments made about DU from many unofficial sources, and tended to believe them. It would make sense, after all. But in the absence of any documented evidence from Apple I now consider them to be hearsay.

To learn more than I care to know about Disk Utility and the Mac OS native format processes, take a look at this article in Apple Developer Connection.

http://developer.apple.com/technotes/tn/tn1150.html#BadBlockFile

Here's the relevant paragraph:

Bad Block File
The extent overflow file is also used to hold information about the bad block file. The bad block file is used to mark areas on the disk as bad, unable to be used for storing data. This is typically used to map out bad sectors on the disk.

Looks to me like mapping out bad blocks is a standard part of the erase and format procedure.
Francine


Francine
Schwieder

I suppose that it's implied, since Disk Utility is what one normally uses to initialize HDs.

When an HFS Plus volume is embedded within an HFS wrapper (the way Mac OS normally initializes a hard disk), the space used by the HFS Plus volume is marked as part of the bad block file within the HFS wrapper itself.

The disk drive does the mapping out, and zero all may, or may not, catch a weak sector. A 7-way erase does, in my experience will and has.

MicroMat (TechTool Pro) surface scan of media will alert on detecting a sector but doesn't take any other action. Has basic SMART tests.

Speedtools by Intech has a media scanner that has the option to map out, but my results with it are mixed at best. I had better luck using Windows chkdsk with option to attempt to map out bad blocks. It also has an extensive SMART test.

Zero all is not writing patterns and you really need to see if bits can be switched, which is why a hex x'FF' zeroes and ones might work better.

SoftRAID will alert if there is an error, and you can specify if you only want to know about unrecoverable after multiple atempts.

Disk Warrior has a test, writes the smart status line to system log, and will show how many spare blocks, how many used, how many still reserved, which in one study was found to be a good indicator of when to replace a drive if it starts using from the pool of spares.

Would I rely on the zero-all? that is what Micromat recommends to start with, doesn't take 2 hrs per 100GB like 7-way write will. And subjecting a drive to 7-way, either fix it or kill it? I know those old low=level formats would really heat up SCSI drive to point it would be too hot to touch.

Vista detected a bad block, wouldn't do backup until it was fixed and the partition reformatted - there was 4k file on bad block(s).

The tech notes in ADC contain information that developers need to know to write programs. In this case it is the information necessary to write a disk utility program for working with disks for the Mac, using the HFS+ file system. Which means it is the same information that the Apple engineers used in writing Apple's own program, to wit Disk Utility. I have no desire to write a program, ever, nor any pressing curiousity about disk formats, so did not read the article, merely scanned it for relevant information. But it looked to me like it was saying that mapping out bad blocks is an integral part of erasing and formatting a disk for HFS+ and thus would automatically happen when the program does that thing. Disk Utility does that, therefore bad blocks are indeed taken care of when you erase and format a disk using Disk Utility. I don't know whether one would call this "scanning" or not. But bad blocks are found and removed from use for storing data.
Francine


Francine
Schwieder

No. The drive itself gets an I/O request to do something. DU says "write zero" and unless it gets an invalid read, goes to next block. That to me is not really "testing" by my definition of "test."

Disk drives have their own DSP processors and handle - and have handled - all of the requests. At one time, Apple HD Setup actually did try to "low level" on ATA and of course was quickly updated and the feature - which was designed for SCSI, removed.

The way I read the information in the tech note is that you won't find a "list" of bad blocks, but erasing and formatting the disk does find the bad blocks and marks them as used in such a way that they can not then be used for storing anything else. In short, they are mapped out. So I suppose it is technically incorrect to say Disk Utility "scans" the disk. But for practical purposes it does find the bad blocks and in effect disables them, so nothing you put on the disk will be corrupted by being written to a bad block. Evidently you are interested in something else. I think all most people would want to know is whether using Disk Utility to erase and format a disk will result in bad blocks being eliminated from use, so that the things they then put on the disk will not be corrupted. To that the answer is: yes, bad blocks are found and treated in such a way that no regular files will ever be written to those areas of the disk.
Francine


Francine
Schwieder