Hitachi Harddisks Thread (5K3000, 7K3000, 7K2000)

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

pnkiller78

New Member
Apr 12, 2012
9
0
1
The Intel IOPS333 was OK back in the dark ages, technology has moved on somewhat since then :)

The 3ware 9650SE-8LPML should do better than the Adaptec BUT you are limited to <2TB as the Adaptec is too ?
If this is not an issue then give this a try, its designed to run RAID6

If budget allows get your self a LSI9260/65/66 based RAID controller.
2TB+ no probs
Min 512MB cache
Can handle RAID 6 with ease
OEM versions can be cheaper but lack LSI version standard features (ie IBM M5015 needs a key for RAID6)

I was lucky and got a real cheap Advanced feature key (RAID6/encryption) to go with my IBM M5015 with BBU.
All up was $250 included cables as well of Ebay
Apparently I've found what was causing the volume corruption... the disks (physical hard disk) write back setting was enabled... I changed it to write-through and the volume corruption has stopped, I will test more to see if it's definitively fixed. The raid volume read cache and write cache is enabled, also NCQ.
 

mobilenvidia

Moderator
Sep 25, 2011
1,956
212
63
New Zealand
Apparently I've found what was causing the volume corruption... the disks (physical hard disk) write back setting was enabled... I changed it to write-through and the volume corruption has stopped, I will test more to see if it's definitively fixed. The raid volume read cache and write cache is enabled, also NCQ.
Write back is what the controller does NOT the actual HDD, the cache on the controller can be Writethrough or Writeback
The Drive's read/write cache can be enabled/disabled.
This is often confused.

If the setting to writethrough fixed your problems then the Memory on the controller could be bad.
The controller RAM is in a socket ? you could check it's seated right or replace it.
 

pnkiller78

New Member
Apr 12, 2012
9
0
1
Write back is what the controller does NOT the actual HDD, the cache on the controller can be Writethrough or Writeback
The Drive's read/write cache can be enabled/disabled.
This is often confused.

If the setting to writethrough fixed your problems then the Memory on the controller could be bad.
The controller RAM is in a socket ? you could check it's seated right or replace it.
Ok, I understand what you say. The physical disks have their own write cache setting. And the Controller has it's own settings. In the Adaptec documentation for their CLI utility, they have a command for setting the controller read and write cache modes, but also have another command for changing the "physical drive cache mode", from this link in their KB I understand that changing an individual drive's cache setting overrides the controller write cache mode configured in the controller for that particular disk.
What you say have sense, if I'm getting the info right, if I change to write-through mode on each individual drive, I'm effectively disabling the write-back setting that is enabled on the controller.
Well, apparently in this card the memory is soldered to the board, no chips inserted on any socket, every IC or chips is soldered to the board.
 
Last edited:

mobilenvidia

Moderator
Sep 25, 2011
1,956
212
63
New Zealand
The link you have is all about the Controller cache states.

The controller has a few options that it can cache, read ahead etc
The HDD cache can only be on/off

I'm still thinking something is wrong with the controller RAM.
It should be ECC RAM and correct single bit errors, but it looks to have major issues.
Normally the controller would report errors.

But if writethrough fixes it which is better for data integrity you'd best stick with this.
But I would always be doubtful of the data and if it's getting corrupt
 
Last edited:

pnkiller78

New Member
Apr 12, 2012
9
0
1
The link you have is all about the Controller cache states.

The controller has a few options that it can cache, read ahead etc
The HDD cache can only be on/off

I'm still thinking something is wrong with the controller RAM.
It should be ECC RAM and correct single bit errors, but it looks to have major issues.
Normally the controller would report errors.

But if writethrough fixes it which is better for data integrity you'd best stick with this.
But I would always be doubtful of the data and if it's getting corrupt
Yeah, I think that something is wrong with the card. I'm seriously considering to change for a LSI.
 
Last edited:

rmd3003

New Member
Feb 15, 2011
16
0
0
So what are steps to use update program? Let's say I have updater MKDLBRD.EXE and firmaware MK0NB800.BD.
I connect drive to primary port (IDE mode) and boot with bootable DOS usb stick. What are command line options?
 

kkatarn

New Member
May 24, 2012
6
0
0
Cologne, Germany
Hello kryptex, Hello planetbrian, Forum Members,
I'm a bit curious. :)

I've got two 5K3000 3 TB drives, one with FW 5C0, one with FW 580 and another three already on FW 800. I want to put them all in a RAID array and believe it's just good practice to have them all on the same FW level.

I did receive FW 800 file 'ME0NR800.BD' from Hitachi Support specifically for the pre 800 drives along with a 2.38 version of HiTest but this software ... well, let me put it this way ... I've been messing around with Hardware and Software for 20 years now, I do have a lot of stuff just lying around to check out, and still didn't find any combination of Hard- & Software that would let me use HiTest successfully :(

So, I'd like to give the DOS updater for the 7K3000 3 TB drives which 'planetbrian' mentions in post #146...
http://forums.servethehome.com/show...000-7K3000-7K2000)&p=3631&viewfull=1#post3631
... a shot but unfortunately the download link...
https://apps.hitachibackup.com/d/?PBWF9FBZ
...seems to be down :(

I'm happy to share the FW 800 file for 5K3000 3 TB drives if anybody is interested, of course. The original link from Hitachi Support is...
https://apps.hitachibackup.com/d/?3NX6ZM7R
... - not sure how long that'll work, though.

Can somebody share the 'MKDLBRD.exe' DOS-Updater so I can try if that works with the FW 800 on 5K3000 3 TB's?

Thank you very much in advance,
Kyle
 
Last edited:

kryptex

Member
May 13, 2011
117
11
18
Hi, kkatarn !

If you don't have the DOS updater for the 5K3000 3TB, DO NOT use the DOS updater MKDLBRD.EXE specifically written for the 7K3000 3TB model ! You'll probably brick the 5K3000 drive !!

For the 5K3000 3TB drive, you need to search/ask for the MEDLBRD.EXE DOS updater (which I don't have). Its name is a guess I made from some observations regarding already released DOS updaters...

However, I uploaded the DOS updater+firmware 800 for the 7K3000 3 TB (model 0F12450) here:
http://ifile.it/t6egzxd

Keep us informed, try not to brick a good 3TB drive...
Best regards !
 
Last edited:

kkatarn

New Member
May 24, 2012
6
0
0
Cologne, Germany
5k3000 fw800

Hello kryptex,
Thanks for answering my post.

I've downloaded and tested 'MKDLBRD.EXE' on one of my 5K3000's and since it doesn't detect the drive there isn't much danger of bricking it :)

I've now ordered a HighPoint RocketRaid 1740 controller. This is on the compatibility list of Hitachi's HiTest 2.38 and I'lll check if that allows me to use HiTest successfully. Additionally, I'll try again with Hitachi Support if they can now supply a DOS-Updater as they do with the 7K3000 drives.

I'll keep you posted.

Thanks again,
Kyle
 

1010

New Member
Jun 1, 2012
17
3
3
Additionally, I'll try again with Hitachi Support if they can now supply a DOS-Updater as they do with the 7K3000 drives.

I'll keep you posted.

Thanks again,
Kyle
Hitachi has a dos executable for each series/size of drive, the problem is getting support to release it. From Hitachi support's perspective it's easier for them to troubleshoot one tool rather than infinite variations of the dos tool. Even if the flashing process is the same, they still have to keep track of different executables and then deal with customers who may not be familiar or knowledgeable with a commandline environment. We can argue the merits of the stability and usability of Hitachi Hi-Test (my personal experience is the tool is worthless since it routinely locked up an OS) against the dos tool and I think most would agree that the dos tool is painless and reliable, but trying to get Hitachi support to recognize this position is a challenge.

Your only option is persistence via telephone support. Don't even bother corresponding via email, your request will be ignored. Call back Hitachi support, supply the level1 support rep with your case #. Make sure to document the time/date of call, name of the support rep, and what was discussed on the call. Re-iterate briefly the existing history on the ticket. Supply the Hitachi p/n (in your case I assume 0F12460) and request the firmware and Dos tool. If you had issues with Hitachi Hi-Test (the windows application) then state those issues to the support rep as this will bolster the case for your request. This request will most likely be forwarded to Adrian Cutler. Mr. Cutler has authority to retrieve executables and make a judgment on whether to release them. Your call will end, and it's possible that you may not receive a call back or email the same day or even 1-2 business days later. If Adrian hasn't retrieved the dos tool for your model previously for another customer request, then he will probably have to look for the tool. If he's busy, he may even forget to respond.

This is where patience and persistence come into play. After your 1st request, wait 2 days, call back. Go through the previous steps mentioned, document everything, re-iterate history of the ticket, re-request the dos tool, and BE POLITE. I can't stress that last point enough. This process may go on for 1-2 weeks or longer, but eventually Adrian will recognize repeated requests for the dos tool from his techs and be prompted to respond either with a personal callback or by releasing the tool to a support rep which will then be emailed to you.
 
Last edited:

1010

New Member
Jun 1, 2012
17
3
3
Kryptex or Odditory,

Have you updated drives to FW 800? What was your experience with this firmware with drives in a RAID environment? I'm hoping someone already has played around with FW 800 so I won't have to be a guinea pig. :eek: I'm also hoping that WDC won't sabotage firmwares post merger.
 
Last edited:

kkatarn

New Member
May 24, 2012
6
0
0
Cologne, Germany
Hello 1010,
I've received the HighPoint RocketRaid 1740 controller yesterday and it wouldn't work with HiTest 2.38, I'm tempted to say 'of course' :)

So, I've made my 2nd call to Hitachi and it went exactly as you're describing it ... the 1st level tech even recognized me from my first call 2 weeks back (when I didn't receive a case #), listened patiently to my experiences with HiTest and during the call the tech 'talked to his boss' several times and finally said that his boss would send me the DOS updater for 0F12460. Since my second name is 'patience' and the third 'polite' it was a rather pleasant talk. Should no mail arrive I'll wait until Tuesday before I call again with the case # which I now have.

I plan to run the array I'm setting up for about 5 years, so I'm in no hurry but want to make sure to have everything right from start even if that turns out to take some time.

I'll keep you posted.

Have a Great Weekend,
Kyle
 

kryptex

Member
May 13, 2011
117
11
18
Kryptex or Odditory,

Have you updated drives to FW 800? What was your experience with this firmware with drives in a RAID environment? I'm hoping someone already has played around with FW 800 so I won't have to be a guinea pig. :eek: I'm also hoping that WDC won't sabotage firmwares post merger.
Not yet, but will do the 7K3000@2TB in a couple of weeks. I have several of them on FW 580 and 5C0 (the 5C0 has better seq. performance than 580). The problem is... I haven't had any spare time for testing/benchmark etc during the last few months... crazy times... crazy times ...

Best regards to you all !