Reallocation sector count dramatically increased

[BlackST]

All I'm saying is that I wouldn't trust them.

Well i agree with you there, neither do I.

I like this irony !

[djinn]

Please, read my new post on the previous page (#155101). Due to reply moderation, my post sent yesterday has not been published until today.

One question: How many posts should I have in order to not to wait approval? I see users posts with a lot of them which are directly published without moderation.

Thank you and regards.

[poehere]

Wow, too many responses. But I'm puzzled with mixed up opinions. I think to make clear:

- The drive is NEW.
- The drive has less than 1 year usage.
- I cannot RMA it because SeaTools passes all tests.

Also I don't understand why forcing a firmware update may produce these failures. I did this update following the instructions provided by the mate fzabkar in the Seagate forums (booting a CD and using the command line). The link is HERE. I've used the correct file to update (according to fzabkar) and I've had no stability problems. I forced the update because of the all horror stories about the previous 7200.11 series.

It's worth noting another user in another forum has the same drive and he also has 10 reallocated sectors (and yes, he has not forced any update).

Please, I appreciate your answers but focus on my problem: if my drive is healthy and I can ignore those warning messages. I repeat that Victoria says SMART status is good (10 reallocated sectors are below the thresold, 54).

Thank you again

This is the new way to do things around here. Follow guides of people who are not in DR and just supply information for others to use. No testing on this. No proof it can work.

Do what others say and force updates. Then in the end you have problems come here to get answers.

I would like to suggest that you do this one. Now that this has happened ship your HDD off to Italy to the person who answers here BlackST and let him fix this for you and bring it back to life. Oh but you might not like it then it will be refurbished but working. Not sure. But looks like you either live with the HDD as it is now after what you did or you send it off to Italy and get it refurbished and fixed from this forced updata and use your HDD again. Not sure what you want now from us.

Considering you followed Franc instructions and did as he said maybe he will be willing to help you out in fixing these issues. Considering he is in AU he might opt to pay some postage to help out in fixing this one too. But I would not bet on it at all. I think that if everyone here follows information that is not proven or tested and does this then they will either need to live with it or send it off to Italy to BlackST and get it fix for a small price. Sorry but after reading this and seeing what you have done there is not much more to say on this one.

[djinn]

Well, I see... but I don't understand yet why the forced update of a firmware made for this hard disk model is generating all of these issues. You all are absolutely sure this is the source of my problems, but I've no seen any technical explanation for this.

However, the Seagate page claims the CC49 is suited for the ST3500418AS (my model). Only the serial number didn't match for the firmware update and therefore it refused to flash the update. I had not peace of mind leaving the old CC46 firmware in my disk when there is an important update published on the Seagate web page. Is Seagate going crazy or are they trying to get us crazy?

I'm not a troll, I'm a real user who has a real problem and I'm concerned about the health of my drive. So please, don't misunderstand, I just want a technical answer explaining why the forced CC49 update is the cause for this abnormal? behaviour. And, finally, can I rollback to the old CC46 version? I don't mind if I have to lose all data (backed up previously).

Thanks again

[JWCC]

Those serial numbers can make a world of difference. Don't assume that just because it's the same model, or even the same board revision, that the firmware is 100% compatible. It's entirely possible that the firmware you put on there is not the cause, but there's no way to know. :/

Let's just say that you had a program running on the drive ('firmware') that is optimized for a specific batch of drives but might at least partially work with another batch. Writing requires even more precise timing than reading because if you 'miss', you fry other data. Reading might not be a problem but writing might be running into problems. Of course, it could be that you just have a drive that has physical sectors failing.

If it was me, I'd get all my data off and use the drive's secure erase feature and use it for a DVR or something where you'll not care if it goes belly up. Once burnt, twice shy with lost data. A useless backup taught me very well. ;(

[Vorian]

Totally agree with JWCC.
Another question is why did you force the firmware update? Weren't you happy with the original version or did you experience malfunction? The problem for RMA is that you forced an update on a drive that wasn't meant to be upgraded by seagate.
I would be suspicious if someone propose you to refurbish the drive as it would certainly be "cosmetic" (that is, scan the drive & put defects in th p-list, erase smart). AFAIK, selfscan for the F3 architecture (= your drive) is still not available in the hdd tools on the market.

[BlackST]

Well, I see... but I don't understand yet why the forced update of a firmware made for this hard disk model is generating all of these issues. You all are absolutely sure this is the source of my problems, but I've no seen any technical explanation for this.

Be happy with what more or less JWCC has explained. I could give the EXACT technical explanation but it would mean either disclose some other info that I'm most probably NOT ALLOWED and DON'T WANT to as it would only be beneficial to people I don't want to educate/feed or give away for free this science (exactly !).

However, the Seagate page claims the CC49 is suited for the ST3500418AS (my model). Only the serial number didn't match for the firmware update and therefore it refused to flash the update. I had not peace of mind leaving the old CC46 firmware in my disk when there is an important update published on the Seagate web page. Is Seagate going crazy or are they trying to get us crazy?

Post an exact link to the page and the exact link ( assuming it still exist ) about the FW update forcing procedure so we can comment it.

I'm not a troll, I'm a real user who has a real problem and I'm concerned about the health of my drive.

Without demonstration of contrary it is OK. Nevertheless, there is a thorough check going on on users and their activity to weed out "bad boys". We say "Male non fare, paura non avere" - don't do anything wrong and you don't have anything to be afraid of. Don't worry.

So please, don't misunderstand, I just want a technical answer explaining why the forced CC49 update is the cause for this abnormal? behaviour. And, finally, can I rollback to the old CC46 version? I don't mind if I have to lose all data (backed up previously).

Apparently - I didn't do any test yet as I don't have too many CC46/49 drives at hand right now - it may be possible to rollback to the old CC46 firmware but still need dedicated HW tools to completely rollback also the ORIGINAL configuration (requiring a good copy specific for your drive) that I think has been overwritten or modified. THEN it is necessary to perform a sequence of operation to let's say "reformat" (it's not exact but it gives a good idea) the drive and recalculate some specific parameters just in case.
After that the drive should be OK , I would do a sort of "accelerate aging test" that stresses the drive. Honestly, it is nothing you can do on your own.

Note : this was and still is and want to stay a serious place made by pros for pros. Others should be welcome, but there's a limit to decency that I feel has been crossed too many times and rules to respect.
NO ONE of the seasoned pros will suggest you to do anything that is not approved (and so assumed factory tested, even if it is not a guarantee it's 100% good).
This is because :
1) manufacturers DO still make troubles putting sometimes unfinished stuff on the market and then patch FW later as problems surface up. It is not good but it is partially justifiable by market pressure and there's little to do.
2) We know our stuff , the "hobby spirit" is still intact otherwise it would be quite alienating to work with these things, but there must be a logic on all this.
3) To me it is not fair to give PERSONALLY untested advice or become a circulating pump of internet bullshit/untested stuff/rumors just to rack up posts, rise up on search engines or vent the fingers. Would be different if there was a thread (and still has to be determined if it is good or not as - usually - in wrong hands this technical stuff makes disasters) where the safety of such procedures AND real testing was discussed and performed.
4) Always in my opinion it is not fair to give advice who can potentially harm data or personal property / things (the drive itself) without a strong "caveat" and most of all without being PERSONALLY ACCOUNTABLE , AVAILABLE AND RESPONSIBLE in order to answer and possibly remedy to what goes wrong - in the limit of overall fairness and good faith.

Just my 40 EuroCents

[djinn]

The exact link was posted before. Here you have it again. As you can see, your mate fzabkar tells us how to accurately update the firmware. Based on the reply of the user, I figured out that this is the right way to update the drive. In fact, I've used this firmware for about 5 months without ANY problem (yes, it's the truth), since I bought the drive, and only on the 6th month reallocated sectors appeared. However command timeouts and high fly writes moderately increased with the time, but within the safe values of SMART.

The reason to do this nonsense operation was posted before: I was scared with the horror stories about the 7200.11 series (no drive detected). Now I think I made the wrong decision, but apart from those parameters in SMART, the drive works (apparently) OK.

And now, I appreciate your useful replies. That's what I wanted to hear. Thanks a lot.

BlackST: How much does the repair cost (without shipping, naturally)?

[BlackST]

In fact, I've used this firmware for about 5 months without ANY problem (yes, it's the truth)

No doubt. I always say that these "hacks" , despite it works immediately, make "time bombs"

... and only on the 6th month reallocated sectors appeared.

Bomb triggered...

However command timeouts and high fly writes moderately increased with the time, but within the safe values of SMART.

This is exactly what happens. Internal parameters are fine tuned for a reason.
Some users may notice it, some other no. Unpredictable. Sometimes the bad mix is new parameters + your HW.

Now I think I made the wrong decision, but apart from those parameters in SMART, the drive works (apparently) OK.

The drive CAN or CANNOT develop further problems or crash. As I don't know what kind of parameters are changed and what match / what not in the forcefully update firmware and your specific drive, it is just guessing.
My theory : if it wasn't designed for it, it is not safe. Biased opinion ? Other strictly technical facts are welcome. Period.

Side note : I made a test some time ago and found my notes, it was another kind of drive (Maxtor Calypso series , to be precise) and I used another FW to gain access and repair it (in a nutshell, the SATA FW on the ATA version) . I left it as is and used it as temporary storage for "garage work" : the drive worked but it developed some strange problems including slow sectors and bads, but not limited to. Putting back the original FW it was designed for and recertifying from scratch solved the problem once for all. Certain timings and things maybe don't allow too much space for "experiments", I guess.

P.S. A pearl of wisdom from our labs :
This is the output AT STARTUP on a Seagate where INTENTIONALLY for data recovery purpose I modified the internal parameters . The drive works (OH YEAH !!) and CAN be used but this is what THE DRIVE sends at startup . The drive is a Momentus 5400.5 but it is common on other .

Drive AMPS Configuration has been modified from compiled defaults.
Drive must be re-initialized to controller firmware defaults by re-downloading controller firmware
DO NOT SHIP WITHOUT FIRST RE-DOWNLOADING CONTROLLER FIRMWARE OR RESETTING TO DEFAULTS!


Is it clear enough ? Let's see if (would be pure decency if with credit) this kind of info will be pumped to other place. Pro bono day today.

For now, backup-backup-backup and don't store anything critical on it.

[Vulcan]

@djinn,

You've received lots of advice already. I'll just reply to one of your comments (from your "delayed" posting) which hasn't (as far as I can see) received a reply so far:

I appreciate your answers but focus on my problem: if my drive is healthy and I can ignore those warning messages. I repeat that Victoria says SMART status is good (10 reallocated sectors are below the thresold, 54).

a) FYI you are interpreting the SMART data incorrectly. Specifically the threshold for attribute #5 is 36 decimal, not 0x36 (54 decimal) and the threshold value is compared with the normalised value, not with the raw data. However the conclusion is correct anyway - the overall SMART status is a "pass", based on your supplied information. As always, that does not mean that the disk is truly "healthy".

b) For various reasons shown in the SMART data, personally I would be concerned about the longevity of your disk, in its current state (e.g. read the Google data centre study, which included their experiences with analysing some SMART attributes, and which fits with my own experience over the years). However I don't expect anyone to care about my opinion.

[djinn]

Ok, thank you all again for your advices. I made the wrong decision, but I'd like to know how much does the repair cost if you don't mind, BlackST. If costs are reasonable (less than 60€), maybe I send the drive to you. If it's more, it's not worth and it's better to buy a new drive.

I will backup all content and I will report any new issue.

Thank you indeed, much appreciated.

Regards

[BlackST]

Wait for my tests to end I want to be sure about some things.

[BlackST]

Totally agree with JWCC.
Another question is why did you force the firmware update? Weren't you happy with the original version or did you experience malfunction? The problem for RMA is that you forced an update on a drive that wasn't meant to be upgraded by seagate.
I would be suspicious if someone propose you to refurbish the drive as it would certainly be "cosmetic" (that is, scan the drive & put defects in th p-list, erase smart). AFAIK, selfscan for the F3 architecture (= your drive) is still not available in the hdd tools on the market.

If there is someone into 'cosmetic' works it's not me !

[djinn]

Update on this.

Now my HDD has 68 reallocated sectors and 2 uncorrectable errors (high fly writes have not increased though). The new additional 58 sectors (I had 10 initially) appeared in 2 days. It's worth noting these new sectors arised coinciding with a sudden change in temperature (temperatures increased about 10º C in my city). Is it a coincidence? Previously, I got the first 10 reallocated sectors with another change in temperature. I don't know if this may be related with the fixes in firmware version CC49 according to this LINK. In addition, HD Sentinel told me the health of my drive is now at 31% and that I must backup data often. Surpringsily I cannot RMA the drive yet, even when the drive has exceeded the threshold of 36 reallocated sectors.

Finally, I've ordered a new WD5000AAKX HDD and I'll use the old one for DVR purposes. Is the ST usable as a secondary drive?

Thank you again for your feedback

[Vulcan]

Surpringsily I cannot RMA the drive yet, even when the drive has exceeded the threshold of 36 reallocated sectors.

FYI, as I explained before, you are mis-interpreting the SMART data.

Your drive does not have a threshold of 36 reallocated sectors. For SMART attribute #5, the threshold (36) is compared with the "normalised" value of the attribute (whatever that is now), and not compared with the raw value.

Or to say it a different way, the trigger value for that SMART attribute is not 36 reallocated sectors.

[djinn]

Surpringsily I cannot RMA the drive yet, even when the drive has exceeded the threshold of 36 reallocated sectors.

FYI, as I explained before, you are mis-interpreting the SMART data.

Your drive does not have a threshold of 36 reallocated sectors. For SMART attribute #5, the threshold (36) is compared with the "normalised" value of the attribute (whatever that is now), and not compared with the raw value.

Or to say it a different way, the trigger value for that SMART attribute is not 36 reallocated sectors.

Can you elaborate this with maths (with an example, if possible). I don't understand yet how to know when the reallocated sector count is critical.

Thanks

[BlackST]

Reallocated sector count is not the worst thing in THAT drive (still the same drive ?)

And assuming there is a normalized value of the attribute, does it matter ?!?

[djinn]

Reallocated sector count is not the worst thing in THAT drive (still the same drive ?)

And assuming there is a normalized value of the attribute, does it matter ?!?

Yes, it's the same drive. What about the sudden change in temperature? May it be related or is it just a coincidence?

[JWCC]

Well in theory...
OK I'll tell you about some experience from the mid 1990's. Note that I'm not saying this is true in your drive's case. Without inside knowledge of firmware revisions or reverse-engineering the drive in my hand, there's no d*** way to know. It 'might' be true or it might be that it's just a coincidence. Like others said, messing with the firmware modules is deep doo-doo territory if you don't know what you're doing. Blaming temperature, moon phases, magnetic orientation (due north), crow landing on your porch, sun spots activity, and so on is all cargo cult, heh.

I had some drives that were able to read after being cooled (or heated in some cases - you had to leave them running for a while before they'd work) but would not work all the time. They clearly depended on a specific range of temperatures. I'd imagine that drive companies have learned from and fixed this issue on newer firmwares. So that old 1GB drive and the new 2TB drive in the store have quite a few improvements in software technology and not just 'more density'. In fact, I'm willing to be that 90% of the density improvement is directly connected to software changes to let them get away with the more precise hardware, but you'd need to ask an expert who's reversed newer drives to tell us if that's true. Not a whole lot of those people are willing to talk, for obvious reasons.

Kind of barely related but probably interesting: Back when people still used 3.5" disk drives, it was common to have some drives that due to calibration issues, would write data that couldn't be read on all drives. They would also have problems reading from other drives.

[djinn]

@JWCC:

Thank you very much for your feedback. I thought anybody in this forum would not reply to my post any more.

Yes, I made a big mistake, I know, I thought that updating the firmware as fzabkar said would be harmless. Also I know my drive has no solution without proper professional diagnosis, as many already said.

At this point, I would like to know if a secure erase might mitigate the new bad sector issue. I must say bad sector counter has stopped now at 70 since I'm doing regeneration (write) tests with HD Sentinel (at level 5 of reinitialization), even with ambient temperature changes. High fly writes counter has been stopped too, at 11. In addition, I've discovered a few slow sectors when writing with this test, but these sectors have improved after 2 passes of that test. I'm asking about the secure erase feature not only for its ability to discover new bad sectors, but I've read on some internet sites that secure erase might recalibrate firmware settings (is this true?). Does secure erase have any benefits apart from zeroing the drive? I already know secure erase does not recover reallocated sectors, only overwrites them.

I'm a totally ignorant about low level repair/diagnostics nor I'm a troll, I just want to workaround the problem to live with the drive as a secondary one (not reliable, of course). Pro repair is not worth for me since my main drive is another one: WD5000AAKX, with perfect SMART values and 100% health according to HD Sentinel (I'm happy with it).

Thanks to anyone who can/want reply to this post.