Reply
Yottabyte
fzabkar
Posts: 3,916
Registered: 01-27-2009
0

Re: Newer firmware does not show up when running the "serial number check utility"

Mastakilla, I am just a user such as yourself. I have no relationship with Seagate.

ISTM that you have collected enough data for Seagate to be able to clarify the issue and assuage your concerns.

Kilobyte
Mastakilla
Posts: 26
Registered: 04-27-2010
0

Re: Newer firmware does not show up when running the "serial number check utility"

I just tried to contact Seagate through the chat function

 

I didn't get much further than receiving the recommendation to upgrade the firmware to CC35.(http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=213915&Hilite=)

I explained her I already checked the Seagate website (serial number check utility) and that it seemed like the CC35 firmware was not advised for the serials numbers I have...

 

The person on the chat claims that this page is only for the 7200.11 HDs:

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

Although the page itself also clearly mentions the Barracuda LP...

 

Anyway... She was sure the CC35 firmware was also for my HDs, so I'll try upgrading the firmware tonight...

Yottabyte
fzabkar
Posts: 3,916
Registered: 01-27-2009

Re: Newer firmware does not show up when running the "serial number check utility"

Here is the configuration file I extracted from the update:
http://www.users.on.net/~fzabkar/HDD/HE-CC35.TXT

It is identified as "HEPBURN RE-FRESH CODE DISTY BARRACUDA LP UPDATE CONFIGURATION FILE".

Before applying the update, I suggest you record as much information as you can about the existing firmware so that you can identify any changes. One way to do this would be to run SeaTools for Windows and compare the *.ATA files that are generated for each drive, before and after the update. The asterisk in the file name will be the drive's serial number. The ATA file contains the 512 bytes of information returned by the drive in response to an ATA Identify Device command. It will indicate which features are supported by the firmware, and the various default settings.

If you could upload these files to a file sharing service, one of us will be able to analyse them for you. BTW, I would copy the first ATA files to a safe place, in case they are overwritten when you run SeaTools for the second time.

http://www.seagate.com/support/seatools/

Kilobyte
Mastakilla
Posts: 26
Registered: 04-27-2010
0

Re: Newer firmware does not show up when running the "serial number check utility"

[ Edited ]

oops...

 

Just too late :smileysad:

 

I just tried upgrading to CC35, but also that didn't went very well...

 

Contrary to what the person on the chat told me, 2 of my 4HDs do NOT support the CC35 firmware...

 

So now I have 2 CC34 HDs and 2 CC35 HDs :smileysad: :smileysad:

This is definitely not what I wanted and I hope this doesn't impact the ability to run a stable RAID5 on them :smileysad:

 

not sure if it helps, but here are all *.ata files I have now (you can read what firmware is on inside the file):

http://users.telenet.be/Mastakilla/HDs/ata.zip

 

I just talked with someone from the US chat support and he didn't really understand why I was told that CC35 should be for all my HDs either and suggested me to contact EU support about it again (region differences :smileysad: )

 

He did have an actual look at this thread and gave some valuable input:

"David K.: There is a question you had in the first post though that I can answer: Firmware updates are typically specific fixes, rather than performance updates so it's not always recommended to get the "latest and greatest" version of the firmware. You mention " is this new firmware "not advised" for ALL Barracuda LP HDs (perhaps it is only advised in very specific situations?)?" Firmware is more designed for these specific situations. "

 

He also told me to test if the CC35 HDs still have the clicking issue and then contact EU support again... (that's what I'll do now)

Kilobyte
Mastakilla
Posts: 26
Registered: 04-27-2010
0

Re: Newer firmware does not show up when running the "serial number check utility"

I just finished testing the 2 HDs with CC35...

 

They still click each minute when HD Tune triggers a SMART read out

 

and, a bit harder to test, they also click when HD Tune and all other HD related tools are closed (so really idling in Windows)

It took me about 20 minutes to spot in click in this situation, but they still do it...

 

Guess I'll be contacting Seagate EU tomorrow again...

Yottabyte
fzabkar
Posts: 3,916
Registered: 01-27-2009

Re: Newer firmware does not show up when running the "serial number check utility"

I have a feeling that the clicking you are hearing is normal seek noise, including periodic patrol seeks to avoid lubricant buildup on the heads.

I have looked at the ATA files and the only differences I can find are ...

(1) serial number
(2) firmware version
(3) World Wide Name (an international serial number)
(4) Time required for Normal Erase mode SECURITY ERASE UNIT command
(5) Time required for an Enhanced Erase mode SECURITY ERASE UNIT command

The data are here:
http://www.users.on.net/~fzabkar/HDD/ST32000542AS/

The meanings of the 512 bytes are explained in Section 7.16.7 of 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

BTW, SeaTools has a bug. Notice that the sizes of your ATA files are 513 bytes instead of 512. If you add all the bytes in the buggy files, you get 350Dh or 360Dh. These checksums should in fact be xx00h. This means that the extra byte is a 0Dh. If you examine each of the buggy ATA files, you will find that in all cases the 0D byte precedes an 0A. An 0A is an ASCII Line Feed character, while a 0D is a Carriage Return. It appears that SeaTools mistakenly converts LF data bytes to CRLF pairs. I have therefore amended your data by removing the 0D byte.

As for your CC35 / CC34 mismatch, I doubt that this will impact on your RAID's performance. If it worries you, then it is a simple matter to forcibly update your stubborn CC34 drives with CC35 firmware.

The reason that the update did not proceed is because two of your drives do not match any entry in the following configuration file:

http://www.users.on.net/~fzabkar/HDD/HE-CC35.TXT

See this related thread for more information:
http://forums.seagate.com/t5/Desktop-ATA-and-Serial-ATA/It-is-not-updated-Firmware-with-CC38-on-CC3E...

If you are updating from a bootable CD, then allow the update to fail on the first pass. Wait for the following message:

"The system will power off after pressing any key..."

Abort the batch by typing Ctrl-C or Ctrl-Break. Then type the following line at the FreeDOS prompt:

FDL486A -m Hepburn -f HECC358H.LOD -i ST32000542AS -s -x -b -v -a 20

Power cycle the drive when the update is complete.

Kilobyte
Mastakilla
Posts: 26
Registered: 04-27-2010
0

Re: Newer firmware does not show up when running the "serial number check utility"

You keep on amazing me fzabkar :smileyhappy:

 

I had a look at the PDF and I'm wondering when this SECURITY ERASE UNIT is triggered.

 

If the SECURITY ERASE UNIT times is the ONLY "real difference" and it is triggered only when "emptying" the HD in specific way, then I guess the firmware won't make any difference in day to day use.

Also for the RAID controller it wont make any difference. (I was afraid the firmware might also change things like power saving settings)

 

About forcibly updating the firmware, I understand that this is only advised if I'm really sure it is compatible and necessary (else it may brick my drive)

So unless Seagate support can convince me to do this, I will not do it...

Kilobyte
Mastakilla
Posts: 26
Registered: 04-27-2010
0

Re: Newer firmware does not show up when running the "serial number check utility"

I'm not sure what I'm doing wrong, but I"m having a really hard time getting a decent answer from the EU chat support (in contrary to the helpful US chat support person, who was restricted from helping me with the firmware related questions because of region differences)

 

...

Maarten Van der Auwera: anyway... can you please check the difference between CC34 and CC35? I would like to know if there are any changes in the power saving settings (which could greatly impact the ability to run in RAID)
Ali B: The CC34 has a seek time of 14 ms and hardly any difference
Ali B: CC34 firmware -> seek 14.5 ms
CC35 firmware -> seek 13.5 ms
Ali B: Is there anything else I can help you with ?
Maarten Van der Auwera: ok thanks, so the difference is purely a performance difference?
Ali B: Seek time only
Maarten Van der Auwera: ic... do you have an idea why CC35 isn't compatible with the 9XW HDs then?
Ali B: No sir
Maarten Van der Auwera: Cause I've read tricks to force the CC35 on 9XW drives, but I really don't want to do that unless I'm sure it won't do any damage
Maarten Van der Auwera: can you perhaps forward that question to a technician?
Ali B: I am not sure sir that`s why I suggested to contact the company they are going to set up your raid
Maarten Van der Auwera: I am "the company" that will set up the RAID :smileyhappy:
Maarten Van der Auwera: it's for home use
Maarten Van der Auwera: but again, this is not related to RAID
Maarten Van der Auwera: I just would like to know the difference between the 9XW and 6XW drives
Maarten Van der Auwera: and CC35 isn't compatible with the 9XW drives
Maarten Van der Auwera: Maarten Van der Auwera: and CC35 isn't compatible with the 9XW drives --> I mean "and why CC35 isn't compatible with the 9XW drives"
Ali B: there is no difference between the drive there are same models a Brarracuda LP SATA 3Gb/s 2TB
Maarten Van der Auwera: that sounds very weird, no?
Ali B: no sir
Maarten Van der Auwera: why would there then be a firmware check...?
Maarten Van der Auwera: then CC35 should just work eh...?
Ali B: I have already checked the drives do not require no firmware
Maarten Van der Auwera: but earlier you said the CC34 to CC35 firmware gives better seek performance
Ali B: I did not say that , you asked me what is the difference and I said there is no difference if you look above my answer
Maarten Van der Auwera: so it may not be required, but it does seem something people may want :smileyhappy:
Maarten Van der Auwera: Ali B: The CC34 has a seek time of 14 ms and hardly any difference
Ali B: CC34 firmware -> seek 14.5 ms
CC35 firmware -> seek 13.5 ms
Ali B: hardly any difference
Ali B: anything else I can help you with ?
Maarten Van der Auwera: I guess not then...



I'm not sure if what he sais is true (I'll check the seek time later this evening), but he claims the difference is purely in seek time
Kilobyte
Mastakilla
Posts: 26
Registered: 04-27-2010
0

Re: Newer firmware does not show up when running the "serial number check utility"

In meanwhile I tested the CC35 HDs again and, as I feared / suspected, the EU tech support person lied to me again :smileysad:

 

There is no performance difference what so ever between the HD before (CC34) and after (CC35) in both random read and random write tests of HD Tune

Also the sequential performance is the same (less than 0,5MB/sec difference)

 

I have repeated the tests multiple times

 

Now this is the 2nd time that the Seagate technical support chat of EU lied to me... First they confirm 4 times that all my 4 HDs are candidates for the CC35 (only 2 of them were) and now they lie about what the difference is between CC34 and CC35... (it clearly isn't access time or "no difference at all")

 

Now I know 1st level support is only for solving the easy problems and for getting rid of as much difficult questions as possible, but that doesn't mean they have to lie to me twice in a row... I think I did enough research before contacting them to deserve at least 2nd level support

 

I'll now try to put this all in a nice email and hope a more honest person receives it, cause I just lost all trust in the Seagate EU chat support

Yottabyte
fzabkar
Posts: 3,916
Registered: 01-27-2009

Re: Newer firmware does not show up when running the "serial number check utility"

The Security Erase command is described in section 7.43 of the ATA8 spec.

The CC34 drives require 196 minutes for a security erase (words 89 and 90 = 0xC4), whereas the CC35 drives need only 183 (= 0xB7) minutes. I wonder if a reduction in seek time could account for this difference?

Anyway, ISTM that your drives are performing to spec. I don't see any cause for concern, although, in the light of your recent experience, I understand how any unexpected click could be interpreted as a potential death rattle.

As for the confusion regarding the update candidacy of two of your drives, I wouldn't blame the tech support person for the error. In fact, I know of at least one 7200.11 firmware update which sometimes fails if the Windows version is used, but works fine from a bootable CD. This is because there are two ways to apply the update.

If you unpack the firmware distribution, you will see that there is an update utility that performs the actual "flashing".

The relevant line in the BAT file that calls the utility is similar to one of the following:

(1) fdl464.exe -m %family% -f %firmware% -i %model% %options%
(2) fdl464.exe %options% -h %cfgfile%

In case (1), the drive's model number and family are explicitly specified. In case (2), the drive's existing firmware modules are compared against an update matrix. If there is no match, then the update does not proceed.

In your case, the Windows executable and the CD ISO versions both use method (2). That's why I suggested that you use method (1).

Having said that, I wouldn't update your firmware without a good reason.