Jackie78

Samsung HDD and "Check M.C. ecc error" in HUTIL 2.10

Recommended Posts

Hi,

I recently (4 weeks ago) bought myself a Samsung HD753LJ 750GB harddrive, which works decent so far. However, there are rumours about production faults in some forums, so I thought it might be a good idea to use Samsung's own test-program HUTIL version 2.10, and check this drive, which is connected to the onboard SATA controller of my asus P5B deluxe motherboard.

So I ran a complete test, including a full surface scan in HUTIL (full surface scan took about 3 hours or so), and finally, all tests passed, except for the second one, which says:

"Check M.C. ecc error"

I am now wondering: is my drive really defective, or is this a bogus message? Can other users of the new HD753 or other F1 series harddisks tell if the get that error as well? Should I return the drive, as HUTIL supposes after finishing all the tests? Or keep it?

I am not really sure what this error means, or if the program HUTIL itself is working correctly. I have two other Samsung disks connected to the same computer (two HD401 400GB hdds), which both pass the "Check M.C." test.

Any suggestions appreciated.

cu,

Jackie

Share this post


Link to post
Share on other sites
I am not really sure what this error means,

From http://www.samsung.com/global/business/hdd...ort_Shdiag.html

"Maintenance Cylinder contains physical information of HDD."

I am now wondering: is my drive really defective, or is this a bogus message?

I wonder if anyone knows the answer to this question other than Samsung, and I have not seen any definitive answer from Samsung. (Until then, I think second-guessing what HUTIL supposes might be risky.) There was another drive with this error combined with other problems which I assumed to be defective:

http://forums.storagereview.net/index.php?showtopic=26327

Newegg customer reviews of the HD753LJ seem to be a little better than those of the HD103UJ:

http://www.newegg.com/Product/ProductRevie...N82E16822152100

http://www.newegg.com/Product/ProductRevie...N82E16822152102

The latest version of HUTIL (2.10) has apparently been out since October. If HUTIL was having problems diagnosing Samsung's flagship F1 drive, don't you think Samsung would have changed the utility by now (or at least posted a notice)?

Share this post


Link to post
Share on other sites

I guess we'll call this the official broken F1 thread. A friend of mine tested his 1TB F1 and got the same error. He tried using it anyway and got CRC errors moving files to it. He contacted Samsung Canada and every tier of support knew nothing about the problem or what to do about it.

Because of their ignorance and the frequency of similar errors reported from people online, he has zero confidence in Samsung and the F1 right now, so he didn't bother trying a second F1. He returned it for a refund and bought a 7K1000 instead.

Share this post


Link to post
Share on other sites

Seems like the Samsung F1 series doesn't seem to be reliable compared to the other Samsung lines..... I have both Samsung & WD in my server, but for next 750GB upgrade, I think I'll be going with the 750GB WD instead of the F1 750GB with so many bad reports from various websites.

Share this post


Link to post
Share on other sites

My 1TB F1 fails M.C. check, but I using it w/o any problems as system HDD at P5B Deluxe. Runs fine for a month already.

Likely it's HUTIL fault, so I just don't care.

SMART doesn't show anything suspicious, no reallocated/pending sectors. Drive runs cool and fast.

When I'll need another TB I'll buy another Samsung, or may be WD if they do 7200rpm and fix multi-read speed issue.

Share this post


Link to post
Share on other sites

I have two F1 1TB drives and both show MC and surface errors with HUTIL. SMART doesn't show any errors, Windows CheckDisk doesn't show errors, and full surface scan with HDUTIL doesn't show errors. The drives run perfectly. I have a feeling that something is wrong with HUTILs compatibility with this HDD.

Share this post


Link to post
Share on other sites
To those who experience m.c error in Hutil with F1 1TB: What is your access time?

Results from HDTune Benchmark:

Access Time: 13.6ms

Burst Rate: 125.4 MB/s

Average Rate: 92 MB/s

Share this post


Link to post
Share on other sites

Btw. I'm really amazed at the transfer speed between two F1s on Gigabit network. I average 90MB/s - not bad for a LAN. My older SATA drives averaged only 50MB/s.

Share this post


Link to post
Share on other sites

I bought 4 HD753LJs last week but two have surface errors and every four reports M.C. ecc error...

wtf?

Edited by inerc

Share this post


Link to post
Share on other sites

Unfortunately, Samsung obviously doesn't want to answer me to my simple question, if the Check M.C. ecc error is normal or not - they simply don't reply to my emails. So, what would you do? Should I keep this drive, or return it, as HUtil suggests?

Share this post


Link to post
Share on other sites

I got 4 Samsung F1 750s today. I've only tested one and it shows the same M.C. error as you guys, although all the other tests, including a full surface scan, passed. Since I wanted to use these four for a RAID, I don't think I am going to risk anything wrong being with them and thus not bother with the RAID array yet. I have a local Samsung service center here, but I feel the guys there will most likely (99%) not know of this error or what to do with it either, although I'm sure they would replace the drive. For now, I'll run the tests on the other 3 drives and report back here.

PS Oh yeah, I almost forgot. On completion of the test, HUTIL recommends doing "ERASE DISK" which is basically a low-level format, and then running the test again. That seems weird but I think I might try it (although I would feel kinda crappy if it "fixed" the problem because I'd feel like the drive is actually faulty and I would only postpone the inevitable). Has anyone tried this procedure it recommends?

Edited by Fedor

Share this post


Link to post
Share on other sites
PS Oh yeah, I almost forgot. On completion of the test, HUTIL recommends doing "ERASE DISK" which is basically a low-level format, and then running the test again. That seems weird but I think I might try it (although I would feel kinda crappy if it "fixed" the problem because I'd feel like the drive is actually faulty and I would only postpone the inevitable). Has anyone tried this procedure it recommends?

Yeah, I erased my F1 1TB in HUTIL - No change. It still reported m.c. error.

Share this post


Link to post
Share on other sites

Thanks, I probably won't bother with it at all then. So far 3 of 3 have the MC error and 2 have finished all other tests fine (the 3rd one is running the full surface scan now). If it turns out that all have the MC error, but all other tests pass fine, I will boot into Windows, do some benchmarking to see if anything is off, and then try moving over data with .sfv files and then verify they got moved without corruption. Frankly as it stands with my drives, I can't help but feel that HUTIL is giving a bad diagnosis - no read/write/seek/SMART errors and no bad noises or anything makes the drives look totally fine. If I hadn't run HUTIL, I think I wouldn't even have a clue that something might be wrong!

I'm going to sleep now but tomorrow I will test the fourth drive and then try the Windows stuff, and I'll be sure to report back.

Share this post


Link to post
Share on other sites

So finally, we may assume that this mysterious "Check M.C. ecc error" is an error of Samsung's HUtil itself rather than a hardware error, right? Well, isn't it a pitty that Samsung doen't try to clarify the situation? Up to now, they haven't answered to any of my mails concerning this topic yet, nor have they put that topic in their FAQ.

Has anybody ever returned a drive showing a "Check M.C." error, and received a drive without that error in return?

Share this post


Link to post
Share on other sites
So finally, we may assume that this mysterious "Check M.C. ecc error" is an error of Samsung's HUtil itself rather than a hardware error, right?

I wouldn't assume that an M.C. error is meaningless and ignorable without official notice from Samsung. The maintenance cylinder is the single most important part of the disk surface (even more important than track 0), containing information that may be critical for proper drive operation in both common (every-day) situations and less common (future) situations - like low/high temperatures conditions, high vibration/off-track conditions, sector reallocation, etc.

(If the M.C. errors are being caused by a bug in HUTIL and the drives are actually error-free, I assume Samsung will soon officially make that known if they haven't already.)

My guess (and it's only a guess, assuming that HUTIL is correct) is that Samsung may use the "maintenance cylinder" for the location of spare sectors and/or to track the location of faulty sectors. The M.C. error may be preventing the drive from locating spare sectors and/or "reallocating" faulty sectors to spare sectors when the drive's own diagnostics indicate that data sectors are faulty (e.g. because they have ECC errors).

This may be why many (most?) people with M.C. errors also have ECC errors in the simple and full surface scans of data sectors. (My understanding is: New drives shouldn't have any ECC scan errors. Data sectors with ECC errors should have been reallocated to spare sectors at the factory during the burn-in test just prior to shipment, but perhaps this is not happening because of the M.C. error. Sectors which are reallocated at the factory do not show up as "reallocated" sectors when a SMART utility is run; new drives almost always shows zero reallocated sectors in SMART, the zero meaning no sectors have been reallocated since leaving the factory, even though the new drive may have many factory reallocated sectors.)

Another possibility (guess) for the F1 is that the drives' spare sector areas are being fully filled (or mostly filled) with data from reallocated sectors before the drives leave the factory (due to ECC errors detected during burn-in), and this is giving rise to the M.C. error as a warning message.

Of course, it's best to listen to Samsung on this one (and I haven't seen anything official from them yet).

Samsung's possible use of the maintenance cylinder for faulty sector mapping:

http://v3.espacenet.com/textdoc?DB=EPODOC&...4190185&F=0

Some other Samsung uses of the maintenance cylinder:

http://v3.espacenet.com/textdoc?DB=EPODOC&...6624957&F=0

"The outermost zone, cylinder CYL_M, is reserved as a maintenance cylinder to record essential data necessary for the maintenance of the magnetic disk recording apparatus. This essential data may be, for example, a list of defects in the sectors of the disk, the data of spare sectors, parameters of the read/write channels, etc."

and more:

http://v3.espacenet.com/textdoc?DB=EPODOC&...5078393&F=0

"Information is stored in the annular tracks 134 on the disc 112. Generally, the disc 112 has a data zone where user data is written, a parking zone where a head is located when the drive is not used, and a maintenance cylinder. Values such as a type of a head, write parameter correction values "A" and "B" at low and high temperatures, and write parameter correction values "a" and "b" according to the type of the head, are stored in the maintenance cylinder."

Share this post


Link to post
Share on other sites

So I'm curious... Does ANYONE have a Samsung F1 drive that DOESNT have the M.C. Error? If not, then we would certainly have a sound basis for assuming the HUTIL program is for example not updated to correctly diagnose these models.

For the record, I've now finished testing all of my four 750GB drives in every way I know. They ALL had the MC error, and no other errors at all, including a pass on the full surface scan (btw this means they have no SMART errors either since that is one of the tests). Further, each drive performed well in HD Tach, both in terms of reads/writes and in terms of seek times. I then proceeded to dump data onto them, no less than 20GB each, and then .sfv check that data against the originals, and all of it came out fine. Frankly, so far the evidence is strongly in favour of the fact that HUTIL has a bug or is simply outdated for the F1s from my experience. Since Samsung didn't even HAVE an HDD diagnostic utility for a long time, it seems that it was never a priority and I'm guessing could have even been outsourced and has not been updated since. I feel unsafe trying to build a RAID5 array out of these potentially faulty drives but I have no reason to expect that their replacements would be any different.

I will try to contact Samsung although my hopes aren't high considering the lack of success the rest of you have had in this regard.

Share this post


Link to post
Share on other sites

Hi fellows,

I have now got a second HD753LJ, and it reports the Same Check M.C. error in HUtil. has anybody of you heard anything from Samsung concerning this topic? I have sent two emails some days ago already, none of them has yet been answered. :(

Share this post


Link to post
Share on other sites
has anybody of you heard anything from Samsung concerning this topic? I have sent two emails some days ago already, none of them has yet been answered. :(

There is a thread at Silent PC Review where a poster [andyb at Jan 31, 2008 8:16 am] says he has received word from Samsung of a "problem" (it is not clear whether the problem is with the drives or HUTIL). He says he has been told (by Samsung) that more details will be coming today.

http://www.silentpcreview.com/forums/viewt...0&start=180

Share this post


Link to post
Share on other sites
has anybody of you heard anything from Samsung concerning this topic? I have sent two emails some days ago already, none of them has yet been answered. :(

There is a thread at Silent PC Review where a poster [andyb at Jan 31, 2008 8:16 am] says he has received word from Samsung of a "problem" (it is not clear whether the problem is with the drives or HUTIL). He says he has been told (by Samsung) that more details will be coming today.

http://www.silentpcreview.com/forums/viewt...0&start=180

I meant to say thanks for the great link and apparently forgot. I came back round to this now however, to mention that the spcr thread has been updated and supposedly someone there now has a confirmation from Samsung that HUTIL is incorrectly diagnosing the 1TB drives (or maybe the whole F1 line) and that they are working on a new version. Hopefully it is the whole F1 line because my 4 750s still appear fine in every way except for that M.C. error *knock on wood*.

Share this post


Link to post
Share on other sites

I bought five of these drives (5x 1Tb) here in the UK in the middle of January, which all had serial numbers within a range of fifteen-twenty numbers of each other.

All five showed an ECC error on the M.C. test and four also on the Surface Scan test. I did a low level format on one of the drives which had both errors but it did not help, so couldn't be bothered to run the test for the other four.

I RMA'd all five and got 5 new drives as replacement with totally different serial numbers, although still within a close range from each other (same batch I guess).

They all failed the M.C. test at which point in the sequecne I stopped the test before the Surface Scan since I knew I was not going to keep them anyway (and honestly because I was so p_i_s_s_e_d o_f_f by Samsung and the whole thing that I was out of patience).

I have now been granted a refund from the retailer so all well there (despite the very time-consuming hassle of taking delviery of, installing, testing, emailing, repacking, resending five drives... twice...).

I think it is worth mentioning also that this was not only a theoretical problem for me; the drives simply did not work with my Areca ARC-1231ML PCI-E raid controller card, not even connected one by one, and the card set to work in a simple JBOD configuaration. I have no such problems with two Barracuda 7200.10 750Gb drives that I also own. For this reason, I believe these F1 HDDs suffer other more serious problems than just not being compatible with HUTIL v2.10.

I hope Samsung will sort this out and I am holding off buying other drives, at least till another manufacturer introduces 3x333Gb platter solutions.

Share this post


Link to post
Share on other sites

The incompetence of Samsung in dealing with this issue is enough to put me off buying any current model/revision F1s. If they aren't smart enough to catch their newest drives erroring with their own diagnostic tool, and they aren't smart enough to respond quickly and comprehensively to their customers who report the error, then they also might not be smart enough to produce a properly reliable drive; or find fatal problems in the design; or .... etc.

We could be looking at the next Deathstar here and Samsung wouldn't even have to cover it up -- they might just be totally ignorant of it.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now