DBAN taking forever...is HDDErase better?

[brownstone]

I recently decided to wipe my relatively new and "error free" 250 gb hard drive. It ended up taking like 112 hrs to wipe. I'm no expert on DBAN, but does that sound right to anyone else? I also tried WD's tools that zero out the drive and that was estimated at about the same amount of time. This is a SATA Western digital 250gb drive.

 

[Modelworks]

Are you doing multiple writes over each sector ? If so stop it !
There is no need to overwrite a sector but 1 time. The other settings are for the extremely paranoid with millions of dollars in secret data that someone might place the platters under electron microscopes to recover.

 

[brownstone]

I'm pretty sure DBAN defaults to 3 overwrites, so I think that's what it did. The WD tools on the other hand just does a zero fill and that was estimated at over 100 hours. I'm not looking forward to selling my 500gb any time soon if it will take a week and a half to wipe.

Also, not to give the impression of paranoia, but I'm not convinced that 1 overwrite will do the trick. I was able to recover pictures and some music after formatting and installing a new OS over my old one. Does DBAN overcome this somehow? As I mentioned, I'm no expert in this area so I appreciate any feedback.

 

[razel]

One pass is plenty. 250GB should take between 1-2 hours. 112 hours is about .6 MB/s a sequential write should be over 50MB/s. Maybe there are errors. Did you happen to run a SMART test before hand? A quick one is mostly a quick analysis while the full one scans the entire surface. Maybe the drive isn't healthly enough to sell. As for your previous experience, a format and OS reinstall isn't enough to overwrite the entire disk.

 

[brownstone]

No SMART test was done beforehand. I did run the short and long diagnostic test using the WD tools and they came back error free.

On a side note, I am currently wiping an 80gb drive and 40gb drive and I'm looking at 10 hrs. Not great, but 120gb in 10 hrs is far better than 100+ hrs for 250.

 

[razel]

Wow Brown... it's still too many hours. Good to know the drive is fine. Good to know this occured even two different drives. Nice to know this occuredin Windows and outside windows with DBan. Have you tried a different cable/different SATA channel? Maybe IDE transfers in the BIOS are set to PIO? Acutally maybe you can try wiping using a USB enclosure? You'll be limited to about 35 MB/s, but it least it won't take you 10 hours.

Otherwise, if you got a different computer or a buddy nearby, at 10 hours it's faster to wipe using another computer... that includes removing the drive, calling them up, driving over, installing, wiping, uninstalling and driving back! Still at 10 hours, you got time to take the buddy out for a drink and probably sober up and still have plenty of time left.

 

[brownstone]

It could very well be the computer...

The 250gb is SATA, the 40 and 80gb are IDE and I tried the 40gb in a USB enclosure...no change. The only thing in common is the same computer was used. I'll try it in a different computer and report.

 

[iluvdeal]

What kind of throughput is DBAN reporting? I was doing a wipe on a 1TB SATA drive attached via USB using the Mersenne twister option (random numbers written to drive) and it was taking forever, like 1% done after 6 hours or so? I think the reported throughput was around 3300KB/s.

 

[brownstone]

That's about what mine is. I'm looking at 2600 KB/s right now on each drive (I'm doing 2 of them) and that is hooked up via IDE cable and just doing 1 pass.

 

[Red Squirrel]

Seems slow for only 250GB. I remember doing a 20+ pass "shred" (same idea as DBAN) on a 1TB drive in about a week, which translates to approx 168 hours (was not EXACTLY a week, just an estimate). I think what slows things down is the randomization algorthm. The shred command uses predetermined patterns and alternates them for each pass, then finalizes with a write of zeros.

 

[brownstone]

I was under the same impression on the randomization algorithm slowing things down so I tried the quick erase option on DBAN that just writes 0's, but it still is crawling. In fact, I get the same slow speed that I was getting with the mersenne twist 3 pass algorithm.

 

[zerogear]

A quick note: DBAN 1.0.7 is PAINFULLY slow. Try 1.0.6, it will be quite a bit faster.

 

[brownstone]

A quick note: DBAN 1.0.7 is PAINFULLY slow. Try 1.0.6, it will be quite a bit faster.

Unfortunately, I'm using 1.0.6. Anyone have any tips for running DBAN in a situation with SATA where the program loads up and after starting it, it instantly says that it has successfully completed?

 

[Modelworks]

Also, not to give the impression of paranoia, but I'm not convinced that 1 overwrite will do the trick. I was able to recover pictures and some music after formatting and installing a new OS over my old one. Does DBAN overcome this somehow? As I mentioned, I'm no expert in this area so I appreciate any feedback.

1 write is enough. If a drive didn't return the same information written to a sector it wouldn't be much good as a storage device. They did some theoretical studies where they tried to recover data that was erased and nothing was ever recovered. It remains a theory.

The difference in using something like DBAN is that DBAN overwrites sectors. Formatting under windows usually does not. Instead windows wipes the file table that basically tells windows where the files start and end, but still leaves the data. As far as windows is concerned the data is gone but it can still be recovered with another program. If you overwrite the data 1 time though it is gone for good.

One thing to watch out for is USB flash drives. They are very hard to erase properly where they cannot be recovered. I have shown many friends how easy it is to get back things that they thought they had erased from the drives.

If you want to see the raw data on a drive.
Download a copy of winhex :
http://www.winhex.com/winhex/

Run it and pick tools then open disk, then select physical disk, not logical drive.
It will show all the data on the drive.

 

[RU482]

Unfortunately, I'm using 1.0.6. Anyone have any tips for running DBAN in a situation with SATA where the program loads up and after starting it, it instantly says that it has successfully completed?

have you tried posting to the DBAN message board (if it still exists)
I seem to remember them telling me that v 1.0.6 worked great on old Pentium 3's, but anything newer (in my case, a core duo) did not work so well.

Is version 2.0.0 still around? I was able to use that on the machine that was a turtle with v 1.0.6, and it worked great

FYI, it's been about 2 years since I had to deal with the issue, so my info may be dated

 

[brownstone]

have you tried posting to the DBAN message board (if it still exists)

I checked out the forum, but it appeared to be a lot of questions with not to many answers.

I'm revising my question after doing a little research to whether HDDErase is better than DBAN? Has anyone had any experience? HDDErase sounds like it's quicker if not more secure than DBAN, but that is just from the little that I've read.

 

[razel]

HDDErase is popular with the early SSD crowd. It can erase and restore performance of degraded SSDs in a minute or two. It uses an ATA standard command secure-erase. I tried it on a regular HDD and erase speed was, no surprize, the same as a regular single pass write zero. It requires changing the HDD password which might put your disk out of warranty. It maybe also more through, but I wouldn't say it's better if you just want to simply wipe your disk and move on with life.

Have you gotten a chance to figure out your bottleneck? As others have mentioned, you DO NOT need more than a single, simple pass.

 

[brownstone]

Alright, so things today took a turn for the weird. I have another 250gb drive, same model but manufactured just a few months apart from the previously mentioned one (I'll refer to it as 250-2 for clarity....if that makes it clearer). Long story shorter, I hooked up 250-2 to a different computer and ran DBAN and it finished in 2 hours! At this point I'm thinking that the other computer must be the problem. So just to make sure, I take the original 250gb and hook it up to the 2nd computer and it runs at super slow speeds (estimated 60+ hrs to finish). So now I think it must be the drive that is bad despite tests saying that all is well. I took 250-2 and hooked it up to the original computer and it finished in 2 hrs again.

So now I'm sure that it is the drive. So I take the old 40gb drive that took 10 hrs to DBAN on the original computer and it finishes in like 45 minutes on the 2nd computer. I can see no reason or rhyme here as to what is going on! Obviously the first 250gb is running a little slow, but how do I explain the computer issues? I can't rule the original computer as having a problem because despite doing all the other drives slow it did 250-2 fast....???? Any ideas?

Nutshell:
Computer #1 - All drives done slow except for SATA 250-2 (the other drives were IDE)
Computer #2 - All drives done fast except original SATA 250

 

[Ken90630]

I glanced thru this thread, so forgive me if this has been covered, but:

Never mind SMART at this point; have you run CheckDisk on the suspected bad drive? If it's bad (unrepairable sectors or whatever), it could be causing your PC to shift down from UDMA to PIO mode and that could explain the eon's worth of time for DBAN to run on your PC and also the second PC you tried it on (it could very well have shifted the second PC to PIO also). It could also be why your SATA-2 drive finished fast in your PC -- 'cuz it isn't bad & thus doesn't cause your PC to shift down to PIO mode.

Have you checked the transfer mode from within Device Mgr? I'm sure you know it should read "Ultra DMA Mode" and not PIO.

Just a thought.

 

[brownstone]

I ran checkdisk and it found some errors and claimed that it fixed them. I went to device manager and oddly enough neither drive is listed when I look under my ATA channels > properties > advanced settings...so basically, I'm not sure how I can tell what my drive is currently running at. Any other ways to determine? In bios it doesn't show me what they are running at either.

 

[iluvdeal]

So the drives have identical model numbers? If so, is the firmware on each drive the same?

 

[brownstone]

Same model numbers, but from what I can tell the firmware is not the same. Is it possible to flash a hard drives firmware?

 

[Ken90630]

I ran checkdisk and it found some errors and claimed that it fixed them. I went to device manager and oddly enough neither drive is listed when I look under my ATA channels > properties > advanced settings...so basically, I'm not sure how I can tell what my drive is currently running at. Any other ways to determine? In bios it doesn't show me what they are running at either.

Hmmm ... that's strange that neither drive would show up under your IDE channels at all. As I'm sure you know, you should see "Drive 0" and "Drive 1" or something like that. Have you clicked on both the Primary and Secondary channel entries in the IDE ATA/ATAPI Controllers section? There's gotta be something there, I'd think.

If not, does either drive show up under Disk Management?

Lastly, have you by any chance messed with your mobo chipset drivers recently? Maybe did a reformat or something & forgot to re-install the chipset drivers? Or did a driver update recently and possibly installed incorrect chipset drivers? Just thinking out loud here.

 

[Ken90630]

Same model numbers, but from what I can tell the firmware is not the same. Is it possible to flash a hard drives firmware?

I doubt if it's a HD firmware problem, particularly if the drive worked okay before. Highly unlikely IMHO.

 

[Ken90630]

I ran checkdisk and it found some errors and claimed that it fixed them. I went to device manager and oddly enough neither drive is listed when I look under my ATA channels > properties > advanced settings...so basically, I'm not sure how I can tell what my drive is currently running at. Any other ways to determine? In bios it doesn't show me what they are running at either.

Oh, and I meant to tell you: If CheckDisk found bad sectors -- even if it fixed them -- conventional wisdom is you should plan on replacing the drive real soon. I'm not an expert on this, but those who are (or claim to be) say it means more bad sectors are soon to follow 'cuz the drive is about to die. Just FYI.