Barracuda XT, Barracuda, Barracuda Green, DiamondMax and other desktop SATA and IDE drives

[Mastakilla]

Hi all,

I have 4 Seagate Barracuda LP ST32000542AS, 2TB HDs.

When filling in their serial in "serial number check utility", the utility sais that no new Firmwares are available:

http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207931&NewLang=en

They all have firmware CC34, but while looking in this forum I just found out the CC35 exists as well:

http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=213915&NewLang=en

What does this mean? Is this new firmware not valid for my HDs? Or is this new firmware "not advised" for ALL Barracuda LP HDs (perhaps it is only advised in very specific situations?)?

There is no info on what is changed in this firmware.

I'm currently in the process of RMAing all my 4 HDs for making unacceptable clicking noises while idling and would like to know if this new firmware might be safe / worth to try...

[fzabkar]

I don't understand the clicking at idling. I thought that it may be due to the heads parking off the platters on a load/unload ramp, but your model appears to have a landing zone on the platters.

Barracuda LP Series SATA Product Manual, Rev. C:
http://www.seagate.com/staticfiles/support/disc/manuals/desktop/Barracuda%20LP/100564361c.pdf

Could the clicking be a Background Media Scan, or offline SMART testing?

Try a SMART diagnostic tool. Look for reallocated, pending, or uncorrectable sectors.

HD Sentinel (DOS / Windows / Linux):
http://www.hdsentinel.com/

HDDScan for Windows:
http://hddscan.com/

I extracted all the files from the firmware upodate, both the CD ISO version and the Windows executable.

http://www.seagate.com/staticfiles/support/downloads/firmware/BarracudaLP-ALL-CC35.iso
http://www.seagate.com/staticfiles/support/downloads/firmware/BarracudaLP-ALL-CC35.exe

There is a firmware update matrix in an encoded configuration file (HE-CC35.CFS).

The decoded version is here:
http://www.users.on.net/~fzabkar/HDD/HE-CC35.TXT

The format of each entry appears to be ...

(model);(original firmware modules);(update image file);(updated firmware modules)

I suspect that the updater is attempting to limit the application of updates to drives that were affected by a particular bug, by verifying that your drive's firmware modules match a list of affected ones.

For example, the following lines would suggest that not all ST32000542AS drives with CC34 firmware have identical modules:

ST32000542AS;HE1DDB.CCD3.DK0B2H.CC34.BD45;HECC358H.LOD;HE1DDB.CCD3.DK0G2H.CC35.BD6B ST32000542AS;HE1DDB.CCD3.DK0C2H.CC34.BD47;HECC358H.LOD;HE1DDB.CCD3.DK0G2H.CC35.BD6B ST32000542AS;HE1DDB.CCD3.DK0F2H.CC34.BD4C;HECC358H.LOD;HE1DDB.CCD3.DK0G2H.CC35.BD6B

That said, you can forcibly flash CC35 into your drive by modifying one line of text in the updater. Let me know if want to do this.

[Mastakilla]

Thanks a lot for the extensive reply!

I will not be updating the firmware in that case. Sounds like it is not "officially supported", which is quite important being in an RMA procedure

About the clicking:

I have created 2 very detailed forum topics about the issues I'm having (including 3 youtube videos):

http://www.xtremesystems.org/forums/showthread.php?t=250597

http://www.silentpcreview.com/forums/viewtopic.php?t=58633

http://www.youtube.com/watch?v=5LVwDSgwHpo --> short video (see comment to know when clicking happens)

http://www.youtube.com/watch?v=e4W97j6erms --> longer video (see comment to know when clicking happens)

http://www.youtube.com/watch?v=GWzTHyBdNK4 --> video of the one HD that really "died" in meantime

Short summary:

* 4 HDs

* started with 1 HD attached ICH10 on which I installed Win7 x64 (nothing overclocked yet, all at default and high quality hardware)

noticed distinct clicking sound above all case fan noise (intel and ati stock fans) and even during listening to music at a normal volume (although the clicking is not very loud, it is a type of sound that is easily heard). I have 6 7200rpm currently in my old desktop and none have this type of sound during idling...

* ran all tests of Seatools (including the long ones) and a lot of benchmarks. no faults were found and performance also seemed normal

* reinstalled win7 x64 on SSD, installed all 4 HDs, removed partition from HD which previously had win7 (so all 4 HDs are completely empty)

* ran all tests of Seatools (including the long ones) on all 4 discs and a lot of benchmarks. no faults were found and performance also seemed normal, but clicking still happened

* attached all 4 HDs one by one and listened for clicking sounds. clicking happened on all 4 HDS. Some had it more often and were louder than others. One could even be considered almost acceptable in volume... I was able to trigger the clicking to happen sooner and more often by running a benchmark (eg HD Tune Pro) before the idling

* tested if clicking also happened without SATA cable attached. It did happen, but since I wasn't able to run the benchmark to "trigger" the clicking, it didnt happen very often or loud.

* 1 of the 4 HDs started clicking constantly during non-idling. Also performance went down extremely. This HD really "died" it seems. Makes me wonder when the others will die...

I wasn't really expecting much more valuable info on this, which is why I didn't elaborate in detail about this in this forum yet.

But you did bring up some good points:

1) I did always enable SMART, so perhaps it is indeed SMART that is triggering the clicking. Although that the clicking did also happen without the SATA cable attached. I'll try disabling SMART later today.

2) I'll also try HD Sentinel and HDDScan later today

Thanks again and please let me know if you have more valuable input!

[Mastakilla]

I just discovered something weird

I currently have 3 of the HDs attached and on one of them I'm running HD Sentinel full surface read + write test, while the other  are idling.

I just noticed that currently all HDs are clicking at a constant rate.

They click each 21 seconds.

The 2 idling HDs click quite loud with about 0.3 seconds delay to eachother

The HD doing the test, also clicks, but it is hardly noticeable (I have to put my ear on the HD), but it does "click" with the same rate as the other HDs (meaning they are also doing it while not idling, only it is not noticeable then...)

anyone have an idea what this means?

I havent tried to disable SMART yet (the surface test on this disc alone will take about 16 hours), but that will be the next thing to try.

Also I am 100% sure it wasn't always clicking at a constant rate. Most of the time it is very random (see the youtube videos)

[Mastakilla]

HD Sentinel surface read + write test didnt find anything on one HD (dont have time to test the others, but I'm quite sure I already did a similar test with Seatools)

Disabling SMART also does not help against the clicking

I'm going to unplug the HDs now and send them for RMA

Hopefully the replacement HDs don't click.......

btw: they now stopped clicking at the regular 21 sec interval... not sure what triggered that (the HD sentinel surface test on one of the HDs perhaps?)

[fzabkar]

Sorry for my slow response.

AIUI, random clicking, especially when it is associated with freezing during reading and writing, is symptomatic of weak read heads or bad sectors. This may show up in HD Tune as dips in the performance graph.

I suggest you try a HD Tune performance test in both horizontal and vertical orientations.

FWIW, here is a thread which shows some curious results for a vertically mounted drive:
http://forums.seagate.com/t5/Internal-ATA-and-Serial-ATA/ST31000528AS-7200-12-1TB/m-p/42506

As for the regular clicking, I wonder if that is a "green" APM (Advanced Power Management) issue. "LP" drives are Low Power, so I'm wondering if the drive has an APM setting that retracts the heads and places the drive in a reduced power management mode ???

You may be able to use a utility such as hdparm or HDDScan to set the APM levels for your drive:

http://en.wikipedia.org/wiki/Hdparm
http://hdparm-win32.dyndns.org/hdparm/

HDDScan for Windows:
http://hddscan.com/

The ATA commands, including APM, are documented in the following document.

Working Draft AT Attachment 8 - ATA/ATAPI Command Set (ATA8-ACS):
http://www.t13.org/Documents/UploadedDocuments/docs2008/D1699r6a-ATA8-ACS.pdf

In your other threads you have recorded periodic clicking for one particular drive, the times being ...

2:18
2:51
3:24
4:28
5:00
6:32
7:08

On average, the clicking repeats every 32 seconds. In your current thread you have identified periodic clicking every 21 secs.

You can possibly determine what the drives are doing during their "idle" periods by retrieving their SMART reports.

The raw values of the Raw Read Error Rate and Seek Error Rate attributes are actually sector and seek counts, respectively, so the differences in the data between successive reports will tell you how many sectors and tracks have been accessed.

Attributes F1 (241) and F2 (242) report the number of LBAs written and read by the host, ie your PC, so any increase in these numbers would be related to OS activity.

FWIW, here is another thread that may interest you:

WD Elements with permanent access noise in idle:
http://community.wdc.com/t5/Other-Externals/WD-Elements-with-permanent-access-noise-in-idle/m-p/2156...

[Mastakilla]

No problemo! Thanks for answering now...

The HDs are currently on the way back to the shop for replacement (hope to get the new ones somewhere next week)

I did run many HD Tune performance tests and they all seemed normal (except for that one HD that really broke down last week - HD Tune screenshot in the XS thread)

The HD also made noise when SMART was disabled or even when there was no SATA cable attached.

Could it then still be APM or SMART?

Also, if it is APM, it doesn't make any sense to "redo" every x seconds while idling (there were no partitions on the HD, so nothing todo during idling - that would mean a defective APM in my eyes)

anyway.. I hope the replacement HDs don't have the issue, but if they do I'll make sure to check out the things you just mentioned!

[fzabkar]

Laptop drives have APM issues which result in frequent head parking. I don't know about desktop drives, though. The following wiki might give you some food for thought.

Drives which perform frequent head unloads under Linux:
https://ata.wiki.kernel.org/index.php/Known_issues

See the large counts in this thread:
http://sourceforge.net/mailarchive/forum.php?thread_name=1271978934_27472%40mail.internode.on.net&fo...

FWIW, here is a forum thread that talks about "Background Media Scan" in Cheetah models:
http://forums.seagate.com/t5/SCSI-drives/Cheeath-15K-6-with-knocking-sound-in-idle-state/td-p/26510

[Mastakilla]

Yesterday I received my 4 replacement HDs.

It are the same Seagate Barracuda LP 2TB HDs as before, again with firmware CC34 and again the Seagate firmware check-tool sais that there is no update available.

First thing I did was hook them up to my ICH10 and run ALL tests in Seatools (including the 2 long ones)

All 4 HDs passed those tests...

Then I started doing some HD Tune performance tests, while closely listening to the HDs for clicks...

The good news is that the HDs don't have that same irritating click during idling!!

The bad news is that at least 1, and perhaps 2, of the replacement HDs are again DOA

HD1 (ok!)

Just as a reference:



HD2 (ok!)

Just as a reference:




HD3 (broken)

I noticed about the first broken HD during the performance tests in HD Tune.
When looking in the Health tab, I noticed a warning:

In the read test everything seems fine:


But in the write test there are serious performance drops and the HD starts ticking really hard (especially during random access):




HD4 (possibly broken?)

About this HD I'm not really sure if there is a problem. It doesn't tick and it doesn't seem to have serious performance drops, but it's read performance is weird (different from the others)


As you can see there are performance drops in the 1500-1600GB area. They occur each time. All HDs are empty and don't have partitions installed (so it not because of Windows).

write looks normal:


Conclusions and a Question

1) The clicking during idling with a long interval is NOT normal. The same HDs also exist without the clicking...
2) HD3 seems pretty broken... I'll request another RMA (DOA) for this one...
3) Can anyone tell me if HD4 is ok? I want to use the HDs in RAID5, so I would think it is best that all HDs perform similar. Is this a defect? What other tests can I do to know more about this possible issue?

Thanks again!
Mastakilla

[Grim0x]

fzkabar has taken the lead on this one - I havent read all your posts / linked threads extensively. But my 2 cents: The click during Idle sounds like power to me fzkabar.