AVS Forum banner

Status
Not open for further replies.
1 - 16 of 16 Posts

·
Registered
Joined
·
50 Posts
Discussion Starter #1
I am still having problems with my hacked 80GB Maxtor drive. My Showstopper freezes up occasionally (once every week or two). Sometimes I can reset and it comes backs but once in a while, it needs to be rebuilt. I have been running with the cover off for a couple of weeks so I don't think it is heat. I went back and did a low level format after the first time it crashed and that did not help.


This time I went in to turn off the write verification. But, it seems like it was already turned off. The program seems like you set write verification for a defined amount of power cycles and then it turns itself off. My question is: If the write verification was off when I did the low level format, would that isolate the bad sectors? My latest hope is that if I do the low level format again and make sure write verification is on, it will fix the bad sectors on the drive. I can't imagine what else would cause these intermittent failures. Any suggestions would be appreciated. I love having the extra disk space but the crashes are annoying.
 

·
Registered
Joined
·
4,612 Posts
I was questioning the wvset /off thing myself. I emailed Maxtor and expect an answer in a few days. I'll post back with Maxtors reply.
 

·
Registered
Joined
·
3,715 Posts
Quote:
Originally posted by mendels
My latest hope is that if I do the low level format again and make sure write verification is on, it will fix the bad sectors on the drive.
Low level formatting is unaffected by whether write verification is on or off. Bad sectors will not be fixed or returned for use, they will be flagged as bad so they will not be used in the future.

I think Maxtor normally has write verification set to on for a certain number of power cycles (10?) so that data on a new drive will be verified for a period of time. Write verification slows a drive down somewhat and when a drive is working properly is normally not needed.


-Gary
 

·
Registered
Joined
·
4,612 Posts
As for the crashing do you have any clues?

Is the 80 gig installed where the original drive was?

Is memory dropping below 150/160 during use.

Is power being interrupted to the unit?

Have you tested the drive with Powermax diagnostics?

Are any IC chips getting hot to the touch as opposed to warm?
 

·
Registered
Joined
·
50 Posts
Discussion Starter #5
I have no clue as to why it is freezing up. I never had a problem with the old drive and the new one never stutters or has a faulty picture. I have been running with the case open for a couple of weeks and there is plenty of ventilation around the unit so I doubt it is heat on the drive or the chips. I installed it in the same place as the old drive. My memory has always been around 250 or above. I tested the drive with the Powermax diag. Initially I did the factory certification test. Then after the first crash, I did the low level format and I just ran the low level format again yesterday before putting the disk back in. I also changed the acoustic management to /quiet from /fast. The only thing I can think of is a bad sector is causing it to crash when you hit it. It is very strange.
 

·
Registered
Joined
·
5,119 Posts
Guys isn't the order of the new powermax diagnostics:

a) Quick 90 second test


b) Disk write pack (low level format) provided you have a mirror of your data... takes hours


c) Advanced test recertify? (non destructive takes less time than low level and can be done any time)


I know the order is different on the menu but it thought that was to keep us from accidently plowing over our data if we're not sure what we are doing when we boot up?? I watched each of these and the b) the long one looks to be blindly write zeros while c) is the one that appears to "read... and then think, read and then think" So I think it's c) that marks the sectors. I prefer to always do c) last and I've good luck maybe I should go read up on that but c) always after b) sure can't hurt.
 

·
Registered
Joined
·
4,612 Posts
Kenl, I read Powermax instructions which surmise. Perform the simplest test first. Proceed to the Thorough test if a disk problem still exist after simple test. Perform LLF if all else fails to detect disk problem or upon advice from tech support.



"Thorough test" is Quantum terminology while "Advanced" is maxtors terminology. Both refer to the same in depth test.

Maxtor and Quantum max operating temp is 135 deg F



Below is a paste of Maxtor tech supports answer to my Questions


1- What will happen to data when a hard drive exceeds its max recommended operating temperature?


2- Does "Thorough test" mark bad sectors as Low Level Format does?



At 10/29/2001 12:04 PM Maxtor wrote -


In general, as things heat up they expand. If the drive is being ran hotter than the specifications recommend then the data will not be in the same place as it was before. The platter will expand the hotter the temperature. The hotter the temperature the more warped the data will become. Depending on the circumstances there is also chance of it catching on fire.


The "thorough test" is not the same as the write disk pack test, which is the low level format. It will not mark off the bad sectors. If you have bad sectors, the write disk pack would be the way to go. Thank you for purchasing Maxtor Storage Solutions.
 

·
Registered
Joined
·
5,119 Posts
Wow great info from the horses mouth so I guess we need to run both if we want to be "marked" and certified. I especially like this part:
Quote:
The hotter the temperature the more warped the data will become. Depending on the circumstances there is also chance of it catching on fire.
Showstopper w/ easybake on locked broil/oven clean cycle?
 

·
Registered
Joined
·
4,612 Posts
Mendels, This is the LLF question I sent Maxtor support and at the bottom is their reply.


I am running the 80 gig maxtor with wvset /off to speed up data reading and writing.

I'm still having trouble with data corruption. Since the LLF was done with wvset off did powermax still mark and discard from use bad sectors regardless of the previous or current wvset setting?

In theory if wvset was on and while installed in the replay I recorded a large continuous file that filled most of the hard disk space would that mark and remove from use all questionable sectors it came in contact with?



Summary

---------------------------------------------------------------

Write verificatio off and low level format


Suggested Solution

---------------------------------------------------------------

At 11/15/2001 05:28 AM Maxtor wrote -


Dear Maxtor Customer,


When you run the LLF it should fix all bad sectors. Make sure no hard drive is on the same ribbon as thr 80gig. Make sure the Ultra cable is being used with the hard drive. Run all the Powermax tests to make sure the hard drive is OK. The performance of the drive will Slow down if the wvset is off.
 

·
Registered
Joined
·
3,715 Posts
Quote:
Originally posted by Mikeyboy
The performance of the drive will Slow down if the wvset is off.
Did you quote Maxtor correctly? With wvset ON it will slow down the drive. From the Maxtor web site:

"When Write Verify is enabled, the WRITE performance of the drive is affected as a read occurs for each write. When disabled WRITE performance is improved as, a read is not performed for each write.

When performing streaming write operations (e.g. Audio, Video, downloads) additional verification of the data being written is not as important as the continuous writing of the data to the hard drive and Write Verify should be disabled."


-Gary
 

·
Registered
Joined
·
4,612 Posts
I8er, The information from Maxtors reply is copy and paste.

I did not alter it.


Maxtor support did not address all questions asked and the answer resembles a rush job.


What I get out of this is a strong suggestion to test the drive out.


Your support statement is convincing.
 

·
Registered
Joined
·
5,119 Posts
Quote:
Originally posted by Mikeyboy

Dear Maxtor Customer,
When you run the LLF it should fix all bad sectors. Make sure no hard drive is on the same ribbon as thr 80gig. Make sure the Ultra cable is being used with the hard drive. Run all the Powermax tests to make sure the hard drive is OK. The performance of the drive will Slow down if the wvset is off.
Mikeyboy this is the part I was worried about which prompted my original above rumination. For this application I'd rather not "fix" my bad sectors I'd prefer to just leave them "marked" as "don't use" because it could be that trying to reuse "hard to read" sectors just slows us down and may cause stutters or stopples. I know you practically have to be lawyer to read the fine print there (BTW that was an obvious typo about wvset he/she meant to say on) but just to be safe you could follow my above "order" and then it should try to re-mark the unreadable sectors after the disk write pack. It's not like the "recertify' is gonna break any sectors or anything so there certainly can't be any harm in running step:


"C) (step last) Advanced test recertify? (non destructive takes less time than low level and can be done any time)"


as the final test before you take them out of the computer. I'm not a lawyer and I may be confused here (maybe that's why I depend on you to actually write to Maxtor for me) but it seems like I remember from somewhere that the disk write pack also resets the "skip these bad sectors" table when it LL reformats the whole disk. Don't ask me how recertify can be marking out sectors without affecting the data present but maybe it sorta scan disk 's them off to reserve space? Remember these drives also try to work out the "fine line" between sectors they do and don't use "on the fly" and I think "that" may slow us down while it is trying to decide whether it can easily read certain sectors or not. So I'd prefer to just skip any remotely questionable sectors (hey maybe all of them as that pre-combustion heat warps on) Well... and as long as I'm not also rapidly losing capacity on my drive! :confused: :confused: mad cow hard drive disease?? For this application we want speed over accuracy on p2 (and I think p1 already has built in redundancy making it awfully accurate-- The "stzaske logic niggle" notwithstanding but I still consider that one "user expectation error") The point is it can't hurt to "always recertify" last? Also keep in mind from Maxtor's perspective they never want to make any statements which may cast "bad light" on the reliability of their product so that could be why sometimes they don't exactly come right to the point. (I must say it is a remarkably reliable item when you consider what it is we ask it to do) ditto for these Replay units!


Also because of this "heat warp" it can't hurt to LL reformat the platters (say every year or so) so the new format structure conforms to the current perm warp of the platter surfaces? And thank goodness soon these complexities will be moot once we advance to solidstate non-mechanical forms of mega storage... and can I get a Hallelujah for that day?? :D :D
 

·
Registered
Joined
·
4,612 Posts
KenL,

I'll agree if the data on the disk is toast then the test order doesn't matter. It is desirable to identify disk failure in the least amount of time and so the tests follow short to long, non destructive to destructive.


I believe maxtors "fix all bad sectors" means remove from use. I don't have a more accurate description available and for our use it doesn't matter.


If the drive passes all tests before or after LLF it works for me. The drive is good.


I realize you wanted to take the test order a bit further.


Mendles crashing problem, if it is hard disk related should show up under the Powermax tests. Actually he may be better off running his replay with the cover on. At least then an airflow moves through the case keeping hard drive temp within an anticipated level.


If I was in his shoes and the drive failed testing I'd be thrilled.
 

·
Registered
Joined
·
5,119 Posts
Quote:
Originally posted by Mikeyboy


I believe maxtors "fix all bad sectors" means remove from use.
Well it's not clear is it? And on top of that these utilities keep being revised. If it means "recover" lost sectors in the reformat then that just makes the drive have to start all over again mapping out the bad ones (from areas hard to read physically) which it will do but "recertify" next after LLF will give it a jump start. And of course technically it does remove all the old sectors from use if it recreates the sector structures. Bottom line is, as we have all seen, a drive with "slow" difficult sectors usually passes all the powermax tests but can still be a b**** with "sector read" stutters. Even complete lockups (when it clunks an interval 3x). Too bad the test doesn't give the drive a practical time limit for the sector reads hey maybe that was one feature of that mythical "quantum quickview" technology?


And no don't do a LLF without backing up the entire drive first -- I'm talkin about from a maintenance perspective here not trouble shooting failure as you correctly pointed out.
 

·
Registered
Joined
·
412 Posts
sounds more to me a customer service standard comment .... rember they get alot of e-mail and calls from computer users who dont know much about hardware ... to tell joe shmo that the $150+ hard drive that he grabed off a shelf of a compusa has a 'Bad' sector and will be marked as unuseable will get alot of "what do you mean it's bad .. then your saying the drive is defective .. i want my money back" ..... saying "Fix" seems to be a work around for that ....


BTW most LL formats do a write to a sector , read the sector , format the sector , write to the sector , read from the sector ( up to 5 times ) ... most formats will mark the sector bad if their is more than 1 or 2 bit error during this test ( depending on what the companies QA says is acceptable drive failure ) ... Maxtor is a well knowen company and sells "junk" drives and "Server/Peformance" drives ... i would guess the acceptable error level per platter per sector are under 4% on both ( probly more like 1% for the highend drives ) ....


You will most likely never get these numbers from maxtor as this is normly confidential information.


anyway .. just my 2-cents :):cool:
 
1 - 16 of 16 Posts
Status
Not open for further replies.
Top